Archived Content

High Content Analysis - Day 1


Day 1 | Day 2 | Day 3 | HCA 2012 Final Brochure | Live Cell Imaging

 

Tuesday, January 10, 2012

Sponsored by
Thermo Scientific
7:45 am-2:00 pm Thermo Scientific User Group Meeting

 

 

Sponsored by
Molecular Devices
2:00-6:00 pm Molecular Devices User Group Meeting

 

 

Sponsored by
GE Healthcare
2:30-6:30 pm GE Healthcare User Group Meeting

 

 

5:00-6:30 pm Conference Pre-Registration

 

Wednesday, January 11, 2012

7:00-8:00 am Conference Registration and Morning Coffee

8:00-8:10 Welcoming Remarks from Conference Director

Julia Boguslavsky, Cambridge Healthtech Institute


Sponsored by
Thermo Scientific
8:10-8:15 Introduction from Executive Sponsor
Jeffrey Haskins, Ph.D., R&D Director, LSR-Cellomics, Thermo Fisher Scientific


8:15-9:15 Panel Discussion with End-Users and Vendors

Moderator: Anthony M. Davies, Ph.D., Director, Irish National Center for High-Content Screening and Analysis (INCHA)

Discussion Questions Include:

  • What new applications and image analysis tools are expected to be launched in 2012?
  • What is the progress on imaging standards?
  • How is the need for new probes and protein-protein interaction imaging tools addressed?
  • What are the new application areas undergoing rapid adoption?

Vendors:
Liz Roquemore, Technology Manager, Cellular Applications, GE Healthcare Life Sciences
Grischa Chandy, Ph.D., Product Manager, Cellular Imaging Marketing, Molecular Devices, LLC.
Mark A Collins Ph.D., Director, Global Marketing, LSR-Cellomics, Thermo Fisher Scientific
Jacob Tesdorpf, Ph.D., Director, High Content Instruments & Applications, PerkinElmer

End-Users:

Paul A. Johnston, Ph.D., Research Associate Professor, Department of Pharmaceutical Sciences, School of Pharmacy, Drug Discovery Institute, University of Pittsburgh School of Medicine

Joe Trask, Ph.D., Head, Cellular Imaging Core, The Hamner Institutes for Health Sciences

Eberhard Krausz, Director, Assay Development and Target Validation, Janssen Research & Development

Sponsored by
Thermo Scientific
9:15-10:30 Networking Coffee Break in the Exhibit Hall with Poster Viewing




HCA for Toxicity Assessment

10:30-10:35 Chairperson’s Opening Remarks

Joe Trask, Ph.D., Head, Cellular Imaging Core, The Hamner Institutes for Health Sciences

10:35-11:00 Assessing Potential Etiology for Delayed Renal and Hepatic Toxicity in HAT Clinical Compounds Using HCI

Joe Trask, Ph.D., Head, Cellular Imaging Core, The Hamner Institutes for Health Sciences

Human African trypanosomiasis (HAT) is a devastating disease in regions of sub-Saharan Africa and reached epidemic levels in Uganda in 2008.  Unfortunately to date there are limited safe and easy-to-administer therapeutic inventions with good efficacy for first-line defense against the disease.  Antitrypanosomal compounds such as pentamidine, an aromatic diamidine used for over 60 years, is effective but has known toxicities and requires parenteral administration.  The development of an oral diamidine prodrug, pafuramidine maleate (DB289) showed great promise during Phase II and III clinical trials, before an expanded Phase I trial indicated unexpected delayed renal and liver toxicities in healthy volunteers, which halted further development.  Understanding the underlying mechanisms involved in delayed renal and liver toxicities are key components to further developing and delivering next-in-class drugs to combat HAT and similar diseases such as pneumocystis pneumonia and malaria.   I will discuss present research findings by means of high-content imaging to phenotype cellular responses and identify perturbations that may contribute to drug-induced organ injury following compound treatment using primary human liver and primary human renal proximal tubule epithelial cell models.

