SLAS2021 Digital International Conference and Exhibition

The SLAS2021 course package contains 90 presentations including:

  • Advances in Bioanalytics and Biomarkers
  • Assay Development and Screening
  • Automation and High-Throughput Technologies
  • Cellular Technologies
  • Data Science and AI
  • Drug Target Strategies
  • Micro- and Nanotechnologies
  • New Modalities
  • 'Omics 
  • Precision Medicine Technologies
  • Keynote Speakers
  • Innovation Award Finalist Presentations 
  • Science Ignited Talks
  • Tony B. Poster Lightning Talks

Based on presenter permission, 88 of the 102 total SLAS2021 podium presentations are available on-demand.
The SLAS Scientific Program Committee selects conference speakers based on the innovation, relevance and applicability of research as well as those that best address the interests and priorities of today’s life sciences discovery and technology community. All presentations are published with the permission of the presenters.


Keynote Presentations
Innovate Within – Challenging the Status Quo Through Intrapreneurship
Open to view video.
Open to view video.
Precisely Practicing Medicine from 700 Trillion Points of Data
Open to view video.
Open to view video. There is an urgent need to take what we have learned in our new data-driven era of medicine, and use it to create a new system of precision medicine, delivering the best, safest, cost-effective preventative or therapeutic intervention at the right time, for the right patients. Dr. Butte's lab at the University of California, San Francisco builds and applies tools that convert trillions of points of molecular, clinical, and epidemiological data -- measured by researchers and clinicians over the past decade and now commonly termed “big data” -- into diagnostics, therapeutics, and new insights into disease. Dr. Butte, a computer scientist and pediatrician, will highlight his center’s recent work on integrating electronic health records data across the entire University of California, and how analytics on this “real world data” can lead to new evidence for drug efficacy, new savings from better medication choices, and new methods to teach intelligence – real and artificial – to more precisely practice medicine.
Innovation Award Finalist Presentations
Innovation Award WInner: Defining the breadth and specificity of drug response in heterogeneous immune cell populations
Open to view video.
Open to view video. Human diseases are fundamentally multicellular in nature with many different cell types contributing to disease progression and treatment response. However, how drugs impact each cell type in a heterogeneous population remains poorly understood. Conventional drug response studies, including those using single-cell profiling based approaches, have focused on pure cell types, ignoring population-level effects. Here, we applied highly multiplexed single cell mRNA-seq to study the impact of over 500 immunomodulatory compounds on human primary blood mononuclear cells (PBMCs), a heterogenous mixture of myeloid and lymphoid immune cell-types. We profiled over one million single cells using MULTI-seq to multiplex samples and used PopAlign, a probabilistic modeling platform, to discover cell-type specific responses for each compound in the library. Our conditions include CD3/CD28 stimulation, which activates signaling interactions that unmasks a wide range of drug responses that are not observed in resting cell populations. Our results highlight cell type-specific patterns of response: while many drugs inhibit T-cell activation in a similar manner, drug impact on macrophages are diverse. By classifying cell-type specific drug response signatures across conditions, we could identify two types of immunomodulators: localized and broad regulators of immune activation. We find localized inhibitors that act specifically on macrophages, such as TLR agonists and NSAIDs which induce pro-inflammatory and apoptotic programs, respectively. Broad modulators impact more than one cell-type; they inhibit activation in T cells but can push macrophages into inflammatory (M1), non-inflammatory (monocyte-like), anti-inflammatory (M2) or drug-specific transcriptional states. For instance, while JAK inhibitors and calcineurin inhibitors shift the balance toward non-inflammatory monocytes, some VEGFR and Bcr/ABL inhibitors generate more inflammatory M1 macrophages. Our analysis also reveals novel local activity for previously poorly characterized molecules, including a myeloid-suppressing function of a group of compounds including NSAIDs and an artificial sweetener. By providing new depth and insight into how existing compounds reshape immune populations, our dataset is a promising resource for improving therapeutic strategies, especially in cancer where macrophage state (M1/M2) can advance or reverse disease progression. Our platform can be broadly applied towards understanding heterogeneous cell populations in a wide range of therapeutic and disease conditions.
Reimagining Druggability using Chemoproteomic Platforms
Open to view video.
Open to view video. The Nomura Research Group is focused on reimagining druggability using chemoproteomic platforms to develop transformative medicines. One of the greatest challenges that we face in discovering new disease therapies is that most proteins are considered “undruggable,” in that most proteins do not possess known binding pockets or “ligandable hotspots” that small-molecules can bind to modulate protein function. Our research group addresses this challenge by advancing and applying chemoproteomic platforms to discover and pharmacologically target unique and novel ligandable hotspots for disease therapy. We currently have three major research directions. Our first major focus is on developing and applying chemoproteomics-enabled covalent ligand discovery approaches to rapidly discover small-molecule therapeutic leads that target unique and novel ligandable hotspots for undruggable protein targets and pathways. Our second research area focuses on discovering and exploiting unique therapeutic modalities accessed by natural products. Our third research area focuses on using chemoproteomics-enabled covalent ligand discovery platforms to expand the scope of targeted protein degradation and to discover new induced proximity-based therapeutic modalities. Collectively, our lab is focused on developing next-generation transformative medicines through pioneering innovative chemical technologies to overcome challenges in drug discovery.
Droplet-based microfluidic optical calorimeter
Open to view video.
Open to view video. Isothermal titration calorimetry is the only bioanalytical technique capable of simultaneously determining affinity, thermodynamics, and stoichiometry of binding interactions and enzyme reactions in a single, label-free experiment with a broad range of sensitivity (e.g. nM to mM affinities). It is the gold standard for characterizing thermodynamic properties of binding reactions. Despite the high value of this technique, current instrumentation is low throughput and makes these experiments cumbersome to perform, limiting its use in drug discovery to high-value measurements. To increase the throughput and sensitivity of calorimetry, we have developed and demonstrated a proof-of-concept for a novel microfluidic calorimeter that uses optical methods to measure the temperature change caused by reactions occurring in sub-nanoliter droplets. In this calorimeter, a microfluidic system creates a mixed droplet of reactants while a thermochromic liquid crystal (TLC) transducer converts the temperature change to a spectral shift, and a sensitive optical detector measures the spectral shift. Experimental measurements of the temperature change induced in droplets by the exothermic binding of EDTA to Ca2+ show good agreement with a thermal multiphysics model. Our ongoing work to improve the microfluidic mixing of reactants and increase the temperature resolution of the calorimeter has yielded temperature and energy resolution (16 nJ) for this calorimeter, which is on the same order as commercial isothermal titration calorimeter (ITC) systems and 10-fold better than most nanocalorimeters.
A Universal Chemical Processing Unit for Chemical Synthesis - the Chemputer
Open to view video.
Open to view video. We are entering the era of Digital-Chemistry whereby computer control can be used to monitor reactions, calculate reactivity and predict routes to targets but what next? In this talk I will explain how a programmable chemical computer or ‘chemputer’ can be used to do organic synthesis reactions making the process of chemical synthesis, safer, cheaper, more reliable, and freeing the chemist to focus on the more interesting aspects of synthesis, exploring for new reactivity and reactions – enabled by letting computers program chemistry. But how is this possible? We had to devise a new architecture that would allow all chemistry to be done using a programming code in an universal way. This works because it is possible to make a hard link between the process of chemistry that is done in almost all labs in the world, using batch chemistry, and a programmable architecture that would allow us to make a universal programmable chemical synthesis engine.
An Open-source and Modular Technology Platform for Automated Manufacturing and Screening of 3D Cell Culture Models
Open to view video.
Open to view video. Current three-dimensional (3D) cell culture models provide relevant models to mimic human physiology and study healthy and pathological situations. While much progress has been achieved in developing physiologically relevant 3D cell culture models, current studies are mostly conducted by using extensive human labour and manual working steps. Since manual-based methods are time-consuming and bear the risk of high variances due to human error, the current methods limit the scientific community to widely adapt 3D cell culture models and generate reproducible data sets. Although commercial laboratory automation is partially available, current solutions focus on narrow aspects of the wider problem of automating applications with 3D cell culture models. For example, bioprinters only focus on the small aspect of printing, but still require manual handling and are also not capable of producing a large variety of different hydrogel precursors in an automated fashion. In addition, commercially available automation equipment cannot be adapted easily to changing requirements due to closed software and hardware infrastructure. Here, we present a modular and open-source laboratory automation approach to address the need for customised automated solutions and enable agile and inexpensive development ( The concept was successfully applied to develop two platforms to automate the manufacturing and screening of 3D cell culture models. Firstly, the manufacturing setup converges liquid handling with emerging biofabrication functionalities to enable automated workflows spanning from the preparation of desired hydrogel compositions, to the required mixing steps with additional biochemical molecules and/or cells, and finally the production of 3D cell culture models. Secondly, a screening setup combines optical fibre-based, needle-type microsensor technology with a fully automated sample and a sensor positioning system for label-free and minimally invasive measurements of dissolved oxygen as a metabolic activity parameter within cell-laden 3D constructs. To demonstrate the suitability and the practical relevance of the developed platforms, validation studies were conducted for (i) the automated generation of highly reproducible 3D constructs, (ii) the screening of manufacturing parameters to generate a parameter library to predict material properties, (iii) the characterisation of the oxygen distribution within cell-laden 3D constructs, (iv) the identification of the time-depended drug response on 3D breast cancer models after a chemotherapeutic treatment, and (v) the evaluation of the kinetics and recovery effects after drug exposure over 35 days. In this presentation, we discuss the current workflow limitations for 3D cell culture research: Automation, throughput, and reproducibility. By introducing our open-source approach, we highlight how in-house developed automation outperforms current solutions for 3D cell culture models and is capable of fostering reproducibility and efficacy in academic research. We strongly believe that the concept could become a transformative enabling tool for the life sciences community, empowering scientists to build, share, and replicate the experimental setups.
A multiwell plate-based high-throughput technology for toxicity screening under various oxygen environments
Open to view video.
Open to view video. Introduction. Tumor tissue contains a distribution of static and dynamically changing oxygen microenvironments with levels ranging from physiologically normal oxygen down to anoxia. However, anticancer drug discovery and development initially relies on in vitro screening methods in which drug-response is assayed in cells that are most often cultured in incubators at oxygen levels that are far higher than those found in any region of the in vivo tissue. Hypoxia studies usually test a single low oxygen level. There are far fewer studies that test drug response under physiologically relevant cycling conditions. Therefore, we have designed, fabricated and characterized a high-throughput technology to perform drug screening under multiple static or cycling oxygen levels on a single plate that better represent the range of microenvironments found in a tumor. Growth, feeding and dosing the cells can be performed without the use of sealed chambers or glove boxes and can easily be integrated into current laboratory protocols. Materials and Methods. The gas manifolds were fabricated from laser cut acrylic. Rubber gaskets were used to form gas-tight channels between the permeable-bottom 96-well plate and manifolds. Gas-permeable bottom 96-well plates were fabricated using PDMS cast membranes. Finite element analyses were performed using COMSOL. Gas flow was regulated by Aalborg mass flow meters and was controlled by LabVIEW. Cells were seeded in each well, allowed to attach and cultured in the device under altered pO2 conditions for 24h. Then the cells were cultured with tirapazamine or doxorubicin for 48h. Cell proliferation was determined by MTT.Results and Discussion. Media/drug dosing is performed using standard multichannel pipets with narrow tips. In wells exposed to severely hypoxic conditions (pO2 2 levels. Using this device the drug-dose response of PANC-1 to tirapazamine and doxorubicin cultured under eight different static or cycling pO2 levels was measured. Compared to cells under normoxia, PANC-1 cells show progressively greater sensitivity to tirapazamine as static oxygen levels decrease. Interestingly, PANC-1 cells retained some sensitivity to tirapazamine under cycling oxygen conditions where cells periodically are exposed to non-hypoxic environments for half of the cycle. In contrast, PANC-1 sensitivity to doxorubicin showed little change under a range of static or cycling oxygen environments.Conclusion. We have developed a bench-top technology that shields the cells from the ambient atmosphere and maintains cells under any desired oxygen environment while media is removed and added without the use of a glove box. This new device offers a facile screening approach to determine the toxicity response of cells cultured under varying oxygen conditions.
Drop microfluidics for virus studies
Open to view video.
