2018 Archived Content
Track 4: Software Applications and Services

Track 4 explores the technology and tools that are used to connect data, applications, people, processes, and partners to ensure available, reliable, and actionable information for scientific decision making. Case studies will be presented that address how life science organizations address common problems in utilizing data including analytics, methods and standards, using open source, semantic technology, using in-house vs. customized commercial platforms, transparency, efficiency, security, and cost-effective solutions.

Tuesday, May 15

7:00 am Workshop Registration Open (Commonwealth Hall) and Morning Coffee (Foyer)

8:0011:30 Recommended Morning Pre-Conference Workshops*

W2. Common Statistical Mistakes to Avoid for Data Scientists

12:304:00 pm Recommended Afternoon Pre-Conference Workshops*

W13. Lab Informatics: An Insider’s Guide to Project Succes

* Separate registration required.

2:006:30 Main Conference Registration Open (Commonwealth Hall)

4:00 PLENARY KEYNOTE SESSION (Amphitheater & Harborview 2)

5:007:00 Welcome Reception in the Exhibit Hall with Poster Viewing (Commonwealth Hall)

Wednesday, May 16

7:00 am Registration Open (Commonwealth Hall) and Morning Coffee (Foyer)

8:00 PLENARY KEYNOTE SESSION (Amphitheater & Harborview 2)


9:45 Coffee Break in the Exhibit Hall with Poster Viewing (Commonwealth Hall)

Cityview 2

10:50 Chairperson’s Remarks

Varshal K. Davé, Vice President, Sales & Marketing, L7 Informatics

11:00 Data Architecture and Integration in Pharmaceutical Research Pathology: Guiding Principles, Solution Architecture and Technology Choices

Rohit Sharma, Senior Bioinformatics Architect, Bioinformatics and Computational Biology, Genentech, Inc.

The current Software application that supports Research Pathology at Genentech is a bespoke monolith, with no standards around data collection, metadata management, master data management, or data integration. As data needs grew over the years, the application got abused in numerous places with unstructured user-input fields being overloaded by users for additional data collection. With no data integration architecture in place, it had become almost impossible to track all the systems extracting or exchanging data with this application. As we replace it with the next generation Pathology Informatics solution, we decided to have an upfront data architecture. In this session we will explore the Data Architecture of this new solution, with special emphasis on: • Guiding principles • Metadata management with data standards as input and industry standard tools • Master data management for Ontologies and Dictionaries • Evolutionary Architecture • Use of state-of-the-art technologies like data virtualization to implement the architecture

11:30 Allotrope@Bayer – Enabling Disruptive Ways of Collaboration & Innovation through Harmonization

Henning Kayser, R&D IT, Scientific Development IT, Bayer

Being a founding member of the Allotrope Foundation - granted the 2017 Bio-IT World Best Practice Award in Knowledge Management - Bayer is implementing strategically the Allotrope Framework to escort scientific data throughout its complete lifecycle from method development over acquisition, processing, reporting, archiving to submission, using a common set of standard tools. Allotrope aims to make the intelligent analytical laboratory a reality – an automated laboratory where data, methods and hardware components are seamlessly shared between disparate platforms, and where one-click reports can be produced using data generated on any analytical instrument. Allotrope's vision of an intelligent analytical laboratory will be realized through the creation of an open "ecosystem" in collaboration and consultation with vendors and the analytical community. Tangible examples how this became already reality at Bayer and will shape our future lab environment will be presented.

 Cognizant 12:00 pm Preparing the Lab Today to Become the Digital Lab of the Future

Arvind Ramakrishnan, Venture Leader, Cognizant Lab Insights, Life Sciences, Cognizant

Guardant Health12:15  pm Enterprise-Class Software Engineering - A Practical Path to Agility and Scalability in Life-Science Industry

Hamed Ahmadi, Director, Software Engineering & IT, Information Technology, Guardant Health

Life Science is solving some of the world’s hardest problems. While software is typically an integral part of the solution, a few companies fully embrace innovative technologies and best practices in developing software & services. This talk breaks down the technical, cultural and industry challenges and offers a path to unlock the full potential of the industry using Enterprise-class software development.