11:00-11:25 HCA for Predictive Cytotoxicity Testing in Pharmaceutical and Food Industries

Peter J. O’Brien, D.V.M., Ph.D., Veterinary Clinical Pathologist, Pathology, University College Dublin

HCA applied to HepG2 human hepatocytes in a cytotoxicity model is effective for discriminating between drugs that induce liver injury and those that are generally accepted as safe, and is widely used in the pharmaceutical industry. In one of the first evaluations of HCA in the food industry, this same approach applied to human intestinal and liver cells is shown effective at screening for and assessing comparative toxicity of dietary components and additives.

11:25-11:50 High-Content Imaging for Compounds that Impair Mitochondrial and Lysosomal Function

Sashi Nadanaciva, D.Phil., Principal Scientist, Pfizer Global R&D

Toxicity is a major reason for compound attrition in pre-clinical drug development and post-market drug withdrawals. This talk will provide a general overview on two mechanisms of toxicity, mitochondrial impairment and lysosomal impairment. Compounds that impair the synthesis of either mitochondrial DNA (e.g. nucleoside reverse transcriptase inhibitors) or mtDNA-encoded proteins (e.g. antibiotics that target bacterial protein synthesis) cause mitochondrial impairment leading to various adverse effects. A high-content imaging assay that identifies compounds that impair mtDNA-encoded protein levels will be described. A high-content imaging assay that identifies compounds that accumulate in lysosomes, the cell’s recycling center, will also be presented.

11:50-12:15 High-Content Analysis for High-Throughput Screening of Genotoxicity

Bonnie Goodwin, Ph.D., MPH, Biologist, NIH Center for Translational Therapeutics, NHGRI, NIH

Identification of environmental and pharmaceutical compounds with the potential to induce DNA damage is an important step in investigating compound safety. As part of the Tox21 program, we utilized high-content analysis to classify genotoxicity caused by different chemicals. We present here several high-content assays, including in vitro micronucleus assay and phospho-H2AX detection, which have been adapted to high-throughput screening in 384- and 1536-well format. With these assays we are able to characterize genotoxic compounds that induce micronuclei formation and DNA strand breaks.

High-Content Image Analysis

10:30-10:35 Chairperson’s Opening Remarks
Jeffrey Haskins, Ph.D., R&D Director, LSR-Cellomics, Thermo Fisher Scientific

10:35-11:00 High-Content Data Analysis: Challenges and Solutions

Michael Lenard, Solutions Architect, Discovery Automation, Bristol-Myers Squibb

The complexity and size of high-content screening data may affect the performance of data analysis and therefore should be taken into account when building new or integrating existing data analysis tools. In addition, while there are many commercial analytical tools available for high-content screening, it is often beneficial to extend “out of box” functionality to support customer-specific data analysis scenarios and provide seamless and efficient integration with data management platforms. The presentation focuses on how these problems can be addressed.

11:00-11:25 Robust Pattern Recognition Methods for the Classification of Image-Based Phenotypes: Application to Angiogenesis-Imaging Assay Screens

John Raymond, Ph.D., Senior Research Scientist, Discovery Informatics, Lilly Research Laboratory,
Eli Lilly

It can be argued that assays based on phenotype identification may provide a more meaningful proxy for therapeutic endpoints than the traditional biochemical assay, but comparative, technical hurdles remain such as greater experimental variation in phenotype results and a lack of quantitative resolution capable of supporting customary chemical structure optimization (i.e., PC50 phenotype-concentration curves). Borrowing heavily from the face and pattern recognition literature, we present a robust, phenotype assay paradigm based on machine learning that is capable of probabilistically quantifying multiple phenotype classes (e.g. multiplexed endpoints) as well as identifying novel phenotypes, indicating either previously unobserved phenotypes or concept drift. An application of the proposed methodology to the analysis of ten combinations of growth factor-induced angiogenesis based on the Acumen imaging platform is presented.