Open to view video. Droplet microfluidics offers unique advantages for sensitive assays, high throughput screening, and selective enrichment of target analytes. Using pico-liter drops as individual reaction vessels, drop microfluidics achieves high reaction efficiencies even with tiny input material, such as single cells, viruses, and molecules. Moreover, individual droplets can be rapidly processed to inject new reagents into droplets, an important requirement for multi-step reactions. After performing massively parallelized assays, individual droplets can be screened one by one and the droplets containing the target analytes can be selectively enriched for further analysis. Utilizing these benefits, my group develops new microfluidic-based, bioanalytical tools for high throughput and high resolution measurements. In particular, we have been developing microfluidic platforms for virus discovery and single virus genomics. The virus discovery platform enables selective enrichment of a single virus species from a highly heterogeneous virus mixture and identifies their complete genome sequence. The platform combines metagenomics, drop microfluidic platform, and computational analysis to enable genome sequencing of novel viruses. We also have been developing a high throughput, unbiased single virus genomics platform. The single virus sequencing platform we have been inventing is the first demonstration of high throughput, unbiased profiling of individual virus genomes. By unbiasedly profiling the genome sequence of individual virus particles in an ensemble virus population, we aim to study virus evolution in a quantitative manner.
A high-throughput respiratory viral antigen and antibody profiling platform for COVID-19 surveillance and therapeutic discovery
Open to view video.
Open to view video. This presentation will discuss our recent effort in developing a high-throughput and comprehensive serological profiling platform for COVID-19 surveillance. Our Coronavirus Antigen Microarray (CoVAM) comprises a comprehensive panel of respiratory viral antigens that can probe up to hundreds of different antibody responses simultaneously in a single microsized test. Using patient samples, we recently demonstrated the CoVAM that currently measures seroreactivity against 67 antigens from 14 viruses known to cause respiratory tract infections, including 11 antigens from SARS-CoV-2. This presentation will also discuss our process in developing an integrated and automated system combining microarray probing, imaging, and cloud-based analysis and adopting to a 96 well plate format to enable large-scale and widespread testing. This technology can make immediate and significant impact in order to a) determine true COVID-19 prevalence including asymptomatic individuals and implement surveillance to identify outbreaks during economic re-opening and evaluate the effectiveness of countermeasures; b) identify suitable donors of convalescent plasma for transfusion to treat patients; c) investigate the breadth and durability of the immunologic response and correlate with protection from re-infection with evolving virus strains to inform vaccine development. Using the same microarray platform, we have demonstrated its utility in highly scalable respiratory viral panel analysis for highly-sensitive detection of viral antigens. As we will likely have the flu epidemic and the coronavirus epidemic, along with common cold and other respiratory infections at the same time in a more devastating next winter, our ability to analyze the complete respiratory virus panel is therefore critical in fully discriminating these viral infections and identifying the cause of infection when they share similar symptoms.
CellPreserve: A platform to enable longitudinal cell studies
Open to view video.
Open to view video. We present an innovative apparatus and method for extending the lifetime of isolated rare cells to weeks rather than typical lifetimes of about 24 hours. This opens up a range of new and novel analysis techniques by providing a method for longitudinal studies of rare cells with applications in drug discovery and in-line QC for immunotherapy manufacturing. In many applications, rare cells contain high quality and highly relevant information for making diagnostic and clinical decisions. Although advanced cell sorting methods allow us to isolate rare cells obtained through liquid biopsy, their lifetime post capture is often limited. Whilst this is sufficient for tests which require lysis or other destructive methods, it limits the information that can be extracted to a single measurement quantity or a single point in time. By extending the lifespan of a cell through active control of its environment and nutrient supply, our platform enables an array of different measurements which can be performed over the course of weeks to obtain longitudinal information relating to a single rare cell. This approach has the potential to provide the time-based assessment of DNA release, cell phase monitoring, cytokine release and the longitudinal morphological analysis of cells through imaging. We achieve this by: -Capturing and then encapsulating a single cell in a microdroplet within the platform -Utilizing an electrowetting-on-dielectric (EWOD) system to move the microdroplet to an environmentally controlled storage region -Utilizing EWOD to divide the microdroplet in two -Identifying the empty half of the split microdroplet and extracting the supernatant for analysis/disposal -Replacing the removed ‘half microdroplet’ with refreshed supernatant thus extending the cell lifetime -Repeating steps 2—5 at pre-determined intervals This method provides the flexibility to extend cell lifetimes by adding nutrients and removing waste on demand. Integrated optics allow us to track the cell's location and to perform on-chip optical measurements of the cell and its effect on the microdroplet. In addition, by being able to extract supernatant from the cell storage microdroplets, this method allows the user to perform further assays off-chip. Our system provides a flexible, scalable and adaptable platform for generating longitudinal Omic information from single cell samples.
Advances in Bioanalytics and Biomarkers
The Evolution of MALDI-TOF MS Towards a Working Horse in Label-Free High-Throughput Screening at Boehringer Ingelheim
Open to view video.
Open to view video. Label‑free, mass selective detection and quantification of relevant analytes via mass spectrometry (MS) provides a step further towards more physiologically relevant assays within drug discovery campaigns. Reduced risk to suffer from compound interference and diminished necessity for tailored signal mediators, combined with the potential accessibility of a new target space emphasize the valuable role of label‑free MS-readouts. However until recently, MS‑based detection did not meet high‑throughput screening (HTS) requirements due to the lack of HTS compatible sample loading methods allowing measurement times of less than a second per sample. Significant advancement of liquid handling techniques as well as MS instrumentation have triggered extensive efforts in the drug discovery community to integrate the comprehensive MS‑readout into the HTS portfolio. We recently established and introduced a pioneering screening platform utilizing MALDI‑TOF MS. This approach was successfully integrated into the HTS infrastructure at Boehringer Ingelheim. A key aspect of the implementation of this technology was the development of a fully automated liquid handling concept to combine biochemical assays with MALDI‑TOF MS. HTS compatible cycle times down to 0.6 s/sample for MALDI target plate preparation and 0.4 s/sample for MS analysis are routinely achieved with the established strategy exhibiting outstanding system performance and instrument robustness. In this presentation we will give insights on the experiences we gathered in the course of several full diversity screening campaigns successfully completed with the established MALDI-TOF MS platform. Exemplarily, we will highlight how a differential derivatization strategy enables high quality and unbiased screening of an enzymatic product in the low molecular weight region.
Tandem electrophoresis and multiplexed ion beam imaging for the detection of proteoforms in single cells
Open to view video.
Open to view video. Single-cell analysis tools are necessary to tease apart the cell-to-cell variability driving many important biological processes, such as cancer and drug resistance. Differential expression of proteoforms contribute to the underlying heterogeneity within a cell population. Single-cell electrophoretic assays (scEPs) offer proteoform detection specificity, but often rely on fluorescence-based readout and are therefore limited in multiplexing capability. Detection of up to 12 protein targets has been possible in scEPs by using a method of antibody stripping and reprobing1. However, this approach is sensitive to a ~75% drop in immunoprobed signal after just one round of stripping, which makes detection of highly multiplexed, low abundance proteins challenging2. Among rising multiplexed imaging methods is multiplexed ion beam imaging (MIBI), a mass cytometry imaging technology. MIBI has been used to perform simultaneous imaging of 36 protein targets in fixed tissue by employing metal-tagged antibodies that do not suffer from spectra overlap to the same degree that fluorescently-tagged antibodies do3. We report for the first time multiplexed ion beam imaging of single-cell electrophoretic cytometry (scEP-MIBI). We perform protein separations from individual cells on a microscale device consisting of a 50-µm thick hydrated (3-µm thick dehydrated) polyacrylamide gel (PAG) matrix for protein immobilization. We confirm antibody-protein binding in the PAG with indirect fluorescent readout of the metal-tagged antibodies. Since MIBI is a layer-by-layer imaging technique, and our protein target is immobilized within a 3D substrate, we characterize the protein distribution throughout the PAG depth by fluorescent confocal microscopy and find that the highest signal-to-noise ratio is achieved by imaging the entirety of the PAG depth. Accordingly, we report the required MIBI ion dose strength needed to image varying PAG depths. Lastly, we show that by imaging ~42% of PAG depth with MIBI, we detect two isoelectrically separated TurboGFP proteoforms from individual glioblastoma cells. The characterization workflow presented here has broad applicability to validate and implement MIBI-based readout of other bioanalytical assays and samples where multiplexed detection from a 3D matrix is desirable. By increasing the amount of available antibody labels and eliminating the need to perform antibody stripping and reprobing for multiplexed detection, scEP-MIBI provides a promising strategy to increase the number of low-abundance targets detected with a simplified experimental workflow. References: 1. Sinkala, E. et al. Profiling protein expression in circulating tumour cells using microfluidic western blotting. Nat. Commun.8, 14622 (2017). 2. Gopal, A. & Herr, A. E. Multiplexed in-gel microfluidic immunoassays: characterizing protein target loss during reprobing of benzophenone-modified hydrogels. <em>Sci. Rep.</em> 9, 1–12 (2019). 3. Keren, L. et al. MIBI-TOF: A multiplexed imaging platform relates cellular phenotypes and tissue structure. Sci. Adv. 1–17 (2019).
The power of a systematic approach to mechanism-of-action studies
Open to view video.
Open to view video. Phenotypic screening is a powerful approach to small-molecule discovery; in effect, we are letting the cell reveal what intracellular targets are important for achieving a desired outcome. This approach can be especially useful in areas of biology where a therapeutic target is not immediately clear. Over the last twenty years, phenotypic screening campaigns have yielded more first-in-class drugs, suggesting that this strategy should perhaps be taken even more frequently. A major challenge, however, is to understand the target and mechanisms for compounds discovered by this method. Part of the challenge is that there is no single method that serves as a “magic bullet” to identify a compound’s target. Instead, a systematic approach that uses multiple diverse data types to triangulate upon a mechanism is more likely to result in success. It is important to keep in mind that target and mechanism are not synonyms, though one sees them used interchangeably in the literature. Understanding to what (protein, RNA, etc.) a compound binds is a very important, but not comprehensive, component of understanding mechanism of action. My group focuses on phenotypes important in diabetes, especially in the pancreatic beta cell. A loss of beta-cell mass and biologically active insulin is a central feature of both type 1 and type 2 diabetes. A chemical means of promoting beta-cell viability or function could have an enormous impact clinically by providing a disease-modifying therapy. Here, I will discuss how my group is identifying small molecules that promote beta-cell regeneration, viability, and function, and how we use technologies at the Broad Institute to understand the target and mechanisms of these small molecules. These technologies include quantitative chemoproteomics, cell morphology profiling, and analyses of gene-expression analyses, historical screening performance, or pathway perturbation.
Development of an automated MALDI MS assay for direct analysis of cellular drug uptakevia the organic anion transporting polypeptide OATP2B1
Open to view video.
Open to view video. OATP2B1, a member of the solute carrier (SLC) transporter family, is an important mechanism of substrate drug uptake in the intestine and liver and therefore a determinant of clinical pharmacokinetics and site of drug-drug interactions. Other SLC transporters have emerged as pharmacology targets. Studies of SLC transporter uptake to-date relied on radioisotope- or fluorescence-labelled reagents, or low-throughput quantification of unlabeled compounds in cell lysate. In this study, we developed a cell-based MALDI MS workflow for investigation of OATP2B1 cellular uptake by optimizing substrate, matrix, matrix-analyte ratio, and matrix application and normalization method. This workflow was automated and applied to characterize substrate transport kinetics and to test 294 top-marketed drugs for OATP2B1 inhibition and quantify inhibitory potencies necessary for extrapolation of clinical drug-drug interaction potential. Intra-assay reproducibility of this MALDI MS method was high (CV < 10%), and results agreed well (83% overlap) with previously published radioisotope assay data. Our results indicate that fast and robust MALDI MS cellular assays could emerge as high-throughput label-free alternative for direct assessment of drug transporter function in DDIs and toxicities, as well as enable drug discovery for transporters as pharmacology targets.
Assay Development and Screening
Establishment of a target identification screening platform for DMD, by combining engineered micropatterning and multidimensional image-based cell profiling of patient-derived myotubes
Open to view video.
Open to view video. Duchenne Muscular Dystrophy (DMD) is a severely debilitating disorder that is revealed by proximal muscle wasting eventually leading to respiratory insufficiency and cardiac failure. This is caused by recessive, frameshifting deletions and duplication or nonsense mutations in the DMD gene, resulting in loss of expression and function of dystrophin eventually leading to myofiber degeneration. Here we have used an image based high dimensional cell morphology phenotype analysis to identify biomarkers of dystrophic myotubes as putative therapeutic targets. We have leveraged the MyoscreenTM automated high content screening micro-printed platform from Cytoo to grow and image myotubes derived from healthy and DMD donors. Myotubes were stained for different markers and imaged using an automated microscope. The images were segmented to identify individual myotubes and ~500 features, broadly categorized as intensity, texture and shape, were extracted for each myotube using CellProfiler. A supervised learning model was then generated based on the multidimensional cell profile to separate the healthy and DMD populations. The degree of separation was then used as a method to define a multidimensional muscle marker read-out correlating with the disease state. To our knowledge this is the first demonstration of a clear and reproducible phenotype obtained by cellular imaging of myotubes derived from DMD patients relative to healthy donors. The myotube marker profile analysis can be then used in the future to screen for therapeutic targets for DMD.
