The pharmaceutical industry plays a key role in driving precision medicine, a reality that will require all possible information sources and data being integrated, analyzed, and interpreted across departments, organizations, and geographical regions. Integrated Pharma Informatics will discuss the challenges related to data from a variety of sources, including clinical trials, imaging, sequencing, EMRs, and personal sources such as wearables and social media. This conference will examine a number of real-world projects in data integration and analysis and translational research, driving toward precision medicine projects.
Monday, February 20
10:30 am Conference Program Registration Open
11:50 Chairperson’s Opening Remarks
Tom Plasterer, Ph.D., Director, US Cross-Science, AstraZeneca
12:00 pm All-in-One or One-for-All? Reckoning with the Diversity and Commonality of Research Information Technologies
Peter Covitz, Ph.D., Senior Director, Research & Translational IT, Research and Development IT, Biogen
Biopharmaceutical research is increasingly enabled by information technology. At the same time, life science companies are under increasing pressure to boost R&D productivity without adding significant cost. One example is the choice between consolidating around a limited number of integrated platforms that receive higher levels of investment versus supporting a greater variety of point solutions that serve specific niche requirements. This presentation will discuss the trade-offs inherent in such choices and make some suggestions on which way to lean for different situations.
12:30 How Highly Integrated Informatics Platforms Contribute to Our Drug Portfolio
Juergen Hammer, Ph.D., MBA, Global Head Data Science, Pharma Research and Development Informatics, Roche Innovation Center New York
In a world where genetics, sensor, real-world and image data increasingly influence clinical decision-making, well-designed and highly integrated informatics platforms supporting structured data capturing, integration, and analytics become essential for effective drug development. I will discuss how these informatics platforms are being applied at Roche. Furthermore, I will discuss some of the principles in designing these platforms, and contrast our current approach to previous approaches in biomedical informatics.
1:00 Session Break
1:10 Luncheon Presentation: Agile Data Access to Speed Discovery of Scientific Insights
Quan Nguyen, Senior Director, Client Technology Solutions, Certara
A key bottleneck in discovery and development is the ability to access and understand complex biological, chemical, pre-clinical data from many sources. Thousands of scientists rely on D360 Scientific Informatics Platform to access and analyze data for better informed decisions.
2:10 Session Break
2:30 Chairperson’s Remarks
Ajay Shah, Director, Research Informatics and Systems, Office of Chief Informatics Officer, Beckman Research Institute, City of Hope National Medical Center
2:40 Edge Informatics and FAIR (Findable, Accessible, Interoperable and Reusable) Data
Tom Plasterer, Ph.D., Director, US Cross-Science, Research & Development Information (RDI), AstraZeneca
Edge Informatics is an approach to accelerate collaboration in the BioPharma pipeline. By combining technical and social solutions, knowledge can be shared and leveraged across the multiple internal and external silos participating in the drug development process. This is accomplished by making data assets findable, accessible, interoperable and reusable (FAIR). Public consortia and internal efforts embracing FAIR data and Edge Informatics will be highlighted, in both preclinical and clinical domains.
3:10 Enabling Genomics Beyond the Pipelines
Boris Umylny, Ph.D, CTO, Smpl Bio
As genomic analysis gains widespread commercial acceptance, it is important for analytical tools to evolve from their current siloed setup to become an integral part of the corporate computational environment. In this talk we introduce Kratos – a cutting-edge application that combines capabilities for bulk as well as single-cell NGS data analysis with secure, world-wide collaboration and unparalleled ability to integrate and thrive in real-world setting.
3:40 Integrated Informatics Approaches to Enable Immuno-Oncology
Yun Li, Head, R&D Informatics, R&D Business Technology, Pfizer San Francisco
The advancement of Immuno-Oncology brings new hope for cancer patients but also presents new challenges in several areas of Informatics. This presentation will highlight some of the approaches R&D Business Technology at Pfizer has taken to overcome those challenges.
4:10 Integration from the Ground Up: Transforming Biologics R&D Informatics with Benchling
Sajith Wickramasekara, Founder & CEO, Benchling
Most R&D processes are scattered across disparate software. Benchling unifies experiment workflows, ensuring that cutting-edge science is never held back by obsolete software. We will describe how we worked with scientists to streamline biologics R&D workflow on a single platform.