11:25-11:50 High-Content Analysis Recipes: Flat Field Correction, Machine Learning and Multi-Parametric Regression Models

Peter Horvath, Ph.D., Head, Data Analysis Group, Light Microscopy Center, ETH Zurich

First I will present a novel, reference imageless technique to correct illumination problems of high-content microscopic images. The technique is based on automated background and foreground detection directly from the images. Experimental comparisons show that the proposed method outperforms the currently used correction techniques. In the second part of my talk I will present our new findings using machine learning techniques to more precisely decide between known cell phenotypes and identify unknown cell types using semi-supervised learning and cell population context. The “Suggest a Classifier” (SAC) package, a new high-content analysis recommendation tool, will be briefly introduced. Finally, machine learning-based regression algorithms and their usefulness for HCA will be discussed.

11:50-12:15 PhenoRipper: A Novel Platform for Rapid Image Analysis

Steven Altschuler, Ph.D., Associate Professor, Pharmacology and Green Center for Systems Biology, University of Texas, Dallas

Here we present PhenoRipper, an easy-to-use image analysis software package that enables rapid exploration and interpretation of high-content microscopy data. PhenoRipper requires minimal user input or training, effectively making it a turnkey solution for the comparison and classification of cellular perturbation assays.

Pathway Analysis with HCA

10:30-10:35 Chairperson’s Opening Remarks

Hakim Djaballah, Ph.D., Director, HTS Core Facility, Molecular Pharmacology and Chemistry Program, Memorial Sloan-Kettering Cancer Center

10:35-11:00 Dissecting Oncogenic-Induced Signaling Pathways Using HCA

Hakim Djaballah, Ph.D., Director, HTS Core Facility, Molecular Pharmacology and Chemistry Program, Memorial Sloan-Kettering Cancer Center

Oncogenic activation is one of the hallmarks of transformed phenotype resulting in foci formation. While many attempts to study this phenomenon have been made thus far, they have as yet to be fruitful. We have developed a simple high-content assay platform to study the consequence of signaling pathways in a transformed phenotype induced by an oncogene as compared to the parental phenotype; the platform was optimized and validated against two oncogenes. In this talk, I will describe the development and validation of this high-content assay and discuss the merits of using whole-well imaging together with its applications to both chemical and RNAi screening.

11:00-11:25 Differentiating Cellular Phenotypes in HCS of in vitro Models of Human Disease

Susanne Heynen-Genel, Ph.D., Director, High-Content Screening Systems, Conrad Prebys Center for Chemical Genomics, Sanford-Burnham Medical Research Institute

The analysis of the discovery strategies of newly (1999-2008) FDA-approved drugs, recently published in Nature Reviews, suggests that phenotypic screening approaches are the main contributor in discovery of first-in-class drugs. At the same time, image-based high-content phenotypic screens have become more widely used in medium- and high-throughput screening of chemical probe and drug discovery. These multi-parametric screen readouts allow more detailed analysis of cellular phenotypes than single parameter assay readouts, which may provide additional phenotypic information of biological or disease relevance. This presentation will showcase examples of differentiating phenotypes in image-based screens investigating in vitro models of human disease.

11:25-11:50 Comparison of a High-Content Cell Count and Cell Cycle Assay with Energy Metabolism-Based “Proliferation” Assays

John Moffat, Ph.D., Scientist, Biochemical and Cellular Pharmacology, Small Molecule Drug Discovery, Genentech

Among the most commonly used methods for measuring drug effects on cell proliferation are ATP-dependent luminescence and mitochondrial reductase activity (e.g. MTS). Using high-content assays we have observed that significant deviations from the assumed linear relationship between cell number and ATP or MTS assay readout can arise with agents targeting the cell cycle. Furthermore, the deviations can vary significantly between different drug mechanisms of action and cell lines. By simultaneously quantifying mitochondrial mass per cell we found that much of this discrepancy is due to drug-induced changes in cell size and/or mitochondrial mass.