Bioactivity Screening of Environmental Chemicals Using Cell Painting: Molecular Point of Departure Determination and Mechanistic Prediction Via Profile Similarity
Open to view video.
Open to view video. New Approach Methodologies (NAMs) are any technology, methodology, or approach that provides information on chemical hazard without the use of intact animals. Toxicologists are increasingly considering the use of NAMs data in chemical risk assessment in lieu of or in combination with data from traditional animal-based toxicology studies. In order to increase the efficiency of chemical risk assessment and provide data for poorly characterized chemicals, the Blueprint for Computational Toxicology at US EPA advocates the use of high-throughput profiling (HTP) assays as the first tier in a NAMs-based tiered toxicity testing strategy. In this presentation, the use of one such HTP assay, Cell Painting, will be discussed in the context of two applications for in vitro chemical hazard evaluation: 1) determination of a molecular point of departure corresponding to the threshold concentration for perturbation of cellular biology and 2) mechanistic prediction via comparison of phenotypic profiles across chemicals with known and unknown molecular targets. Cell Painting is an automated imaging-based technology in which organelles are labeled with fluorescent probes to quantify thousands of cytomorphologic features at a single-cell level to define/produce a phenotypic profile. We used this assay to screen more than 1200 chemicals in U-2 OS cells in eight-point concentration response up to 100 µM using a 24 hour exposure duration. Here, we describe the experimental design used to conduct this study, laboratory automation workflows for chemical delivery and sample preparation, image analysis workflows for phenotypic feature extraction, open source approaches for concentration-response modeling of high-dimensional profiling data and methods for evaluating assay reproducibility. Overall, 41% of chemicals screened were classified as hits using the Cell Painting assay based on changes in cell phenotype. Where possible, micromolar potency values from active chemicals in the assay as determined by concentration-response modeling of Mahalanobis distances were used to estimate administered equivalent doses (AEDs) using in vitro to in vivo extrapolation and reverse dosimetry. For many chemicals, AEDs based on in vitro potencies for bioactivity were higher than predicted human exposures, indicating low risk. Many AEDs were lower or comparable to in vivo effect values from mammalian toxicity studies, indicating that this approach is potentially conservative. Using phenotypic profile comparison methods, we observed that profiles for retinoic acid receptor (RAR) agonists (e.g., arotinoid acid, bexarotene, AM580) and glucocorticoid receptor (GR) agonists (e.g., betamethasone, budesonide, prednisolone) were similar to their respective model compounds (retinoic acid for RAR and dexamethasone for GR). In addition, profile similarities were also observed for several structural classes of pesticides (e.g., organochlorines, strobilurins, dinitroanilines). This abstract does not reflect USEPA policy. Mention of trade names does not constitute endorsement for use.
A Novel High-Throughput Multi-Phenotype Approach to Discovering Therapies for Age-Related Dysregulation of the Immune System
Open to view video.
Open to view video. Age-related immune dysregulation contributes to increased susceptibility to infection and disease in older adults. Spring Discovery has built an assay that combines high-throughput laboratory automation, high-throughput protein expression assays, high-content imaging, and machine learning to identify a suite of phenotypic features capable of discriminating young vs old donor immune cells in-vitro. The core approach of our system involves generating a number of phenotypic features that cluster with either young or old donor cells. In contrast to most high-throughput immunological assays that rely on surface markers and flow cytometry, we use high-content imaging with broad morphological cell staining palettes coupled with machine image recognition to classify features such as cell type, cell-to-cell interactions, cell activation and cell death. By using the FirePlex-HT 10-plex human cytokine panel we are able to couple the features derived from our imaging with cytokine expression data. Our analytical models incorporate both the features derived from our imaging analysis with the cytokine expression data to determine an age score that serves as a response variable for screening potential age-related therapies. By altering the feature sets included in our model and the weighting for those features we can target our screens to address specific therapeutic aims, such as reducing SARS-CoV-2 infection immune dysregulation in older patients. We have used this system to screen thousands of compounds for their ability to make old immune cells respond more like young immune cells to viral infection. Our findings to-date have revealed beneficial effects from promising therapy candidates for a number of age-related conditions including severe coronavirus disease 2019 (COVID-19).
Phenotypic screening in monocytic cells unravels molecular mechanisms of NLRP3 inflammasome activation
Open to view video.
Open to view video. Immunooncology has now emerged as a successful therapeutic strategy in cancer treatment. Tumour suppression is achieved by T cell mediated cytotoxicity, which relies on T cell primed to recognise tumour antigens. However, current therapeutic strategies do not lead to immunologically active “hot” tumours across and within all tumour indications, and as a result there is the need for novel therapeutic interventions that promote anti-tumour responses in immunologically cold tumours. The NLRP3 inflammasome, a component of the innate immune system, has been identified as an interesting target to convert uninflamed cold tumours into responsive hot cancers. NLRP3 has a major role in the initiation and progression of inflammatory cytokine maturation and release, leading to innate immune induction and anti-tumour response in the tumour microenvironment. Therefore, NLRP3 inflammasome pathway activation is a rational therapeutic target for treatment of certain cancers. Unfortunately, the underlying mechanisms of NLRP3 inflammasome activation and its role in tumour progression are still poorly understood. Elucidation of mechanism of action of modulating agents that lead to inflammation and anti-tumour immune response will be crucial to better inform drug and therapy design. We have generated models of NLRP3 inflammasome activation by developing robust IL18/IL1b cytokine release assays in a human monocytic cell line. Recently we screened a library of small molecule Fully Functionalised Fragments (FFF) in LPS-primed monocytic cells. NLRP3 activation was monitored by assessing the release of pro-inflammatory cytokines in a dose-dependent manner, as well as induction of pyropoptosis, a unique form of cell death. Target deconvolution of screening hits by chemoproteomics pull-down assays identified novel targets potentially involved in NLRP3 activation and cytokine response. Preliminary data indicates that most targets are linked to the fatty acid oxidation (FAO) pathway, which has been reported to have a link with inflammation and the NLPR3 pathway. Moreover, we were able to demonstrate that genetic perturbation by CRISPR of NLPR3 inflammasome complex genes in monocytic cells led to complete inhibition of NLRP3 inflammasome-mediated cytokine release. This optimised method will not only accelerate hit validation from chemoproteomic pull-down assays, but will also enable bespoke pathway screening by using our well-established arrayed CRISPR platform.
Developing an Arrayed Co-Culture Screen for Immuno-Oncology Target ID
Open to view video.
Open to view video. Immunotherapies including PD-L1 blockade have shown remarkable increases in the T cell–directed antitumour response; however, efficacy is seen only in a sub-population of patients. Recently, pooled CRISPR-Cas9 knockout (CRISPRn) screens in tumour/immune co-culture systems have identified a number of genes that confer resistance to T cell killing in pathways including antigen presentation and cytokine signaling, providing insight into tumour mechanisms that cause resistance to immunotherapies. The development of a tumour/immune co-culture system in an arrayed plate-based format would allow the identification of novel targets for immuno-oncology through CRISPR, small molecule and secretome screening, as well as characterization of hits from pooled co-culture screens. Arrayed screens such as these will allow multiple assay endpoints to be measured per treatment, allowing inbuilt orthogonal validation. Here, a small-scale arrayed CRISPRn screen was successfully developed to investigate the effects on a co-culture of T cells and Cas9-expressing PC9 lung adenocarcinoma cells modified to express anti-CD3 antibody on the cell surface (PC9-OKT3 T cell system). A focussed CRISPRn library was designed to target genes involved in known resistance mechanisms (including antigen presentation, cytokine signaling, and apoptosis) as well as genes involved in immune synapse interactions. The viability of PC9 cells was assessed in two-dimensional adherent co-cultures via longitudinal imaging analysis. Knockout of epidermal growth factor receptor (EGFR) and PLK1 in tumor cells cultured alone or with T cells resulted in increased tumor cell death, as expected, whereas knockout of the test gene ICAM1 showed subtle donor-specific resistance to T cell killing. Taken together, these data provide proof of concept for arrayed plate-based tumor/immune co-culture systems and have led to further investigation of in vitro co-culture models.
Applications of functional genomics to improve novel target identification
Open to view video.
Open to view video. Functional genomics is an emerging field of research that aims to deconvolute the link between genotype and phenotype by making use of large -omic data sets and next-generation gene and epigenome editing tools to perturb genes of interest. This area is of major interest to the drug discovery industry, as functional genomic tools, particularly CRISPR-Cas9, can be used to better understand the biological interplay between genes, improve disease modeling, and identify novel drug targets. Incorporation of functional genomic capabilities into the drug discovery pipeline at UCB has facilitated identification and early validation of novel targets across multiple disease areas, and examples of how these capabilities has improved the relevance of cellular models, allowed for interrogation of targeted hypotheses in disease samples, and facilitated the understanding of the mechanism of action of compounds of interest will be discussed.
The Identification of Novel Drug Target Candidates using Pooled CRISPR-Cas9 Gene Editing in T cells
Open to view video.
Open to view video. Advances in functional genomics methods are enabling the identification of novel drug target candidates. The use of CRISPR based screening systems as a functional genomics tool, can be used for both target identification and validation. Such screens provide platforms for the rapid identification of multiple therapeutic target candidates which when targeted result in a desired phenotypic response. When combined with multiparametric readouts, these screens can provide a wealth of data for the drug discovery field. We have developed a pooled CRISPR screen workflow in primary CD4+ T cells with a multiplexed intracellular cytokine readout. This screening workflow will enable the identification of gene targets which, when silenced, cause T cell subsets to elicit immunomodulatory effects. These data will provide information for multiple disease areas when selecting targets for further development.
Expanding the activity-based chemoproteomic toolbox
Open to view video.
Open to view video. Bioorthogonal chemistry is a mainstay of chemoproteomic sample preparation workflows. While numerous transformations are now available, chemoproteomic studies still rely overwhelmingly on copper(I)-catalyzed azide–alkyne cycloaddition (CuAAC) or 'click' chemistry. In this seminar, I will discuss the use of Suzuki–Miyaura cross-coupling for gel-based activity-based protein profiling (ABPP) and mass-spectrometry-based chemoproteomic profiling. Key findings highlighted will include the identification of reaction conditions that proceed in complex cell lysates and a comparison of Suzuki–Miyaura cross-coupling and CuAAC for chemoproteomics, including for target deconvolution studies. Uniquely enabled by observed orthogonality of palladium-catalyzed cross-coupling and CuAAC, I will also discuss applications of this chemistry to dual labeling experiments.
Building the efficient discovery of actionable chemical matter for from DNA-encoded libraries
Open to view video.
Open to view video. DNA-encoded libraries (DELs) provide an exciting opportunity to build diverse chemical collections unbiased from previous organizational therapeutic areas of focus and promises to enable the search for small-molecule ligands to first-in-class targets of academic and industrial importance. While screening DELs and proposing hits from the resultant data has proven to be a straight-forward exercise, the physicochemical properties of resultant hits has often made them disadvantaged vs. those arising from other hit identification technologies that draw on historical chemical collections. Additionally, a poor conversion rate of NGS-identified hits to confirmed ligands off-DNA further complicates the realization of value expected from the technology. We will describe our initial exploration of DEL affinity selection and demonstrate the robustness of the binding assay that underwrites this technology. This talk will also highlight approaches to address the above-mentioned challenges, specifically (1) the incorporation of drug-like properties to the hit proposal process to prioritize the follow up of hits with drug-like properties and (2) our hit validation workstream that includes characterization both on-DNA and off-DNA to limit our investment in series whose initial signal arose from aberrant chemistry during library synthesis. Combining these methods has provided a strategy for DEL application that is delivering value to our organization in the form of actionable chemical starting points that are being taken up by project teams.
Quantitativeness of DNA Encoded Library Technology
Open to view video.
Open to view video. DNA Encoded Library (DEL) technology has been a disruptive selection technology in the drug discovery field for almost two decades. Encoding each of billions of chemical molecules with specific DNA barcodes, the technology can rapidly discovery novel and potent affinity ligands to target proteins. As the technology evolves, many alternative implementations are deployed. However, a consensus method or standard to evaluate compare different platforms is not in place. In an attempt to establish a common language among the DEL practitioners and users, this seminar will analyze several frequently used check points throughout the entire DEL process. The quantification of these check points are often contributing to the quality of the library and the hit prediction generated during selection campaigns.