4:25 The Paradigm Shift in Integrated Informatics
Marc Siladi, Senior Business Analyst, Core Informatics
This presentation will discuss the convergence between big data and drug discovery driving the neo “Cambrian Healthcare explosion.” This vertex is creating a new landscape with new approaches and tools for healthcare including informatics, analytics, and data management strategies.
4:40 Refreshment Break and Transition to Plenary Session
5:00 Plenary Keynote Session
6:00 Grand Opening Reception in the Exhibit Hall with Poster Viewing
7:30 Close of Day
Tuesday, February 21
7:30 am Registration Open and Morning Coffee
8:00 Plenary Keynote Session
9:00 Refreshment Break in the Exhibit Hall with Poster Viewing
10:05 Chairperson’s Remarks
Boris Umylny, Ph.D, CTO, Smpl Bio
10:15 Universal Registration: Centralized Data Management and an Intelligent Network of Relationships between Records
Farida Kopti, Ph.D., Director, Chemistry/Pharmacology/HTS Informatics IT, Merck Research Labs, Merck & Co.
Merck intends to capture and secure the intellectual property around our various entities in the course of drug discovery. Hierarchical and other pipeline driven relationships can be identified between these entities, plus direct links to assay test results depicting the functional activity of the entities. This will enable traceability, lower the cost associated with duplicate production of the entity, and allow scientific data to be accessed in real time across the research organization. Scientists will be able to make decisions based on a more complete set of data that is not broken down by application silos.
10:45 Alexion’s Content Analysis Project: Mining Content for Actionable Insight
Martin Leach, Vice President R&D IT, Enterprise Data Management & Analytics, IT, Alexion Pharmaceuticals
Unlocking content from internal and external sources is key for many different use cases. Competitive analysis, business development, external scouting, patient identification and just making sense of the vast amount of information within an organization. In our presentation, we will share some of our findings and methods for doing this at Alexion using off the shelf technology and approaches with advanced data visualizations that help explain information sources.
11:15 SPIRIT-SA: Machine Learning Platform for Scientific Data Analysis
Ajay Shah, Director, Research Informatics and Systems, Office of Chief Informatics Officer, Beckman Research Institute, City of Hope National Medical Center
SPIRIT (Software Platform for Integrated Research Information and Transformation) seeks to integrate basic, clinical and translational data and analytic tools. SPIRIT-SA (Scientific Analytics) component of SPIRIT platform provides data normalization, data cleanup, and results validation using multiple machine learning algorithms simultaneously. The latest version of SPIRIT-SA guides the users to the most suitable algorithms for their dataset and the ideal visualization methods.
11:45 Aligning Data Analytics for Translational and Personalized Medicine
Dan Weaver, Ph.D., Senior Product Manager, PerkinElmer
PerkinElmer Signals for Translational helps discover, analyze & align high-dimensional datasets with clinical outcomes. We will elaborate on functional components, simplified data loading model supporting structured and unstructured data and TIBCO Spotfire visualizations designed to answer Translational Researcher’s questions.
12:15 pm Session Break
12:25 Luncheon Presentation(Sponsorship Opportunity Available) or Enjoy Lunch on Your Own
1:25 Refreshment Break in the Exhibit Hall with Poster Viewing
2:00 Chairperson’s Remarks
Peter Covitz, Ph.D., Senior Director, Research & Translational IT, Research and Development IT, Biogen
2:10 Connecting Tumor Genomics with Therapeutics through Multi-Dimensional Network Modules
Sourav Bandyopadhyay, Ph.D., Assistant Professor, Bioengineering and Therapeutic Sciences, University of California San Francisco
Recent efforts have catalogued genomic, transcriptomic, epigenetic and proteomic changes in tumors, and a challenge is to integrate these data to identify a consensus catalog of the unique molecular events or modules in cancers and connect them with new therapeutics. We present a new approach called MAGNET to integrate such data based on functional networks. Evaluation of network modules in cancer cell lines reveals a preserved subset that can be used for biomarker development.
2:40 ADC Target Identification and Validation
Sean Liu, Ph.D., Global Head, Translational Science Systems, Research IT, Takeda Pharmaceuticals
Takeda developed an analysis pipeline to use publicly available microarray and RNASeq data for antibody-drug conjugate (ADC) target identification and data integration and visualization platform to validate and prioritize targets based upon biological relevance and druggability. This presentation will discuss considerations and criteria for ADC target identification and public data sources to address each of the criteria.