11:50-12:15 Superiority of Dye-Based Assay over Antibody-Based Assay in Multidimensional Pharmacological Response Studies

Tiao Xie, Ph.D., Senior Image Analyst, IDAC, Harvard Medical School

In an effort to study multidimensional pharmacological responses across a large panel of human tumor cells to anti-cancer drugs, we developed and compared several antibody- and dye-based assays. While the conventional antibody-based assay allows more flexibility and easier image analysis, its handling incurs loss of cells, especially apoptotic cells. By comparison, the dye-based assay involves no washing. It provides more accurate cellular phenotypic scoring, and is quicker, easier, and more economical. Combined with a customized image analysis method, this dye-based approach enables us to collect a multidimensional pharmacological response signature for a better understanding of drug mechanisms.

 

 

Luncheon Technology Showcase: High-Content Screening

Sponsored by
Molecular Devices
12:30-1:00 Accelerating Your High-Content Screening Work Flow

Grischa Chandy, Product Manager, Molecular Devices, LLC

Evan F. Cromwell, Ph.D., Director, Research, Molecular Devices, LLC

Accelerate your discovery with our latest advances in high-content screening, as we present data on assays for drug discovery and toxicity testing:  high resolution, 2-color GPCR activation (Transfluor® Assay) that doubles data generation rates and increases the assay window;  phenotypic neuronal toxicity assay with improved statistics and reduced sensitivity to cell plating artifacts;  3-color multi-parametric cytotoxity assay that measures both cytoskeletal and mitochondrial integrity and was accelerated to assess greater than 10 million cells/hour.

Sponsored by
GE Healthcare
1:00-1:30 Advances in HCA for Drug Safety Assessment and Screening

Liz Roquemore, Technology Manager, Cellular Applications, GE Healthcare Life Sciences
Cytotoxicity screening is becoming one of the most prevalent HCA applications, particularly in the pharmaceutical industry, where the negative impact of drug recalls and late-stage failures is driving the need for earlier more predictive toxicity testing.  HCA can provide a more sensitive and robust means of evaluating cytotoxic potential than conventional approaches, while providing additional information about mechanism of toxicity.  Results from a live-cell toxicity screen demonstrate the power of HCA for early cardiotoxicity assessment.

Sponsored by
Thermo Scientific logo
1:30-2:00 Analyzing Complex Biologies Using Multi-Dimensional High-Content Imaging

Scott Keefer, Manager, Product Management, Thermo Fisher Scientific
Richik Ghosh, Ph.D., Director, Applications, Thermo Fisher Scientific
Explore the cell without limits with the new Thermo Scientific Arrayscan Infinity HCS Reader. We will present several case studies on stem cells, angiogenesis and other biologies using our latest platform, which integrates an innovative confocal module and a range of software enhancements to provide a variety of multi-dimensional imaging capabilities. We will demonstrate that these enhancements extend the benefits of high-content into new areas of cell biology, drug discovery and toxicology without sacrificing productivity, flexibility or ease-of-use.

Luncheon Technology Showcase:
High-Content Data Analysis

Sponsored by
Gene Data
12:30-1:00 A Corporate HCS Knowledge Platform for Systematic Analysis and Management of Images, Single-Cell Data, Compound and Biological Results

Jon Tupy, Ph.D., Head, Professional Services, Genedata, Inc.

We present an HCS knowledge platform serving as a central hub for all HCS images and data produced in an organization, across different HCS imagers and image analysis packages. It performs standardized QC and analysis, from single cells to end results, and manages data and images for enterprise-wide access and interpretation.

Sponsored by
DeNovo Software
1:00-1:30 Analysis and Reporting for High-Content Screening with FCS Express 4 Image Cytometry by De Novo Software

David Novo, Ph.D., President & CEO, De Novo Software

De Novo Software has developed premier analysis and reporting tools for research and clinical flow cytometry for a decade.  Leveraging this experience, we developed an image cytometry package for HCA.  A flow cytometry analysis environment for image cytometry data allows fully interactive data and image review at a single-cell level.  Create sophisticated reports with 1-click.