A generalised framework for rapid, end-to-end automated assay characterisation and analysis using a space-filling Design of Experiments model
Open to view video.
Open to view video. Assay development groups face constant pressure to develop, characterise and optimise robust assays in short periods of time. One successful method used for assay characterisation and optimisation is the implementation of Design of Experiments (DoE). DoE investigates a set of variables (“factors”) within a defined range, and their interactions by applying a statistical framework that creates an experimental “design”. Each design contains a set of experimental “runs”, where each run is made up of different factor combinations. Space-filling designs (SFD) are a type of DoE that sample a “design space” uniformly across each factor level, therefore, making them a suitable method of investigating high-dimensional, non-linear biological systems. However, executing an SFD is significantly more challenging and laborious than executing a traditional DoE design as multiple reagents need to be sampled at a wide range of levels with speed and precision. We used the Antha software platform (Synthace Ltd) to translate liquid handling instructions of an SFD created using JMP (SAS Institute Inc) and executed using a dragonfly® discovery liquid dispenser (SPT Labtech). Our goal was to execute a single SFD to successfully characterise and potentially identify optimised design spaces of a spectrophotometric enzymatic activity assay for three key bioprocessing parameters: initial rate, product yield and enzyme stability. This was done by testing wide ranges of 11 factors critical to an enzymatic assay. These include enzyme, substrate, co-factor, solvent, salt, buffer, and product concentrations, pH, temperature, and salt type. To measure enzyme stability, the enzyme was incubated in reaction conditions that excluded substrates for 8 and 24 hours, prior to kickstarting the reaction. Additionally, every experimental run was replicated 8 times to get a precise measure of noise. This experiment spanned 4 days, requiring >30,000 dispensing actions across thirty six 96-well plates. Time-course absorbance data was processed and analysed by a pipeline created using MATLAB (MathWorks). The MATLAB scripts used the raw data to group, blank-correct and time-correct, and perform outlier analysis on replicate reactions. A model was then fitted to the processed data to interpolate values for initial rate and product yield at each stability level. These responses were added to the original JMP design table and further analysed to produce models that best describe each of the bioprocessing parameters. This body of work represents a generalised framework that can be followed by assay development groups for planning, executing and analysing complex DoEs to rapidly characterise and optimise assays within a matter of weeks.
Strategies for the Discovery and Optimization of Small Molecule Ligands and Targeted Degraders for PCSK9
Open to view video.
Open to view video. Proprotein convertase subtilisin-like/kexin type 9 (PCSK9) is a serine protease that inhibits the clearance of low-density lipoprotein (LDL) cholesterol via a protein-protein interaction with the LDL receptor. Both human genetics and clinical validation support that blocking this protein-protein interaction prevents LDL receptor degradation resulting in a reduction in LDL cholesterol levels. Among many approaches attempted to discover small molecule direct binders of this challenging PPI target, affinity selection/mass spectrometry enabled the identification of one confirmed hit compound. X-ray crystallography revealed that the compound binds in an unprecedented allosteric pocket located between the catalytic domain and the C-terminal domain of PCSK9. To generate ligands with high binding affinity to PCSK9, optimization of this initial hit applied two distinct strategies: structure-based drug design and kinetic target-based ligand optimization. A cellular thermal shift assay was applied to confirm direct target engagement of the optimized compounds in cellular lysates. Finally, the high affinity allosteric ligand was modified with a proteasome recruiting tag and shown to induce protein degradation of PCSK9.
High-throughput flow cytometry in drug discovery: current practice and future prospects
Open to view video.
Open to view video. Flow cytometers are powerful optical instruments that acquire multiple optical signals from cells or other particles of interest as they pass one-at-a-time through laser interrogation spots. Acquisition rates of up to tens of thousands of events per second are possible, and commercially-available cytometers capable of collecting as many as 18 fluorescence and 2 scatter parameters are widely available. Unlike high-content imaging platforms, with which cytometers share some key similarities, acquiring more channels does not require more time. Cytometers can be modified for automated sample acquisition from 96-, 384- and even 1536- well plates, either by transferring samples one at a time into the cytometer’s fluidic system, or by sampling continuously, which results in a single data stream with air bubbles acquired as the probe moves between wells serving to demarcate wells. The former is more common and available commercially on many instruments. The latter is considerably faster. In either implementation, the result is a powerful medium- to high- throughput screening platform with capabilities not easily matched on other instruments. Two elements inherent to cytometer function contribute to the technology’s power. First and most important, acquisition and quantification of many optical parameters from individual particles at very high event rates makes it possible to identify and analyze signals from specific subpopulations within complex samples, enabling multiplexed assays of many kinds and facilitating analysis of the behavior of rare populations within a sample without prior separation or enrichment. Second, as samples are injected into sheath fluid for hydrodynamic focusing, particles traverse the laser interrogation points accompanied by vanishingly small amounts of sample solution, enabling no-wash fluorescent ligand-binding assays that do not depend on anisotropy or polarization. In this talk, I will provide a brief overview of the basics of flow cytometry hardware and software. I will then illustrate the capabilities and limitations of high-throughput flow cytometry using examples from the literature and from my own laboratory’s efforts to develop screens for immune-modulating small molecules and multiplexed assays based on expression of intramolecular FRET sensors. I will outline whether other technologies could have been used, and what the tradeoffs to doing so would be. Finally, I will highlight advances in hardware that might make the technology even more powerful in the future.
Small Molecule Rescue of Phenotypes Disrupted in Autism Spectrum Disorder
Open to view video.
Open to view video. Small Molecule Rescue of Phenotypes Disrupted in Autism Spectrum Disorder Sumitha Rajendra Rao Post-Doctoral Research Associate The Scripps Research Molecular Screening Center Department of Molecular Medicine Scripps Research, Florida, USA Co-Authors Ana Kostic, Virneliz Fernandez-Vega, Kelty Fletcher, Louis Scampavia, Joseph D. Buxbaum and Timothy P. Spicer Autism Spectrum Disorder (ASD) is a neuro-developmental syndrome that is characterized by deficits in social and communication skills with restricted and repetitive behavior. ASD has a high prevalence rate of 1 in 59 children and has no effective drug to treat the core symptoms. Large exome sequencing has identified 102 risk genes that are associated with severe neuro-developmental delay and ASD. Etiological heterogeneity of ASD and unavailability of human neurons remain a major hurdle in understanding the pathophysiology and testing of new drug candidates. Hence, the relevant model to screen potential drugs would be patient specific neurons derived from induced pluripotent stem cells (iPSCs). In this study, we established a cell-based High Throughput Screening (HTS) assay with glutamatergic neurons generated from hiPSCs with ASD mutations (ADNP, FOXP1 and SHANK3) introduced by CRISPR/Cas9 technology. hiPSCs were induced to neurons (iNs), grown in 1536 well plates, and assessed for neurite outgrowth using high content screening. These mutants displayed significant phenotypic differences compared to isogenic control in terms of neurite number and length per neuron basis. We then obtained patient derived iPSCs carrying mutation in the ADNP gene, to demonstrate the real-life pathology. ADNP mutant and control-sibling iNs were screened against 5088 drug-like compounds. Neurite count per neuron was quantified and the average % response per well was calculated using control compounds for neurotoxicity and neuroprotective effects. We found a total of 343 (neuroprotective) and 120 (neurotoxic) hits for the sibling iNs. Similarly, we identified 1236 (neuroprotective) and 123 (neurotoxic) hits with patient derived ADNP iNs. To our knowledge, this is the first study in developing a robust HTS- amenable phenotypic assay for patient derived neurons and testing thousands of drug-like compounds for ASD. Future studies will focus on validating and determining the functional aspects of these compounds on the patients derived neurons thus paving way for potential early drug discoveries.
Automated Analysis of Biphasic PROTAC Dose Response Data at Scale in Drug Development
Open to view video.
Open to view video. A precise, quantitative and reproducible estimation of drug parameters is essential for robust Structure Activity Relationship (SAR) driven drug development. Traditional small molecule screening for inhibitors of the majority of targets yields monotonic dose response curves with a sigmoidal shape, characterised by a plateau at high drug concentrations. The compound is evaluated by fitting to a Hill model, leading to measurements of the maximum effect (efficacy) and the concentration for half maximal effect (IC50). In contrast, some drug modalities lead to a distinct dose response profile, marked by a loss of efficacy after the plateau. Such a “hook” effect is particularly relevant for PROteolysis TArgeting Chimeras (PROTACs). PROTACs are an exciting new modality that induce degradation rather than inhibition of targets, thus enabling the targeting of proteins without defined active sites such as scaffolding complexes. PROTACs induce ternary complex formation between the target, an endogenous E3 ligase and the PROTAC leading to protein degradation. Consequently, at high concentrations, PROTAC assays often exhibit a hook effect due to formation of independent binary complexes. Application of the standard Hill model to data with a hook effect results in misestimation of both IC50 and efficacy. Accounting for the hook effect is essential to extract the best understanding of the data and making informed SAR decisions. We have developed a bell-shaped biphasic model to allow double sigmoidal curve fitting, with parameters to describe both sigmoidal parts of the data and crucially, confidence levels around the efficacy. This algorithm overcomes the challenge of achieving convergence unlike complex non-linear statistical models, making it usable in high-throughput settings ( >100 compounds/run). The advanced algorithm we have developed can thus achieve more accurate estimation of IC50 and efficacy to drive medicinal chemistry SAR. We have implemented the fit method in Genedata Screener for the analysis of high throughput experimental data. This method also features automated selection of the correct model, Hill or bell-shaped, and automatic data masking, as appropriate. Application of this method to real world PROTACs data has demonstrated its ability to reduce manual intervention and deliver accurate curve fit parameters, leading to greater confidence in drug SAR insights. Further, the novel curve fit algorithm is able to quantitatively describe data from non-PROTACs targets too, showcasing its flexibility and applicability across projects independent of the underlying biology.
Enabling High Speed Analysis of High Throughtput Experiment Products by using Pulsed Gas Ambient Ionization of Sub-microliter volume samples
Open to view video.
Open to view video. Analysis of complex and often high concentration samples in viscous solvents associated with synthesis route optimization often requires time consuming sample dilution and LC/MS technology. Rapid analysis in most cases is therefore 1-2 samples per minute requiring considerable time to analyze samples perpared in 96- and 384-well plates. We have enabled direct sampling from source plates of sub-microliter volume samples which are subsequently analyzed is seconds per sample. The Direct Analysis in Real Time Mass Spectrometry based ionization permits near instantaneous thermal desorption and ionization to generate intact molecules as detected by using either positive or negative ion LC/MS instrumentation. Method The control program for a Direct Analysis in Real Time ambient ionization source is modified to release ionizing gas for a short time period in conjunction with the arrival of sample in the desorption ionization region. The thermal desorption profile generated by using the pulse gas method is observed to be more uniform when compared to results obtained from the analysis of the same samples using the conventional continuous gas flow experiment. The pulse gas method generates mass spectra dominated by ions derived from the sample immediately and the relative abundance of the protonated drugs in a mixture is demonstrated to increase by an order of magnitude. The absence of background related ions which are present in the conventional DART mass spectra is notable virtually eliminating the requirement for background subtraction when processing the data. Implications of the use of this method include; reduced the consumption of the non-renewable reagent helium by over 90%, a decrease in analysis time per individual sample, faster analysis of multiple samples, and simplified data interpretation. Results from the analysis of biocatalysis products and synthesis route optimization experiments demonstrate a throughput of 6 minutes for a 96 well plate and only 22 minutes for a 384 well plates. The method is effective for analysis of samples prepared in aprotic solvents such as DMSO making short time turnaround of information to the chemist possible.
Identification and Validation of Drugs that Inhibit SARS2-CoV-2 Entry
Open to view video.