12:30 Session Break


12:40 Luncheon Presentation I: Democratizing Access to the Broad’s GATK Best Practices Pipelines, Optimized and Certified across Platforms

Geraldine Van der Auwera, Associate Director, Outreach and Communications, Data Sciences and Data Engineering, Broad Institute

As genomic data generation explodes, so does the need for workflows that are scalable, reproducible across infrastructures, and empower researchers to apply cutting-edge analysis methods. We democratize access to such workflows by providing versions of our production pipelines optimized for a range of platforms and priorities (e.g. cost vs. speed), validated by our methods developers to ensure scientific equivalence.

1:10 Luncheon Presentation II (Sponsorship Opportunity Available)

1:40 Session Break

Cityview 2

1:50 Chairperson’s Remarks
Rich Lysakowski, PhD, Senior Business Analyst & Informatics Engineer, Astrix Technologies; Professor of Bioinformatics & Data Science, Network Technology Academy Institute

1:55 Lab & R&D Informatics Systems Selection - Quality Practices
Rich Lysakowski, PhD, Senior Business Analyst & Informatics Engineer, Astrix Technologies; Professor of Bioinformatics & Data Science, Network Technology Academy Institute

2:25 How Editas Is Using Modern Software to Transform R&D
Michael Dinsmore, Associate Director, Informatics, Editas Medicine

Lab7 Informatics2:55 Operational Informatics: A Whole Lab Approach to Actionable Data Insights

Christopher Mueller, PhD, CTO, L7 Informatics

From initial observations through groundbreaking discoveries, data powers science. However, many labs generate more data than they can handle. New opportunities - “that’s odd” moments - are easily lost in the clutter. An operational approach to data management allows seamless tracking of all lab data, ensuring analyses that yield new insights are not just possible but routine.

Tetrascience RED3:10 Leveraging Cloud-Based Platforms to Drive Data Strategy for the Life Sciences

Alok Tayi, CEO, TetraScience

Life science companies want to accelerate drug discovery using data analytics and machine learning. Scientific data, however, is not centralized nor standardized and is fragmented: from instrumentation to CRO/CMOs to legacy software. Here, we will discuss how biopharma companies are advancing their data strategies by deploying new data platforms.

3:25 Refreshment Break in the Exhibit Hall with Poster Viewing (Commonwealth Hall)

Cityview 2

4:00 Advancing Clinical Pharmacology with an Analytics-Based and Performance-Driven PK/PD Platform

Taylor Hamilton, Manager, Emerging Technology, Janssen R&D IT Innovation Enablement, Johnson & Johnson

Johnson & Johnson Global Clinical Pharmacology’s High Performance Pharmacometrics Platform combines standard and custom tools into a system for population PK/PD evaluations that’s both manageable to validate and flexible to change. In this presentation, see how the team responded to the needs of pharmacologists and pharmacometricians to create a system with both cloud-burstable computing and an intuitive UI.

4:30 Enabling Cloud-Based GxP Applications for Modeling and Simulation in Clinical Pharmacometrics

Jobst Loeffler, PhD, IT Operations, Scientific Development IT, Bayer Business Services GmbH

The adoption of Cloud-based computing services by life-science companies is increasing at a fast pace. In terms of reliability and security Cloud services are at least comparable to on-premises services but in terms of flexibility and scalability Cloud services outperform what is possible using on-premises infrastructure. In addition, commercial Cloud providers support use of their services in GxP regulated environments very actively. Considering these opportunities, on-demand Cloud computing is of high relevance for Clinical Pharmacometrics as well as for other business functions within Bayer which conduct projects with very high computing demands. Over the past months we provided flexible and scalable computing resources in the Cloud that are used by Clinical Pharmacometrics for complex modeling and simulation of virtual patients in a regulated environment. Our approach to application validation and infrastructure qualification will be presented as well as concepts developed for operation of Cloud-based applications. Implications of Cloud deployment characteristics on computer system validation processes will be outlined.

5:00 Increasing Productivity through Cloud Computing  

Gurpreet Singh Kanwar, MBA, PMP, Project Manager, Information Management, NAV CANADA

With increase in usage of cloud technology, organizations are looking to adapt to cloud technology and benefits in term of productivity and profitability. Various version of clouds either private, public or hybrid are available for usage. According to IDG Enterprise’s 2016 Cloud Computing summary 56% of businesses are working on transferring more IT operations to the cloud. This presentation will help you 1) understand how Cloud computing can help you achieve business objectives; 2) determine if Cloud is right for your organization; 3) decide if you want to move application to Cloud; 4) identify multiple techniques and strategies that organizations can use and identify for successful SaaS delivery; 5) resolve multiple issues on the Cloud projects; 6) identify when SaaS is a good a solution for an organization; and 7) view a sample Agile project through this approach and identify lessons learned.