Sponsored by
Fluofarma
1:30-1:45 Take-off on the HCS Data Wave with Cytosurfer®

Victor Racine, Ph.D., Team Leader, Biocomputing Division, Fluofarma
CytoSurfer® by Fluofarma is designed for the cell biologist to easily perform comprehensive HCS data analysis. Image datasets produced by BD-Pathway, Metamorph, ImageJ are visualized as population scattergrams and quantified using interactive gates with real-time update (plots, stats, images). It includes 96/384 wells plates capability, heatmap, dose response, IC50, ZFactor, t-test and multi-plates processor.

 

 

High-Content Screening

2:15-2:40 High-Content Screening in Lead Discovery

Marjo Götte, Ph.D., Research Investigator III, Lead Finding Platform, Novartis Institutes for Biomedical Research

In lead discovery, high-content screening becomes a more and more important tool for hit and lead identification. Challenges in high-throughput screening in primary screening as well as recent experience in screening projects will be presented. Utilizing sophisticated image analysis algorithms as well as multi-parametric data analysis will be discussed.

2:40-3:05 Facing the Challenges of HCS Implementations into an HTS Workflow

Dana Nojima, Ph.D., Senior Scientist, Discovery Technologies, Amgen

Image-based HCS technologies enable recording and analysis of 2-dimensional structural information that is inaccessible to conventional screening technologies. Often this information is either essential for detecting biological processes or at least facilitates understanding and description of these processes. However, due to the generation of high data volume, the throughput of HCS systems is limited. As a result, the storage and analysis of the recorded data contribute to temporal constrictions. Recently substantial progress has been made automating the computational analysis of the corresponding data, resulting in a much higher throughput and making implementation of HCS systems into the HTS workflow possible. Here we describe the setup and application of an automated and integrated HCS system in the context of a high-throughput screening campaign. Using a receptor internalization assay, we completed a small molecule library screen to identify compounds that inhibit the formation of the initial receptor-ligand complex. Assay optimization, screen development and results of the HTS will be discussed, in addition to the automation and IT infrastructure challenges we faced over the 6 months of implementation.

3:05-3:30 How to Deal with All the Data when Organizing Comprehensive HCA Support in Pharmaceutical Research: A User’s Perspective

Stefan Prechtl, Ph.D., Senior Scientist, LDB-Screening, High-Content Analysis, Bayer Pharma AG

After 10 years of assay development, our HCA lab provides numerous comprehensive image-based cellular assays that are optimized to Bayer Pharma specific drug discovery projects. Our HCA lab supports early target discovery and validation projects focusing on siRNA K.D. studies and cDNA overexpression studies as well as primary screening campaigns analyzing 3 million compounds. Extensive functional and mechanistic profiling of drug candidates during lead evaluation and lead optimization phase is our main task, but phenotypic biomarker profiling during late pre-clinical studies is also supported by HCA approaches. Our HCA assays consist of classical multiplexed fixed cell culture assays, extensive living cell time-lapse studies and even compound impact studies by utilizing 3-D spheroid technology. The setup of our image storage and backup system, the setup of our HCA network for constant and fast image flow, and the setup of our image analysis and data retrieval tools are vitally important to ensure a consistent and reliable project support.

High-Content Image Analysis (continued)

2:15-2:40 An Automatic Overlap-Based Cell Tracking System

Joe Chalfoun, Ph.D., Research Scientist, Bio-Robotics, ITL, National Institute of Standards and Technology

An overlap-based cell tracking system that has the ability to track multiple cell lines across a set of time-lapse images based on the amount of overlap between cellular regions in consecutive frames will be presented. It uses the overlap to identify mitotic cells and has the ability to detect touching cells and separate them by feedback from tracking to segmentation. This cell tracker is tested on manually segmented and tracked NIH 3T3 data set with accuracy over 95%.