Open to view video. To date the pandemic caused by coronavirus disease 2019 (COVID-19) has nearly 40 million confirmed cases and more than 1.1 miliion deaths have been reported worldwide. Entry of coronaviruses requires the sequential functions of two subunits of spike protein (S), S1 and S2 on the viral surface. S1 subunit or more precisely a smaller domain in this region called receptor binding domain (RBD) is primarily responsible for binding to functional receptors on target cells. S2 subunit facilitates the fusion of cell membrane with viral membrane, and its efficiency is determined by priming steps by cellular proteases as well as following conformational change. SARS-CoV-2, the virus causes COVID19, uses angiotensin-converting enzyme 2 (ACE2) as a functional receptor for entry. Cleavage of S2’ site in S2 subunit by proteases, such as a serine protease named TMPRSS2, demonstrates a priming function and leads to a more efficient entry mediated by SARS-CoV-2 S. Therefore, inhibitors that either block the interaction of S1 or RBD with cellular receptors or the fusion mediated by S2 domain could inhibit entry of the virus. In this study, we developed and initiated a high throughput cell-based screen that identifies inhibitors of the S-mediated entry by monitoring the decrease in a firefly luciferase-based reporter delivered by a pseudotyped viral system. The luminescence signal is dependent on entry efficiency mediated by S, therefore a decrease in signal will indicate inhibition. In order to identify SARS2-CoV-2 specific inhibitors, we used VSV-G pseudovirus which will non-specifically infect any cell. Parallelly, we tested cytotoxicity of selected hits compounds from HTS. We have screened ~15K clinically useful drugs or compounds and identified total of 183 that demonstrated nominal potency (IC50 < 10 µM). Several of these drugs were extremely potent and were further validated in SARS1 and 2 entry assays as well as whole cell SARS2 infectivity assays.
FulcrumSeek: Empowering Rare Disease Drug Discovery
Open to view video.
Open to view video. We launched Fulcrum Therapeutics with a bold vision: To change the course of genetically defined diseases by treating them at their root cause. Our approach to drug discovery generates significant insights into disease biology and allows us to think creatively about the best way to modulate and balance gene expression. Our FulcrumSeek platform resides at the nexus of exciting advances in cellular modeling, high content technologies, pharmacogenomics, and computer science, and is used to generate high dimensional data using 3’RNA-seq and high content imaging to create a database of profiles generated in response to our highly curated and annotated chemical library and functional genomics in complex cell models. We then use computational methods such as AI and machine learning to classify chemical probes and targets, selecting those that can modulate gene expression to treat the known root cause of genetically defined diseases. We demonstrate that this approach can rediscover known targets for genetically defined diseases such as Facioscapulohumeral muscular dystrophy (FSHD) and identify novel targets for future drug discovery.
Binding and activity-based mass spectrometry assays compliment hit identification from DEL screening
Open to view video.
Open to view video. Introduction: In the pursuit of difficult to drug targets in the Oncology space, screening of DNA-encoded libraries (DEL’s) is becoming more prevalent in our drug discovery efforts. DEL’s are highly diverse collections of billions of combinatorially synthesized small molecules attached to large encoding DNA molecules. We are applying DEL technology to sample diverse chemical space distinct from that which is typically represented in small molecule screening libraries, to identify new leads for multiple target families. Methods: Here we report on multiple Mass Spectrometry based biochemical activity and direct binding assays, as well as Surface Plasmon Resonance (SPR), developed to validate and characterize the molecules exemplified from DEL screening efforts. Affinity Selection Mass Spectrometry was implemented as an orthogonal assay to both validate binding of DEL screening hits, as well as to screen compound libraries designed from primary DEL hits. Preliminary Data: In addition, in order to translate compound binding to direct target engagement through enzyme inhibition for enzyme targets, MALDI Mass Spectrometry biochemical activity-based assays were developed to evaluate structure-activity relationships. Utilizing multiple activity based and direct binding assays in parallel, allows us to thoroughly evaluate the mechanism of action of inhibitors derived from DEL screening. Novel Aspect: Using Mass Spectrometry based orthogonal assays to validate the binding observed in DEL screens and it’s translation to enzyme inhibition.
Automated High-throughput ChIP-seq Workflow using Covaris Adaptive Focused Acoustics (AFA) System for Accelerated Epigenetic Drug Screening
Open to view video.
Open to view video. Automated High-throughput ChIP-seq Workflow using Covaris Adaptive Focused Acoustics (AFA) System for Accelerated Epigenetic Drug Screening Jonathan Young1, Tianyao Xu1, Eugenio Daviso2, Jim Laugharn2, Hamid Khoja2 & Alon Goren1,#1 Department of Medicine, University of California San Diego, La Jolla, CA USA2 Covaris Inc. Woburn, MA­# Correspondence should be addressed to Mapping the epigenomic landscape is essential for understanding gene regulatory mechanisms during cell fate changes, disease onset and progression, and during treatment response. Such epigenomic mapping holds huge potential for identification of biomarkers and targets for therapy in general, and epigenetic targets in particular. However, most existing ChIP-sequencing (ChIP-seq) workflows are very time-consuming, difficulty to optimize and standardize for different input material and across different laboratories, and involve many manual steps which makes it inefficient for high throughput screening approaches. To overcome these issues, we combined our validated automated ChIP-seq (Busby et al., Epigenetics & Chromatin, 2016) with the Covaris Adaptive Focused Acoustics (AFA) technology to streamline and significantly reduce the overall workflow time for ChIP-seq. In our study, we tested the ability of AFA to enhance the binding kinetics of antibody-epitope association, improve signal-to-noise ratios, and decrease total processing time. We evaluated a range of epitopes including major histone modifications such as H3K4me3 which is associated with open chromatin and forms narrow peaks, H3K27me3 which is linked to repressed chromatin and binds to wide regions, and H3K27ac associated with open chromatin, and both wide enhancer loci as well as narrow promoter regions. Additionally, we studied the capacity of the AFA to augment the IP step of DNA associated proteins such as CTCF and validated our test conditions using two cell types. Furthermore, we validate each of our test conditions using a variety of cell types. Our preliminary results demonstrate that AFA expedites the immunoprecipitation process to an hour or less while also improving the signal to noise ratio in ChIP-seq, and has the potential to be accelerate similar drug discovery workflows that are based on an immunoprecipitation step, such as MeDIP-seq. Altogether, we have developed a novel use of AFA for accelerated IP that allows enhancing and simplifying ChIP-seq processes. As the process can be performed in a 96 well plate format and thus integrated into other automated workflows, we expect this key advancement would be highly useful for both research, clinical diagnostic applications and discovery pipelines for epigenetics drugs.
Automation Technologies
Array -Based Cellular Assays Miniaturized Beyond the 1536 Well Plate
Open to view video.
Open to view video. Miniaturized assays for high throughput screening (HTS) in drug discovery are typically carried out in microtiter plates using 96, 384, and 1536 well formats. For cellular assays, the lowest volumes achievable may be impacted by the cell types used and the biological outcomes that are measured. Here we present data on an array-based ultraminiaturized format where assay volumes were reduced from uL to nL volumes. Utilizing a high content imaging approach, comparisons to 384-well data showed similar cell health, cell morphology and pharmacological responses in the small volume assays. Challenges included maintaining a humidified environment and handling small volumes for cells and compounds. With the use of more physiologically relevant cell systems such as stem cells and primary human cells, the supply of sufficient material can be a challenge for high throughput drug discovery endeavors. Reducing the volume and numbers of cells required for drug screening can enable the use of better cell systems for drug discovery. Cost savings in reagent usage, and the ability to use disease relevant cell systems are paths toward reduced attrition in drug discovery. As we look to the future to design new drug screening systems, what equipment, device design, and automation will be needed?
Open Source Bio hardware Company Lesson Learned
Open to view video.
Open to view video. 5 areas : Patent Off the Shelf - really shelf Open Source community Supply Chain Sales Channel
Functional Immune Mapping and Drug Discovery with AI-enabled, Image-Based Cellular Phenomics
Open to view video.
Open to view video. Development of accurate disease models and discovery of immune-modulating drugs is challenged by the immune system’s highly interconnected and context-dependent nature. We developed an automated platform for treatment of hundreds of immune perturbations in primary, immune-relevant cell types for use in the identification of dose-dependent, high-dimensional relationships at scale. High-throughput screening on this platform demonstrates rapid identification of hits for TGF-β- and TNF-α-driven phenotypes, and triage of high-risk scaffolds by relating primary screening data with phenotypes from an annotated compound library. The platform can be rapidly adapted for a range of therapeutic models, as demonstrated with models of active SARS-CoV-2 infection (in a BSL-3 environment) and of COVID-19-associated cytokine storm, which surfaced several mechanisms for potential drug repurposing. Recursion's immune profiling and COVID-19 work was released as a library of images and deep learning features for others to explore new AI model building.
Entrepreneurial Experiences with Open Source Components for Clinical Applications
Open to view video.
Open to view video. Clinical laboratory tests directly impact patient care treatment decisions. This is particularly applicable in the field of organ and bone marrow transplantation as the patient’s immune status changes over time. Laboratory automation significantly reduces run-to-run variability resulting in higher quality patient care. Unfortunately, small clinical laboratories do not have the money, time, expertise, or sample number to justify and implement liquid handling automation. The MVO (Minimal Viable Option) was a bootstrapped attempt to use open source software and hardware in conjunction with low cost components to build a cost effective, out of the box solution to address the limitations encountered by small clinical laboratories. The MVO was designed with clinical personnel and constraints in mind. This lecture discusses clinical laboratory considerations, entrepreneurial experiences, use of open source and low-cost components, industry feedback, and market factor considerations encountered during the iterative development of the MVO.
Application of Multiparametric High Content and Pathway Profiling Technologies to Advance Phenotypic Screening Strategies Against Diseases of Unmet Therapeutic Need
Open to view video.
Open to view video. Recent advances in high-content imaging, genome editing, induced pluripotent stem cell technologies and high throughput genomic and proteomic capabilities are converging to stimulate the new discipline of Phenomics Drug Discovery. I will describe how our Phenomics Drug Discovery platform incorporates multiparametric high content imaging and pathway profiling technologies applied across panels of genetically defined cell models to support more mechanistically-informed phenotypic screening. I will present recent data demonstrating how multiparametric high content imaging combined with machine learning methods have enabled a phenotypic screen to stimulate new drug discovery and development programs is oesophageal cancer, a complex disease of unmet therapeutic need. I further describe how NanoString and Reverse Phase Protein Array technologies support deconvolution of the mechanism-of-action of phenotypic hits at the pathway network level. Finally, I will provide a case study of how our Phenomics Drug Discovery platform has enabled the rapid discovery of novel drug combinations which display potent anti-tumour activity across 2D, 3D and in vivo models.
iHEX - The Mobile and Connected Laboratory of the Future is Bridging the Gap Between Chemical- and Bio-Labs Using the Internet of Things
Open to view video.
Open to view video. Digitalization, Automation and Miniaturization currently change the way we live and work. It also affects the daily work in laboratories creating what we perceive as the Lab 4.0 or the Lab of the Future. The disruptive development of new technologies such as open source automation technology, the Internet of Things (IoT) and 3D-printing offer endless possibilities for rapid engineering of new laboratory devices, which are compact, adaptable and smart. In conjunction with automated 3D-image analysis or deep learning algorithms, powerful instruments emerge to create and resolve research data. At the SmartLab systems department of the Technische Universitaet Dresden, Germany approaches towards the laboratory of the future have been developed and implemented. This includes the iHEX system which bridges the gap between conventional chemical and biological laboratories which are static and usually one setup only matches one work process. iHEX adapts to a wide variety of work processes found in both worlds that can be executed on the installed devices. This maintains full efficiency because of the flexible layout. The innovative hexagon design of the tables enables a new way to form your lab. Changes in lab architecture can be done simply by attaching hexagonal static and mobile iHEX elements with the relevant device to form a new working process – iHEX takes care of the workflow process management. iHEX is an interface between the users, devices, control software and the infrastructure itself while it provides for the physical connection between iHEX elements and devices on top, inside or even function integrated combined with the necessary software infrastructure. With function integration we manage to only expose necessary parts of device to the user saving valuable space on the surface. Currently robotic arms, mixers, balances and whole autosamples are integrated into the platform. This provides for a very intuitive workspace equipped with LED status rings which indicate the current status of each iHEX element enabling quick reaction for troubleshooting. The iHEX system - built in the SmartLab systems lab in Dresden, Germany already operates in several labs. The systems brings lab processes to a new level providing intregated data storage and assessment combined with a new approach to laboratory processes.
Accelerating Drug Discovery Using Label-Free MALDI-TOF Mass Spectrometry in uHTS
Open to view video.
Open to view video. Mass spectrometry (MS) is a label-free analytical technique which enables direct quantification of molecules relevant to (patho) physiology, but the low throughput of MS used to be a bottleneck in using this technique for (ultra) high-throughput screening (uHTS). However, the recent introduction of rapifleX MALDI Pharma Pulse by Bruker, which is able to measure up to 10 samples per second, has been a complete game changer in uHTS and early drug discovery space. This technology not only holds the promise of reduced false positives and negatives caused by e.g. fluorescent labels, but also makes the hit-to-lead phase more cost-effective. Pivot Park Screening Centre (PPSC) is happy to announce the integration of rapifleX in its advanced and fully automated robotic uHTS system, handling 1536 samples per plate. During the presentation, we will give you an introduction to ultra-high throughput MALDI-TOF MS, show our fully automated set-up including spotting and measuring the sample plates, present the results of our first screens and discuss the opportunities and challenges of this promising technique.