3:10 Drug Target Validation Using Genomics and Data Mining
Peng Yue, Associate Director, Research Bioinformatics, Gilead Sciences
Target validation using in vivo animal models is a critical step in drug discovery at the pre-clinical stage. However, the poor predictive value of animal models often correlates with the lack of efficacy in clinical trials. The ever-growing collection of big data offers a complementary way to validate the biological role of a given target in human. This talk will discuss the challenges and potential of such an in silico way for target validation.
3:40 New Approaches to Target Identification
Richard K. Harrison, Ph.D., CSO, Clarivate Analytics (Formerly the IP & Science business of Thomson Reuters)
This talk will highlight the use of omic and pathway based analysis for identifying new targets and the use of new analytical tools and analytics to rapidly triage targets to identify the most promising from both scientific and commercial perspectives.
4:10 Hollywood Oscar Dessert Reception in the Exhibit Hall with Poster Viewing
5:00 Breakout Discussions in the Exhibit Hall
These interactive discussion groups are open to all attendees, speakers, sponsors, & exhibitors. Participants choose a specific breakout discussion group to join. Each group has a moderator to ensure focused discussions around key issues within the topic. This format allows participants to meet potential collaborators, share examples from their work, vet ideas with peers, and be part of a group problem-solving endeavor. The discussions provide an informal exchange of ideas and are not meant to be a corporate or specific product discussion. Pre-registration to sign up for one of the topics will occur a week or two prior to the Event via the App.
Target Validation Using Genetics and Genomics in the Era of Precision Medicine
Peng Yue, Associate Director, Research Bioinformatics, Gilead Sciences
- What are the new opportunities provided by precision medicine?
- What are the major challenges?
- What would be the solutions?
Operational Requirements for Big Data Solutions
Ajay Shah, Director, Research Informatics and Systems, Office of Chief Informatics Officer, Beckman Research Institute, City of Hope National Medical Center
- Data quality and standards; privacy and country-specific concerns
- Hardware and architecture – where to put it and how to put it
- Ownership, management and governance
Why is Data Integration So Challenging?
Tom Plasterer, Ph. D., Director, US Cross-Science, Research & Development Information (RDI), AstraZeneca
- Connecting Data (making it accessible)
- Describing Data (making it reusable)
- Data Governance (making us all Data Scientists)
6:00 Close of Day
Wednesday, February 22
7:00 am Registration Open
7:00 Breakfast Presentation (Sponsorship Opportunity Available) or Morning Coffee
8:00 Plenary Keynote Session
10:00 Refreshment Break and Poster Competition Winner Announced in the Exhibit Hall
10:50 Chairperson’s Remarks
Peng Yue, Associate Director, Research Bioinformatics, Gilead Sciences
11:00 Learning Real-World Evidence of Drug Efficacy and Safety from the EHR
Nigam Shah, Ph.D., Associate Professor of Medicine (Biomedical Informatics) at Stanford University, Assistant Director of the Center for Biomedical Informatics Research
With the widespread availability of Electronic Health Records (EHR), it is possible to examine the outcomes of decisions made by doctors during clinical practice to identify patterns of care-generating evidence from the collective experience of millions of patients. We will discuss methods that transform EHR data into real world evidence for comparative effectiveness, drug safety and Phase IV surveillance studies for a learning health system.
11:30 Bioinformatics Approaches for Functional Interpretation of Genome Variation
Kai Wang, Ph.D., Associate Professor, Psychiatry and Preventive Medicine, University of Southern California
We developed Phenolyzer, which analyzes clinical phenotypes on a given patient and predicts the most likely candidate genes that are responsible for the phenotypes, by integrating multiple sources of gene-pathway-disease-phenotype information. Based on Phenolyzer, we also developed iCAGES (integrated CAncer GEnome Score), which is an effective tool for prioritizing cancer driver genes for a patient using genome sequencing data. We illustrate case studies where iCAGES can facilitate selection of optimal treatment strategies based on predicted personal driver genes.
12:00 Visualization, Characterization and Mining of Real-World Patient Data
Andreas Matern, Vice President, Partnerships & Innovation, BioReference Laboratories, GeneDX
Real World Patient Data (RWPD) is plagued with a lack of data management strategy. In this talk, I will discuss the construction of a data repository and visualization tools used to mine and characterize RWPD from clinical patient records. The discussion will include overcoming the complexities of RWPD, modeling the data for use in clinical and pharmaceutical research, and visualizations and data mining techniques used to allow end users to interrogate the data in ways never before possible.