Sponsored by
GE Healthcare logo small
2:40-3:05 Workflow for Improved Hit Identification with a 1536-Well Plate Screen Run on GE IN Cell Analyzer

Robert Graves, Ph.D., Senior Applications Scientist, GE Healthcare Life Sciences
High Content imaging platforms like GE IN Cell Analyzer now provide sufficient sample throughput to enable large-scale screens to be performed.  We describe a 1536-well format screen to identify chemical modulators of a reporter of tumor cell metastatic phenotype.  The workflow is described from imaging with IN Cell Analyzer, image analysis with IN Cell Investigator and image and data archiving using IN Cell Miner.  Software tools such as data filters and supervised machine learning algorithms to identify positive hits versus artifacts derived from dispensing and washing inconsistencies, sample toxicity or autofocus errors are highlighted

3:05-3:30 Automated Construction of Generative Models from Time Series Cell Images: Tools for More Complete Analysis of Perturbagen Effects

Robert F. Murphy, Ph.D., Professor, Computational Biology and Biological Sciences, Biomedical Engineering, and Machine Learning; Director, Ray and Stephanie Lane Center for Computational Biology, Carnegie Mellon University

The complexity of cell responses to perturbation has led to increased collection of time series images in high-content analysis. While variations on traditional features can be used for analyzing spatiotemporal subcellular patterns, the problems of comparing results between assays and systems that are well known in static imaging are intensified with time series. We have therefore developed approaches for building a generative model from a collection of time series images to describe how a particular component distributes itself within cells over time. The model parameters can be used to compare different targets or the same target under different conditions.

HCA FOR FUNTIONAL ANALYSIS AND RNAi

2:15-2:40 RNAiAtlas Database: Libraries and Off-Target Prediction Network for HCS RNAi Experiments

Karol Kozak, Ph.D., Head, Data Handling Unit and High Content Screening, ETH Zurich

RNAiAtlas allows us to find genome-wide scale specific siRNA information (sequence and metadata) related to target gene (search possibility). RNAiAtlas is regularly updated according to new RNAi annotations provided by the RNAi suppliers and also updated based on bioinformatics databases. RNAiAtlas gives an opportunity to navigate between different annotation versions of specific siRNA. RNAiAtlas visualizes interactions between siRNA oligo and predicted off-targets.

2:40-3:05 Functional Gene Discovery Using RNAi-Mediated Gene Silencing in Host-Pathogen Interactions

Elizabeth Hong-Geller, Ph.D., Staff Scientist, Bioscience Division, Los Alamos National Laboratory

To identify host proteins essential for pathogenicity, we have established a platform for high-throughput screens employing RNA interference and have identified 18 host kinases that are required for suppression of Yersinia-induced NF-kB-regulated gene expression. We demonstrate that kinase inhibitors to SGK and AKT block intracellular growth of fluorescently-labeled Burkholderia in macrophages using the Amnis Imagestream. We are developing an in-house automated confocal microscopy strategy to screen kinases that are required for pathogen-induced formation of multinucleated giant cells. Identification of host genes that are targeted by multiple bacterial species can enable the development of broad-spectrum therapeutics.

3:05-3:30 Clustering and Classification to Identify Hits in High-Content RNAi Screens

Rajarshi Guha, Ph.D., Informatics Scientist, NIH Center for Translational Therapeutics

High-content RNAi screens generate multidimensional signals that provide a rich description of phenotypes. We have developed a hit selection framework that employs a random forest classification model to identify genes with negative or positive control-like phenotypes and then a series of tiered models that refine these groups into finer-grained subpopulations. We also enrich gene clusters with GO annotations to prioritize genes for follow-up based on their high-content “signal,” cluster membership and biological function. I will discuss an application of this to a screen for identifying genes involved in DNA rereplication, highlighting biologically sensible gene rankings and reconfirmation rates.