Pipelines, Databases, Workflows—Making Phenotypic Data FAIR in a Single Facility and a Global Resource
Open to view video.
Open to view video. The UK National Phenotypic Screening Centre (NPSC) provides phenotypic screening services using advanced cellular models for academic and commercial partners based around the UK and Europe. Assays include quantitative live cell assays for sperm motility, bronchoepithelial damage responses, inflammatory responses in skin, and T cell exhaustion, to name a few. All these assays produce large heterogeneous data collections that combine imaging data, and chemical and analytic metadata. The scale and heterogeneity of these data, and the fact that most critical data is stored in proprietary file formats, present acute informatics challenges. To store, process, share, analyse and publish these data, we employ tools from the Open Microscopy Environment (OME; an open-source software consortium that builds data management and access platforms. OME releases specifications and software for managing image datasets and integrating them with other scientific data. OME’s Bio-Formats and OMERO are used in 1000’s of labs worldwide to enable discovery with imaging. OME-TIFF is an open, metadata-rich, multi-dimensional, multi-resolution data format for modern bioimaging that has been widely adopted across the bioimaging community. We will present example workflows and queries showing how Bio-Formats and OMERO can be used to make data Findable, Accessible, Interoperable and Reproducible (FAIR) in the context of a single screening facility. Once data is processed and analysed, it must be published—either within an organisation or if part of a publication, in a public data repository. OME has used Bio-Formats and OMERO to build solutions for sharing and publishing imaging data. The Image Data Resource (IDR; includes image data linked to >90 independent studies from genetic, RNAi, chemical, localisation and geographic high content screens, super-resolution microscopy, single cell profiling, light sheet microscopy of developing organisms and tissues, and digital pathology. Datasets range from several GBs to tens of TBs. Wherever possible, we have integrated image data with all relevant experimental, imaging and analytic metadata. These annotations make it possible to re-use IDR data, and to connect independent imaging datasets by molecular perturbations and phenotypes. We have also built cloud-based analysis tool portals to catalyse the re-use of published imaging data. These include notebooks and Docker containers that package well-known tools like CellProfiler and Ilastik, making it easy to view and interact with IDR data. All these efforts are open source and available for anyone to access and re-use.
At First I Was Blind - Incorporating Vision Into Synthetic Chemical Workflows
Open to view video.
Open to view video. Chemists spend an inordinate amount of time performing low-level tasks based on visual observation. For relatively routine synthetic organic workflow sequences, deploying automation has been quite successful. However, the full potential of integrated automated processes has proven challenging due to the dynamic and ever evolving environment of a chemical research lab. Thus, many processes remain outside the realm of routine deployment of automated technology due to very low perceived return of time investment brought about by lengthy tweaking and adaptation. In many cases, adding simple vision algorithms to control and monitor most of the processes in a standard synthetic lab can unlock the potential of automated workflows, and allow them to be rapidly and efficiently deployed to replace time and labour intensive tasks. Furthermore, computer vision can play an even more impactful role when combined with advanced robotics. In these systems, computer vision inspection provides both qualitative and quantitative data extraction with high precision and accuracy, even for systems that are nearly impossible for humans to resolve. Image analysis algorithms can now transform simple digital cameras into powerful analytical tools for laboratory use. This presentation will introduce Heinsight; a simple, modular vision package which can help to deploy intelligent automation to the dynamic discovery chemical lab.
Organoids: Patient in the Lab
Open to view video.
Open to view video. HUB (Hubrecht Organoid Technology) is a non-profit company that was founded by the Hubrecht institute (KNAW) and the UMCU with the aim to translate the Organoid Technology, invented in the lab of Hans Clevers, to preclinical and clinical applications. Key to the development of the Organoid Technology was the discovery of LGR5+ intestinal adult stem cells by the Clevers lab. With the identification of adult stem cells, we were able to develop a culture system that allowed for the virtually unlimited, genetically and phenotypically stable expansion of the epithelial cells from animals including humans, both from healthy and diseased tissue (Sato et al., Nature 2009, 2011; Gastroenterology 2011; Huch et al., Nature 2013, Cell 2015; Boj et al., Cell 2015). Since the initial development, we have generated HUB organoid models from most organs. Furthermore, the organoids have been shown to be a very powerful tool for the study of Cancer, Cystic Fibrosis and other diseases (Dekkers, Nat Med 2013; van de Wetering, Cell 2015; Drost et al Nature 2015). More recently, we were able to demonstrate that the in vitro response of organoids directly correlates with the clinical outcome of the patient from which the organoid was derived (Dekkers et al., Sci Trans Med 2016; Sachs et al., Cell 2018). In addition, we have now developed a novel system that allows the co-culture of Organoids with immune cells to study the effect of immune modulating drugs.
Tumour Organoid Cultures as a Platform for Functional Genomics
Open to view video.
Open to view video. In an era of genomics-guided precision medicine there is an increasing need for models that reflect the hallmarks of cancer and the molecular diversity of patient tumours. New cell culturing methods are transforming our ability to derive cell models from healthy and diseased tissues, with increased success rates, and linked to patient genomic and clinical data. I will update on our efforts as part of the Human Cancer Models Initiative (HCMI) to create a new biobank of molecularly-annotated tumour organoids as a community resource. Furthermore, I will present how we are beginning to use these organoid cultures for chemical and genetic screen to identify new molecular targets and biomarkers of therapy response. These early studies are laying the foundation for a next-generation tumour organoid functional genomics platform to help guide the development of future precision cancer medicines.
ScanDrop - Single cell function analysis
Open to view video.
Open to view video.
Cellular Technologies
Validation Of An MPS Human Tumor Model: Vascularized Colon Cancer Micro-Tumors Recapitulate In Vivo drug Responses
Open to view video.
Open to view video. Over 90% of anti-cancer drugs that show promise during preclinical studies fail to gain FDA-approval for clinical use. This failure of the preclinical pipeline highlights the need for improved, physiologically-relevant in vitro models that can better serve as reliable drug-screening tools. The vascularized micro-tumor (VMT) is a novel 3D model system that recapitulates the human tumor microenvironment, including perfused vasculature, stromal cells and a complex matrix, all within a transparent microfluidic device that allows for real-time study of drug responses and tumor-stromal interactions. Critically, all nutrients and drugs reach the tumor through living, human blood vessels. We have validated the VMT platform for the study of colorectal cancer (CRC), the second leading cause of cancer-related deaths, by showing that gene expression, tumor heterogeneity, and treatment response in the VMT more closely model CRC tumor clinicopathology than current standard drug screening modalities, including 2D monolayer culture and 3D spheroids. Moreover, based on scRNAseq data we identified the TGFβ pathway as being specifically upregulated in the VMT compared to other in vitro geometries. Consistent with this a TGF-β receptor antagonist uniquely blocked tumor growth in the VMT. These studies demonstrate the potential of vascularized microtissues for improving drug discovery and validation.
A Plug-and-Play Approach to Create Chimeric Antigen Receptor Expressing Allogeneic Cell Therapies
Open to view video.
Open to view video. IO Biosciences has developed a proprietary GEMSTM technology allowing cell lines, including universal therapeutic cells, to be converted into a plug-and-play gene expression system. GEMSTM contains a multitude of pre-engineered CRISPR/Cas9 integration sites allowing for precision-guided gene insertion. Once the source cell is outfitted with the GEMSTM system, single or multiple genes can be integrated in a site-specific fashion. The GEMSTM expression system can be used in biomanufacturing to produce complex proteins, such as monoclonal antibodies, that require the simultaneous expression of multiple genes in CHO cells. Here we will focus on the potential of GEMSTM to express one or more Chimeric Antigen Receptors and how the technology can simplify processes creating these cell therapies and enable rapid portfolio development. Cell therapies have entered a new era with the advent of widely available and constantly improving gene modification techniques. Gene modification of cells allows for genetic properties to be deleted, corrected, or added in a transient or permanent fashion. For example, the addition of chimeric antigen receptors (CAR) to patient’s white blood cells, ex vivo, has led to personalized cell therapies that specifically kill targeted tumor cells in the field of immune-oncology. Several early products show the potential and confirm the dawn of a new era in oncology. Nevertheless, the success of the current therapies also exposed the significant need for improvements in patient access, treatment cost, and the manufacturing process. IO Biosciences uses its GEMSTM technology in combination with a unique, proprietary, trophoblastic stem cell (TSC, Licensed from Accelerated Biosciences). The TSCs are unique, pluripotent stem cells derived from the trophoblast tissue of an ectopic pregnancy. Ectopic pregnancies require removal of the surgical intervention and the discarded trophoblast tissue is harvested while the fetal tissue is sent for pathological review. These ethically sound pluripotent stem cells can differentiate in all 3 developmental tissues (endo, ecto, and mesoderm). Most interestingly, the TCS do have the ability to kill various tumor cell lines and have a favorable immunogenicity profile. Therefore, TSCs are a potentially potent stem cell source for a universal killer cell platform. TSC in combination with GEMSTM creates a powerful off-the-shelf, gene plug-and-play, base for CAR therapies. The presentation will address the advantages of the revolutionary GEMSTM system and our vision of streamlined, off-the-shelf CAR therapy development and manufacturing. Furthermore, we shall present several examples of engineering GEMS into the genome of different types of cells, including pluripotent human trophoblastic stem cells (hTSC) which can be further differentiated into, tumor killing, NK-like cells.
Fully Functionalised Fragments: A New Paradigm In Phenotypic Screening For Rapid Target Identification
Open to view video.
Open to view video. Phenotypic screening faces the challenge of identifying the molecular target(s) of active compounds, especially in cases where the screening hits display moderate-low potency. Recent reports suggest that embedding photoreactive and biorthogonal reporter groups into bioactive small molecules can facilitate the chemical proteomic analysis of protein targets in cells1. AstraZeneca have recently generated a library of small molecule Fully Functionalised Fragments (FFFs), which are pre designed for compatibility with chemical proteomics, enabling a novel capability for both ligand and target identification. The additional advantage of FFFs is a demonstrated ability to bind to a diverse array of protein targets, most of which had no previously known ligands, suggesting the ligandable human proteome is larger and more untapped than previously thought1. The following examples show how deployment of the AZ FFF library has highlighted our early successes in using FFFs to find novel targets for a diverse range of phenotypes including pancreatic β-cell redifferentiation, and downregulation of receptors driving resistance in pancreatic cancer. FFF hits were identified which redifferentiate modified β-cells by re-establishing expression of a key β-cell marker, MAFA. Chemical proteomics identified a set of secretory granule proteins as targets of the active FFF. Subsequently, siRNA knockdown of granin protein SCG3 increased MAFA, validating SCG3 as a potential target to modulate MAFA expression and demonstrating the applicability of this approach as a target ID capability. Application to a diverse range of phenotypic platforms was demonstrated by the identification of structurally distinct FFF hits driving a decrease in a key receptor driving resistance in pancreatic cancer and in cell lines expressing clinically relevant variants of the receptor. Chemical proteomics for both active FFFs identified unique sets of targets with functional links to the receptor biology, further demonstrating the value of FFF as a new target identification capability. We believe FFFs will be especially useful in screens of primary or rare cell types, and are an important complement to our current methods to identify new targets. 1. Parker et al., Ligand and Target Discovery by Fragment-Based Screening in Human Cells, Cell, 2017, 168, 527–541
3D Organotypic Models As Secondary Screening Tool To Identify PARPi Combination Therapies For Non Small Cell Lung Cancer
Open to view video.
Open to view video. Poly(ADP-ribose) polymerase (PARP) inhibitors have demonstrated promising clinical outcomes for ovarian cancer patients and are now increasingly explored for other tumor indications including non small cell lung cancer (NSCLC). However they are limited as a single agent due to eventual emergence of tumor resistance, thus, drug combinatorial strategies are being investigated as alternate viable options to overcome drug resistance. Traditional high throughput screening utilizes tumor cell lines in two-dimensional well plates that fail to recapitulate the complex tumor microenvironment (TME) morphology and do not translate well to <em>in vivo</em> and clinical outcomes. With the ultimate goal of improving clinical success with drug-drug combinations, and, to bridge the translational gap, we employed complex human-derived <em>in vitro</em> models that are adapted to a 384 plate-multi well high throughput assay. We developed a versatile triculture model of NSCLC that incorporates cancer associated cells such as normal lung fibroblasts and normal bronchiolar epithelial cells, and, components of the extracellular matrix (ECM), in the context of developing robust, scalable and functional assays. To assess cytotoxicity tumor cell viability was determined either via cell titer glo or inherent GFP expression in the tumor cells by high content analysis (HCA). This allowed the screening of PARP inhibitors across drug sensitive and resistant cell lines in traditional versus triculture platforms. Taken together, we have developed a unique 3D triculture system with excellent Z’ ( >0.5) and S:B ( >30) that can support a long-term (7 day) combinatorial drug screening approach and highlights the impact of testing in a more complex but translationally relevant cellular model.