12:30 Enjoy Lunch on Your Own
1:10 Refreshment Break in the Exhibit Hall and Last Chance for Poster Viewing
1:50 Chairperson’s Remarks
Farida Kopti, Ph.D., Director, Chemistry/Pharmacology/HTS Informatics IT, Merck Research Labs, Merck & Co.
2:00 Leveraging ‘Omics Data from Deeply Phenotyped Clinical Studies to Inform Target and Biomarker Validation
Janna Hutz, Ph.D., Senior Director, Head, Human Biology & Data Science Engine, Eisai AiM Institute, Eisai, Inc.
Beyond oncology, there have been few documented successes in using genome scale sequencing from clinical trials to inform design of subsequent trials. Rather, it is emerging that these datasets’ greatest value may lie in feeding back into earlier stages of drug discovery. I will share Eisai’s efforts to use NGS data from well-characterized clinical cohorts for target validation and biomarker identification.
2:30 Disease Signatures to Drug Discovery
Deepak K. Rajpal, Ph.D., Director, Computational Biology-Target Sciences, GSK
We present how we have used clinical transcriptomics based generation of disease signatures and their application in drug discovery. We have identified disease areas of interest and then clinical transcriptomics datasets from published literature associated with the diseases of interest. We have then generated disease signatures and by integrative informatics approaches, and have applied these datasets in our drug discovery efforts. We present here a case study in dermatological disease area.
3:00 Target Identification and Validation Using Genomics and Genetics
Vinod Kumar, Ph.D., Senior Scientific Investigator, Computational Biology (US), Target Sciences, R&D, GSK
In practice, the identification of a novel disease target is an integrative step combining many lines of evidence, but may often be triggered by a key, highly publicized finding. Though relatively little attention has been paid to systematically evaluate the multiple lines of evidence that have proven effective in choosing a successful target for that disease. I will present how we use informatics approaches to leverage genetic, genomics and phenotypic data to prioritize targets and validate them experimentally.
3:30 Session Break
3:40 Chairperson’s Remarks
Andrew C. Fish, J.D., Executive Director, AdvaMedDx
3:45 Big Data – The Devil’s in the Details
Mike Barlow, Vice President, Operations, MolDX Executive Lead, Specialty Contracts, Palmetto GBA
Linking effective therapies and expanded trial designations are the expected benefit of the ever expanding capabilities of genomic biomarker and gene expression identification. More and more data is being generated every day. Keeping that data ‘valuable’ will require we maintain a critical focus on the quality and comparative values of the data, especially in the area of genomics and more specifically outcomes. Other questions will arise around where the data is collected, how it is curated, and who has access. As a Medicare payer, we support the concept of data collection/aggregation if that data can be effectively mined to create ever improving treatment protocols and more importantly improved outcomes.
4:00 Efficiently Leveraging Commercial and Open Source Bioinformatics Tools for Clinical Interventions and Research Discoveries from Very Large Datasets
Ben Busby, Ph.D., Genomics Outreach Coordinator, NCBI, NLM/NIH
In precision medicine, it is often the case that efficacy does not depend on the appropriate computational intervention, but on the morphology of the data that informs the problem. For example, different strategies should be employed when calling short variants in stable versus unstable regions of the human genome, or when looking for pathogenic effectors in well-characterized versus newly discovered bacterial or viral pathogens. Pragmatic solutions from existing commercial and open source resources will be presented.
4:15 From Bits to Bedside: Developing a Learning Digital Health System to Evaluate Pigmented Skin Lesions
Dexter Hadley, M.D., Ph.D., Assistant Professor, Pediatrics, Institute for Computational Health Sciences, University of California, San Francisco
Melanoma accounts for less than one percent of skin cancer cases but the vast majority of skin cancer deaths. Early screening and diagnosis significantly improves patient outcomes, yet no systematic framework exists for clinical evaluation of common pigmented lesions for risk of melanoma. In this talk, I will describe our approach to develop a learning health system focused on precision screening and diagnosis of skin cancer. We first leverage medical students in the clinics to capture digital samples of pigmented skin lesions at scale with mobile health technology. We then leverage this clinical big data to train state-of-the-art deep learning image classification algorithms to better screen for cancer. Our work translates routinely documented electronic health data into an impactful digital health initiative to directly improve clinical outcomes and advance clinical knowledge.
4:30 PANEL DISCUSSION
5:15 Close of Conference Program