Sponsored by
Thermo Scientific
3:30-4:45 Networking Refreshment Break in the Exhibit Hall with Poster Viewing



 

High-Content Screening (continued)

4:45-5:10 Talk Title to be Announced

Robert Singleton, Ph.D., Lead Research Investigator, Sanofi

5:10-5:35 Development, Validation and Implementation of the 97,000 Compound HCS Campaign to Identify Selective Inhibitors of the STAT3 Signaling Pathway in a Head and Neck Squamous Cell Carcinoma Cell Line

Paul A. Johnston, Ph.D., Research Associate Professor, Department of Pharmaceutical Sciences, School of Pharmacy, Drug Discovery Institute, University of Pittsburgh School of Medicine 

The Signal Transducers and Activators of Transcription (STATs) are transcription factors that mediate the effects of growth factors and cytokines to regulate the expression of target genes involved in cell proliferation, differentiation, inflammation, migration and apoptosis. Activated STAT3 is an oncogene that directs tumor cells toward proliferation and survival, induces angiogenesis to alter the tumor microenvironment, and promotes tumor metastases through its effects on cell migration and invasion. In sharp contrast, activated STAT1 is considered a tumor suppressor because it is a potent inhibitor of tumor growth, promotes tumor cell apoptosis, and enhances anti-tumor immunity. Therefore, a selective inhibitor of the STAT3 pathway would be a highly desirable goal for the development of an anti-cancer drug. The development and implementation of a 97,000 compound HCS campaign to identify selective inhibitors of the STAT3 signaling pathway in a head and neck squamous cell carcinoma cell line will be described.

5:35-6:00 Evaluation of High-Content Assay Technology Platforms to Re-Initiate Drug Discovery Targeting Protein-Protein Interactions

Eberhard Krausz, Ph.D., Director, Assay Development & Target Validation, Janssen Research & Development

Protein-protein interactions (PPIs) are an integral part of many processes in the cell. Discovery of drug-like modulators that disrupt or stabilize such interactions are of tremendous value to understand protein-protein interactions and ultimately address correlating diseases. However, targeting PPIs is not a “low-hanging fruit,” and many disappointments have been reported. Nevertheless, with innovative assays and more complex library molecules covering new chemical space, we should give it another chance. Here, a number of imaging-based technologies such as the fully-reversible Fluorescent 2-Hybrid System or the Proximity Ligation Assay will be presented to screen for PPI-disrupting compounds or analyze protein-protein interactions intracellularly. The challenges, advantages and disadvantages will be discussed in the context of case studies that have been run at Janssen Pharmaceutica or in collaboration with external partners.

High-Content Data Management

Chairperson's Remarks

Caleb Foster, Product Manager, Cellomics Software, Thermo Fisher Scientific

4:45-5:10 Image Analysis, Data Management and Statistical Learning in High-Throughput, High-Content Screening: A Case Study
Xian Zhang, Ph.D., Research Investigator, Novartis Institutes for Biomedical Research
In drug discovery we regularly perform high-content screening with a library of more than one million chemical compounds. These screening projects face common issues of high-content analysis, as well as unique challenges in high-throughput. With a large-scale screening project as an example, this talk will present our current pipelines for image analysis and processing, data flow and management, and multi-parametric analysis and visualization. Various state-of-the-art software tools, informatics infrastructures and statistical algorithms, as well as their advantages and challenges will be presented and discussed.

5:10-5:35 A Data Management Framework for High-Throughput, High-Content Screening

Matthew Smicker, Research Investigator, Research Data Management, Sanofi

The development of affordable and reliable HCS instrumentation has outpaced the evolution of software tools that perform sophisticated management, analysis and interpretation of results. Our data management framework will be described with an emphasis on recent progress using openBIS, an open-source HCS data management application. With openBIS, results from HCS instrumentation and image analysis algorithms are robustly managed and associated with critical experimental metadata such as project, bioassay parameters and test substances. An evaluation was recently completed and has led to a decision to implement at two sites within our organization.