Neuronal Cells And Brain Organoids From Patient iPS Cells For Modeling of Lysosomal Storage Diseases
Open to view video.
Open to view video. Neuronal cells and brain organoids from patient iPS cells for modeling of lysosomal storage diseases Wei Zheng, Ph.D., Group Leader National Center for Advancing Translational Sciences, National Institutes of Health, Bethesda, MD, USA. Disease modeling for neurological disorders is a still major challenge. Incomplete representations of human diseases in cell lines and animal models hinder therapeutic drug development. Recent advancements with iPS cell technologies have enabled modeling of neuronal diseases with patient-derived neural cells and brain organoids. We have applied iPS cell derived neurons and brain organoids for modeling of several lysosomal storage diseases. The disease specific phenotypes are present in these neurons and organoids that have been used to evaluate the therapeutic effects of drug candidates and to screen compounds for drug development. These patient derived cells have genetic backgrounds of patients and thus can more precisely model disease-specific pathophysiology and phenotypes. We think that his approach can be used broadly for neuronal disease modeling for drug discovery and development.
Automation Of Neuronal Assays Using New Platform Compatible With High Throughput Screening and High Content Analysis Of Neurons Derived From Central And Peripheral Systems
Open to view video.
Open to view video. Neurological disorders (ND) affect approximately 20% of the world population and are among the top ten leading causes of disability and death. The prevalence of NDs, such as Autism, Multiple Sclerosis and Alzheimer’s Disease, has been increasing every year, causing massive socio-economic impact worldwide. More recently, COVID-19 patients have also been added to the list of patients with ND. To accelerate the development of new therapies targeting these devastating diseases, we have developed technology and tools compatible with High Throughput Screening (HTS) and High Content Analysis (HCA) to grow physiologically relevant models of neuronal tissue in vitro. First, we have developed technology to grow and rewire neuronal networks in vitro 60x faster than in vivo (Magdesian et al., JOVE 2017). Next, we combined this technology with microfluidics to miniaturize cellular assays and to organize human neurons in physiologically relevant networks in Neuro-HTSTM microplates compatible HTS and HCA (optical readout systems, liquid handling robotics). Our award-winning technology enables to reproducibly grow and rewire over 3,000 neurons in Neuro-HTSTM microplates using 90% less reagents and samples. Neuro-HTSTM microplates can be used to organize central and peripheral nervous system neurons derived from humans and rodents. Moreover, our technology enables independent assessment of toxicity to the soma or to the axons. This is much closer to what happens in vivo, when axons innervating the skin and muscles are much more exposed to toxic compounds (e.g. skin absorption), than the soma localized in the spinal cord. The standardized neuronal networks built with Neuro-HTSTM increase plate to plate reproducibility and accelerate acquisition of more predictive results. The Neuro-HTSTM is a versatile tool, enabling robust analysis of over 3,000 axons per plate and reproducible evaluation of thousands to hundreds of thousands of compounds in miniaturized assays. Different from the 3D spheroid models, Neuro-HTSTM microplates offer a rapid, 100% animal free approach to grow human neurons in 2D cultures with single cell precision, ready for HTS and HCA at the single cell level. Here we used Neuro-HTSTM microplates to quantify two parameters impossible to evaluate in HTS using standard multi well-plates: axonal length and neurite connections. We cultured neurons in Neuro-HTSTM microplates with two different media, traditional neuronal culture medium and BrainPhysTM Neuronal medium (StemCellTechnologies). Our results show that traditional medium promoted 31% increase in axonal length, while BrainPhysTM medium increased neuronal connectivity and network formation by 76%. Neuro-HTSTM is the first multi-well microplate to enable compartmentalized testing of axons and soma and precise analysis of axonal growth, regeneration and network formation in HT. The main advantages of our technology are faster acquisition of neuronal data and generation of more predictive data of compounds’ safety and efficacy prior to exposure to humans while reducing animal experimentation.
iPSC-based Models for Testing and Removing Cardiotoxic Liabilities of Drugsvvvvvv
Open to view video.
Open to view video. Heart disease as an unintended consequence of drug treatment is a major concern, especially in oncology where nearly a third of cancer patients experience heart conditions as a consequence of their treatment. Moreover, cardiotoxicity is a major reason for attrition during drug development and drug withdrawals after approval. Patient iPSC-derived cardiac cells could in principle make it possible to determine adverse cardiac effects of drugs early in their development and guide medicinal chemistry optimization to remove off-target liabilities. I will discuss developing patient and genome-edited iPSC disease models and their use integrate patients’ clinical presentations into drug development. I will provide two examples of large-scale medicinal chemistry campaigns for drug optimization: 1) Reengineering of a small molecule antiarrhythmic drug based on the phenotypic response in iPSC-derived cardiomyocytes from patients with a congenital electrophysiological disorder and 2) Reengineering a small molecule cancer therapeutic to decrease its cardiomyopathic liabilities while preserving its anti-cancer effects through parallel testing in iPSC-derived cardiomyocytes, primary vascular endothelial cells, and genome-edited leukemia cells.
Data Science and AI
Modular, not Monolithic: Flexible Laboratory IT Methodologies for Change and Growth
Open to view video.
Open to view video. Laboratories must flexibly adapt to new challenges while maintaining experimental rigor and exceptional quality standards, however many of the software systems that support these efforts are monolithic in nature, making them difficult to change. In this presentation, we will discuss a laboratory IT ecosystem methodology for flexible plug-and-play integration and orchestration of laboratory workflows from sample management, through wet and dry lab protocols, to data processing, analysis, and results using modular components. We will demonstrate the initial application of this methodology for in situ viral strain identification using an Oxford Nanopore Technologies MinION and a previously developed bioinformatics pipeline for next generation sequencing. We will describe the current components of the system including LIMS, workflow orchestration, instrument integration, bioinformatics and data processing pipeline, and data analysis tools. We will detail how each component can be swapped in and out of the ecosystem to support additional workflows, instrumentation, and bioinformatics efforts, among others. Further, we will demonstrate the technology and technical philosophies used to sustain system flexibility for future reconfiguration, innovation, and application to other scientific initiatives. Finally, we will outline the practical strategies and primarily open source technologies used to support and create this modular laboratory IT ecosystem. Attendees will be provided with practical considerations and strategies that can be implemented their own laboratories to promote flexibility and reusability within their existing software ecosystems.
SiLA & AnIML: Lessons Learned From Scaling the Use Of Standards In Digital Transformation
Open to view video.
Open to view video. Standard protocols and data formats can serve as infrastructure enablers for integration of instruments, systems, and seamless data flow. Several initiatives are gaining in this area. We review the progress of the AnIML and SiLA standardization initiatives, which can provide key building blocks to establish interoperability. AnIML specifies an XML-based standard data format for analytical data, while SiLA provides webservice-based communication standards for interfacing with instruments. Two case studies will investigate the use of SiLA and AnIML at different levels of scale. We start with small local deployments on embedded instruments. At the other end of the spectrum, we review a global enterprise deployment for integration of chromatography data. A new digital lab ecosystem is emerging that allows end-to-end integration of instrument control, data capture, and enterprise system (ELN, LIMS) connectivity.
Introducing the Assay Dev Toolkit
Open to view video.
Open to view video. The Assay Dev Toolkit is an application custom-built at Eli Lilly for processing plate-based data. Initial requirements focused on defining plate layouts for ELISA experiments, doing calculations, dose-response curve fitting, plotting, and quantification. Deeper research into the requirements uncovered the fact that tools already existed to perform all the functionality mentioned, including GraphPad Prism, Excel, Genedata Screener, and custom tools from other departments in Lilly. Before development of the application began, there were four different tools available for calculating EC50s. The pain-point for the scientists was not lack of tools, but rather lack of easy-to-use tools. One set of unwritten problems revolved around the manual processes that the scientists had to perform to format their data as proper input for an EC50-calculating tool like Prism. At the other extreme, tools were available that analyzed data with one click if the assay format matched the input format of the tool. If the assay was changed in a way that was not originally parameterized by the tool, a new tool had to be released, resulting in a collection of similarly functioning tools. With this context in mind, we developed an application that has the flexibility of Excel but has a knowledge of 96 and 384-well plates. Any type and amount of data can be captured using customizable data layers. Calculations can be easily performed using formulas where entire layers of data across multiple plates are represented by a single letter. Common transformations are built-in, such as breaking out 384-well plates to 96-well format, converting between plate and table formats, aggregation of replicates, back-calculation of unknowns, and normalization. Curve fitting of 4-parameter dose-response curves and global fitting of equilibrium-binding curves are available. Other analyses can be added in the future.
Automated Science Education During a Pandemic and Beyond
Open to view video.
Open to view video. Biological research is increasingly reliant on automation for the generation of large experimental datasets which require sophisticated computation to analyze and reach conclusions. In fall 2019, Carnegie Mellon launched a Masters in Automated Science program which is specifically focused on training students to use various modern lab automation technologies including liquid handling, automated confocal microscopy, microplate reading, nucleic acid extraction, and QPCR among others. Students are also taught computational techniques to analyze the data generated. Finally, students are taught to use artificial intelligence to control the robots and automate the selection of experiments. In order to give students hands-on experience applying these techniques, a multi-purpose integrated lab automation robot was purchased. This is the first robot of its type to be used primarily for educational purposes. Prior to the shutdowns during the pandemic, all of CMU’s automated science laboratory courses were taught in person with the students working directly on the system to accomplish the goals of the lab activities. Across the world, science classes based on laboratory experience had to be changed substantially to facilitate following pandemic safety guidelines which generally involved abandoning those lab experiences in exchange for video demos or simulations. The availability of the fully integrated lab automation robot presented the unique opportunity to offer a fully remote course for students while teaching students the same skills they would have learned in the laboratory. In order for this to be successful, various problems unique to automated science education needed to be addressed. First, the students needed remote access to the system in such a way as to be able to control as much as they would have been able to in the lab itself. Second, students needed to be able to track the progress of their experiments, detect problems, and debug procedures remotely. Third, students needed effective access to data generated by the robot and systems capable of analysis. Finally, lab exercises had to be redesigned to take into account bottlenecks in procedures which could easily detract from the educational experience of remote students. During this presentation, solutions to these problems as well as new issues discovered will be discussed. Furthermore, student feedback from the experience will be shared.
Cutting out the Middleman: Leveraging Clustering Tools Power to Circumvent Compensation and Facilitate Complex Sorting and Analysis.
Open to view video.
Open to view video. Fluorescence compensation is a process brimming with user-associated error, not amenable to post-acquisition rescue, and is the primary cause of poor interpretation of results. This problem is compounded when a particular cell type must be isolated from a complex heterogenous sample; the panel used to identify the sample often must be adapted for the specific sorter, and users are limited to fewer colors than what is available with analyzers. Given the current unquenchable demand for sorting cells and analysis of heterogenous samples in the rapidly growing fields of immunology, immune-oncology and -omics, plus the increasing need for more depth and higher throughput, compensation continues to thwart the novice cytometrist and slow the integration of cytometric platforms with other methodologies. Circumventing problems associated with compensation significantly reduces barriers for scientists inexperienced with flow cytometry and cell sorting, and promotes widespread adoption of these techniques. Computational methods have been developed and employed for use in flow cytometry, genomics and other high-parameter fields to facilitate analysis for complex and high throughput applications. These tools in principle, do not require compensation for population identification and can be used as a first step for sorting and/or analysis. In addition, if used properly, some of these tools can eliminate data pre-processing steps where data must be manually gated to remove dead cells, debris, and multiplets; as these populations get separated in the final cluster repertoire. Previous work has shown that clustering algorithms can be applied to uncompensated data to identify known populations in a sample. However, identifying unwanted populations, comparing the effects of different transformation settings, and speed considerations for sorting were not evaluated. Herein we tested several recent clustering algorithms available for flow cytometric analysis (including FlowSOM, X-shift and Phenograph) for their ability to (1) cluster unwanted populations, (2) isolate common populations with high precision, (3) identify known rare populations, and (4) evaluate their speed at several sample sizes. A diverse array of publicly-available data was used with samples ranging from 6 to 27 colors and comprised of mouse or human tissues. Importantly, we extended previous findings and compared results from various data transformation settings, as these algorithms are sensitive to linear/non-linear transitions. Surprisingly, we have discovered that data scaling and transformations commonly used for visualizations (including Biex/Logicle, Log, Hyperlog, ASinH) aren’t necessarily ideal for computational processing as they tend to lead to over-clustering in the negative and dim regions. We compared the algorithms’ results with published results using compensated data, and rank the algorithms’ consistency and speed on uncompensated data. Ultimately we propose a new workflow utilizing these clustering algorithms with uncompensated data as a possible starting point for cytometrists with any level of experience.