5:35-6:00 Enterprise IT Services and Data Management for Large-Scale High-Content Research

Normand J. Cloutier, Ph.D., Support Manager for Scientific Applications, Research Informatics Services, Bristol-Myers Squibb

In large-scale pharmaceutical research for high-content screening, additional services and processes need to be established to handle both the complexity of software and volume of data involved. Delivering the variety of different software and services needed, by the many different possible profiles of users, presents a significant management challenge for traditional IT departments. At BMS, we utilize HCS in traditional compound screening as well as large-scale genomics work (from 96-well to 1536-well formats) using three distinct HCS platforms. We have accommodated this diversity and developed a framework that delivers these tools and services to scientists in a manner that scales with our different research areas. This talk will describe areas where we have strategically placed resources to address the issues that align with the most impact to the research.

Neuronal Imaging

4:45-5:10 High-Content Analysis of Primary Neurons to Uncover “Control Node Kinases” in Axon Growth

Vance Lemmon, Ph.D., Professor, Basic Science Research Labs, Miller School of Medicine, University of Miami

Protein kinases are part of a regulatory network controlling neuronal differentiation and axon growth.  Existing data suggest that the output of this network is mediated by a relatively small number of effectors, indicating that axon growth could be increased by identifying and manipulating activities of a few critical kinases. Identification of such “nodes” in the molecular network that controls axon growth is difficult as the network and its dynamics are poorly described.  We are using high-content analysis of primary neurons to investigate the activities of a large collection of protein kinase inhibitors (KIs), to determine which KIs promote axon growth.  Although the KIs hit multiple targets, we are uncovering relevant targets by combining data from neurite growth regulatory activities, kinase-inhibitory activity profiles, and in silico modeling of compound interactions with kinase domains. The targets we identify will be used for iterative testing with new and more focused KIs.

5:10-5:35 High-Content Screening to Uncover a Novel Mechanism for Antipsychotic-Induced Metabolic Dysfunction: Modulation of the TGFβ Pathway

Fred Levine, M.D., Ph.D., Professor and Director, Sanford Children’s Health Research Center, Sanford-Burnham Medical Research Institute

Antipsychotics cause metabolic side effects. Through a high-content screen, we identified antipsychotics as insulin promoter modulators. A high-content counterscreen of a library of small molecule inhibitors identified the TGFβ pathway as being involved, finding that antipsychotics activated SMAD3, a TGFβ pathway effector, through a receptor distinct from the TGFβ receptor family.  Of note, antipsychotics that do not cause metabolic side effects did not activate SMAD3. In vivo relevance was demonstrated by analysis of SMAD3 responsive genes from published data on human brains treated with antipsychotics. This work raises the possibility that antipsychotics could be designed that lack metabolic side effects.

5:35-6:00 NeuriteQuant: An Open Source Toolkit for High-Content Screens of Neuronal Morphogenesis

Shelley Halpain, Ph.D., Professor, Division of Biological Sciences, University of California, San Diego

Most physiologically relevant cell culture models of neuronal development display high cell density and overlapping cellular structures, which poses challenges for morphology-based screens. We therefore created a novel bioinformatics pipeline called NeuriteQuant, which facilitates automated morphological analysis of large-scale image data. The toolkit automates the quantification of cell bodies, neurite number, neurite length, and branching. An efficient, automatically generated web-based data browser facilitates data evaluation. NeuriteQuant is based on the open source analysis tool ImageJ, and is thus readily extended or customized for specific applications. We performed proof-of-concept automated screens for modulators of neuronal development in primary neurons and neuronally differentiated stem cells.

Sponsored by
Thermo Scientific
6:00-7:15 Reception in the Exhibit Hall with Poster Viewing



Day 1 | Day 2 | Day 3 | HCA 2012 Final Brochure | Live Cell Imaging