Drug Target Strategies
A 3D Tumor Model of Resistance of Colon Cancer to Cyclic Treatments with Kinase inhibitors
Open to view video.
Open to view video. Astha Lamichhane, Pradip Shahi Thakuri, Hossein Tavana Department of Biomedical Engineering, The University of Akron, Akron, Ohio 44325, United States Although targeted therapies of solid tumors using mutation-specific molecular inhibitors generate an initial response, cancer cells often adapt to the treatments and develop resistance. Activation of compensatory survival pathways is a major mechanism of drug resistance. Physiologic tumor models of drug resistance are crucial to understand mechanisms of treatment failure and develop more effective treatment strategies to improve outcomes for patients. Using our aqueous two-phase system microtechnology, we developed tumor spheroids of KRASmut and BRAFmut colon cancer cells and cyclically treated them with MEK inhibitors (MEKi) to mimic how patients receive chemotherapy. Our results showed that the tumor spheroids in long-term cyclic treatments develop resistance to MEKi. Considering the established role of cancer stem cells (CSCs) and epithelial mesenchymal transition (EMT) in cancer drug resistance, we investigated whether the resistant cell had these phenotypes. Our molecular analysis of tumor spheroids in cyclic treatments with MEKi showed significant upregulation of several CSC gene markers including ALDH1A3, CD166, and CD133, and enhanced clone forming capability of the cells. Additionally, treatments with MEKi significantly upregulated EMT markers ZEB1 and E47, and promoted invasion of cancer cells from spheroids into a human collagen matrix. To study whether CSC and EMT phenotypes could be suppressed using drug combinations, we identified activation of prominent oncogenic signaling pathways such as PI3K/AKT, JAK/STAT, and WNT/β-catenin in tumor spheroids treated with MEKi. Interestingly and despite a strong anti-proliferative effect of combinations of (1) MEK/ERK and PI3K/AKT inhibitors, and (2) MEK/ERK and JAK/STAT inhibitors, they were ineffective against CSCs and EMT in cancer cells. We are currently evaluating Wnt/β-catenin pathway due to its major role in colorectal cancer. Our objective is to identify drug combinations that suppress growth of tumor spheroids, block compensatory pathway signaling, and inhibit stemness and EMT in cancer cells. Our approach to use 3D tumor model and identify drug combinations that block pathways cross-talk and processes implicated in tumor progression will facilitate treatment selection for validation in animal models and progress to clinical trials.
Unique Set Of Biochemical Assays To Enable Discovery Of Novel Small Molecule SERCA2 Activators
Open to view video.
Open to view video. SERCA (Sarco/endoplasmic reticulum Ca2+-ATPase) is a membrane-bound calcium ATPase and responsible for active transport of calcium (Ca2+) from the cytosol into the lumen of the sarcoplasmic/endoplasmic reticulum (SR/ER). SERCA plays a key role in maintaining intracellular Ca2+ homeostasis and facilitates muscle contractility. Dysfunction or low level of SERCA in cells have been related to several kinds of disease. Thus, activation of SERCA-dependent ER Ca2+-uptake has great therapeutic potential for the treatment of diabetes and heart disease. However, discovery of SERCA activators has been challenging and hindered due to a lack of robust assays suitable for high throughput screening (HTS). In this presentation we will describe a novel screening cascade for finding small molecule SERCA2 activators, and development of a robust in vitro biochemical assay for HTS screening (Z’ factor = 0.5-0.6, CV = 3-4%) using high quality recombinant SERCA2 protein. Orthogonal and functional assays were also developed for validation of SERCA2 activators. The development of the unique set of biochemical assays offers a great opportunity in discovering novel small molecule SERCA activators for pharmaceutical applications.
Targeting RNA With Small Molecules Using Affinity-Selection Mass Spectrometry
Open to view video.
Open to view video. Approximately 60% of DNA is transcribed as RNA, but only 1-2% codes for proteins. Genetic linkage studies of single nucleotide polymorphisms show that the numbers of non-coding RNAs (ncRNA) involved in cellular processes are similar to the numbers of protein coding genes. The function of ncRNA and its interactions with small molecules is relatively unexplored, and understanding the druggability of ncRNA with small molecules will likely open up new therapeutic approaches for various diseases. In order to understand ncRNA-small molecule druggability, we have investigated whether the Automated Ligand Identification System (ALIS) can be validated for detection of RNA-small molecule interactions. ALIS is an affinity-selection mass spectrometry platform capable of high-throughput screening for small molecules that bind to proteins, and has been routinely used to detect hundreds of thousands of protein drug target-small molecule interactions per instrument per day. We have validated the ALIS system with naturally occurring ncRNA bacterial regulatory elements (i.e. riboswitches) and known small molecule drug leads. We next identified over 40 ncRNA sequences from a range of ncRNA classes and disease correlations. Using ALIS, these 40 ncRNA targets have each been screened against chemically diverse small molecule collections, functionally annotated collections from previous phenotypic screens, and collections enriched in RNA-binding properties (60,000+ compounds total). To date, we have generated millions of screening data points from which we are revealing new targets and mechanisms involving small molecule-ncRNA interactions. Here, we outline our approach and results, and discuss their implications for wider small molecule drug discovery efforts.
The Discovery and Translational Development of the KRAS G12C Inhibitor Adagrasib (MRTX849)
Open to view video.
Open to view video. KRAS is a key mediator of growth factor receptor-mediated signal transduction and this signaling mediates cellular growth and proliferation. The RAS family is the most frequently mutated oncogene family in cancer and the majority of these mutations involve KRAS. KRAS is commonly mutated at codons 12,13, and 61. Mutations at codon 12, including KRAS G12C, function by impairment of intrinsic RAS GTPase activating activity and result in increased GTP-bound active state of KRAS mutant protein. This event results in dysregulated cellular signaling and oncogenic transformation of cells harboring mutant KRAS. Iterative strucutre-based drug design, synthesis adn SAR of >2000 molecules and over 100 co-crystal strucutres led to the discovery of MRTX849. MRTX849 was identified as a potent, covalent inhibitor of KRAS G12C that irreversibly and selectively binds to the inactive confirmation and locks KRAS G12C in its inactive state. The blockade of KRAS inhibits cell proliferation, induces apoptosis and demonstrates marked tumor regression in selected KRAS G12C mutant tumors. The ability of MRTX849 to demonstrate tumor cell death and tumor regression is dependent on maximization of KRAS target coverage, maximization of prolonged suppression of KRAS-dependent ERK signaling, and other co-occurring genetic alterations and characteristics of KRAS mutant tumors. High throughput combinatorial small and large molecule combination screens plus, CRISP-based drug-anchored screens provided insight toward feedback and bypass pathways that cooperate with KRAS in oncogenic transformation or modify its oncogenicity. The ability to co-target oncogenic vulnerabilities in KRAS mutant tumors has resulted in identifications of additional pathways that modify KRAS signaling. The study of tumors that exhibit a robust response of MRTX849 monotherapy and study of additional vulnerabilities in KRAS mutated tumors in nonclinical and clinical trials will provide key insight toward optimal strategies to develop this class of inhibitors.
Building a β-propeller drug discovery platform
Open to view video.
Open to view video. β-propeller protein domains modulate diverse protein-protein interactions across human cell types, and these domains are implicated in many disease areas. While specific β-propeller PPIs have been targeted by small molecule therapeutics, a focused investigation of β-propeller properties for drug discovery has not yet been successfully championed. By combining scientific and drug discovery knowledge across diverse technical expertise, Civetta Therapeutics is building a β-propeller drug discovery company. In this presentation, Civetta’s initial efforts to understand and target β-propellers will be shown.
Micro- and Nanotechnologies
Science Ignited Talks
New Modalities
Tony B. Poster Lightning Talks
Metabolic profiling of macrophages, down to single cell resolution
Open to view video.
Open to view video. Macrophages are immune cells with remarkable plasticity that is critical to their function. Macrophage populations at the sites of inflammation or tumors are often heterogeneous; this heterogeneity is reflected in their phenotypes and functions, and has implications for disease progression and response to therapy. An important aspect of phenotypic heterogeneity is due o metabolic plasticity of these immune cells. Here, we conducted comprehensive metabolomic analysis of single macrophages. We used live single cell mass spectrometry to sample, and measure the metabolic profile of individual cells. The results from this study will provide novel insight into macrophage biology.
Clinical Application of HILIC-MRM Method with SERRF Normalization in Healthy and Stable Coronary Artery Disease Subjects Reveals Differences in Levels of Phosphatidylinositols
Open to view video.
Open to view video. Quantification of endogenous biomarkers in clinical studies requires careful evaluation of accuracy, precision, selectivity, specificity as well as reproducibility. Herein, we characterize the performance of a highly multiplex bioanalytical method for quantification of phosphatidylinositols (PI), a class of low-abundance plasma phospholipids. As a negatively charged phospholipid, PI may significantly impact the charge-dependent interactions of HDL with lipases and other proteins. Oral administration of PI was shown to increase HDL-C in normolipidemic subjects. Therefore, careful characterization of the quantitative properties of a high throughput bioanalytical method for the measurement of human plasma PI is valuable. Due to high levels of endogenous PI levels we utilized odd-chain PI species that are not normally present in human plasma as surrogate analytes (SA) to assess assay performance parameters and establish a definitive dynamic linear range for PI lipids. A high throughput method was developed, qualified, transferred to an automation platform and applied to sample testing in two clinical trials in healthy (NCT03001297) and stable Coronary Artery Disease (CAD) subjects (NCT03351738). The method employs hydrophilic interaction chromatography (HILIC) for separation of various phospholipids based on the head-group and multiple reaction monitoring (MRM) for targeted multiplex quantification. The extraction method was optimized based on the physicochemical properties of PI to simplify the procedure and to improve reproducibility. The extraction procedure employed isopropanol ? a less toxic and volatile solvent. The method was evaluated with comparison to other lipid extraction methods, followed by assessments of method linearity with both SA and endogenous PI, as well as extraction recovery at several concentrations. The method demonstrated acceptable precision and accuracy (±30%) over linear range of 1 - 1000 nM for SA and dilutional linearity (8-fold) for endogenous PI. To correct for batch effects, Systematic Error Removal using Random Forest (SERRF) normalization algorithm was employed. Moreover, we employed SERRF to bridge the raw values between the two clinical studies, enabling quantitative comparison of absolute values (Figure 1). As part of the batch effects correction evaluation, we examined the most suitable quality control (QC) level. We determined that mean-adjusted average QC performed best (Figure 2). The method was transferred to the BRAVO automation platform (Agilent) to further improve its reproducibility and enable testing of ~1000 samples as part of the clinical trial NCT03351738. Introduction of automation significantly improved method reproducibility (Figure 3). The comparison of the two studies revealed that healthy subjects levels of PI are consistently higher across PI species compared to CAD subjects (Figure 3). Treatment with MEDI5884, endothelial lipase neutralizing monoclonal antibody being developed for the treatment of coronary artery disease by increasing HDL quantity and function restores near-normal levels PI in CAD patients on intensive statin therapy. Further characterization of the underlying biological mechanisms responsible for the decrease of the PI biomarker in CAD patient population relative to healthy subjects as well as in conjunction with pharmacological intervention by MEDI5884 may reveal more information on this clinically-relevant biomarker. Importantly, approaches presented within this work exemplify the development, characterization and qualification of analytical methods for the quantification of clinically-relevant lipid and metabolic biomarkers using highly multiplex, metabolomic-style methods not amenable to full validation as prescribed by regulatory guidance. Furthermore, this work demonstrates how advancements and integration in several areas of technology such as liquid chromatography, mass spectrometry, automation and data processing can enable discovery of clinically relevant pharmacodynamic biomarkers that can inform our understanding and treatment of cardiovascular disease.
Precision Medicine Technologies