5:30 Best of Show Awards Reception in the Exhibit Hall with Poster Viewing (Commonwealth Hall)


7:0010:00 Bio-IT World After Hours @Lawn on D
 **Conference Registration Required. Please bring your conference badge, wristband, and photo ID for entry.   

Thursday, May 17

7:30 am Registration Open (Commonwealth Hall) and Morning Coffee (Foyer)

8:00 PLENARY KEYNOTE SESSION & AWARDS PROGRAM (Amphitheater & Harborview 2)

9:45 Coffee Break in the Exhibit Hall and Poster Competition Winners Announced (Commonwealth Hall)

Cityview 2

10:30 Chairperson’s Remarks
William Hayes, PhD, CTO, BioDati

10:40 Expanding the Roche Knowledge Graph with Semantic Terminology Management

Gunther Doernen, Senior Scientific Software Engineer, Research Informatics, Roche Pharmaceuticals

A pharmaceutical enterprise covers a large and diverse space of knowledge objects that are referenced by all its divisions and external partners. To accelerate knowledge integration, we introduced a new terminology management platform for Roche Pharma Research and Early Development based on semantic technologies.

11:10 BEL.bio Semantic Terminology Services

William Hayes, PhD, CTO, BioDati

BEL (http://bel.bio) is heavily dependent on unambiguously defined terminologies and easy access to terms for BEL curators and users. Much of the effort in supporting the BEL.bio open source platform is related to collecting, converting and deploying terminologies in the BEL.bio API to support term completion and search, equivalencing, and orthologization where appropriate. These terminology services are also used in semantic validation of BEL Assertions to make sure BEL Assertion arguments match the allowed semantic types in the BEL Language specification. In the near future, query support for hierarchically arranged terminologies like GO and Uberon will be added.

11:40 SELECTED POSTER PRESENTATION: From Knowledge Assembly to Hypothesis Generation in Systems and Networks Biology  
Charles Tapley Hoyt, MSc, Research Fellow, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI)

The volume, velocity, and variety of experimental and clinical data within the fields of neurodegenerative and psychiatric diseases are increasing at an unprecedented rate - with no signs of declaration. Hidden within these data are elusive patterns of cause-and-effect through both time and space that are relevant to the genesis and progression of these complex pathologies. However, as data become increasingly heterogeneous, multi-modal, and multi scale, so increases the intellectual and temporal burden of the procedures of assessment of its veracity, analysis, and interpretation. Concurrently, the knowledge in the biomedical domain necessary to execute these procedures is also increasing unmanageably. Without the assistance of information extraction, retrieval, and automated reasoning approaches, it is overwhelming, if not impossible, for individuals or groups of researchers to be knowledgeable of the state-of-the-art in any but a specific domain. Here, the Department of Bioinformatics at Fraunhofer SCAI (https://www.scai.fraunhofer.de) presents its research workflow and its suite of tools for knowledge discovery. It comprises six main components relying on natural language processing to support combine knowledge- and data-driven hypothesis generation. First, biomedical literature is preprocessed with several natural language processing techniques. Second, ontological named entities are tagged (ProMiner), indexed (SCAIView), and relations are automatically extracted (BELIEF). Third, relations are formalized in Biological Expression Language (BEL) to be made computable (with PyBEL) and manually curated to verify its veracity (BELIEF Dashboard). Fourth, available structured knowledge and annotated data are harmonized and integrated using BEL as a semantic framework (Bio2BEL) and other approaches pioneered in the IDSN project (https://www.idsn.info). Fifth, generation of hypotheses and interpretation of data-driven, machine learning-based analyses (e.g., longitudinal modeling, event-based modeling, etc.) of clinical and experimental data is supported by reasoners over BEL. Additionally, dedicated visual analytics applications support exploration of linked data, analyses, and hypotheses. Sixth and finally, experiments must be developed to test the generated hypotheses. Two current and highly relevant research projects take complementary approaches - in the Human Brain Pharmacome project (https://pharmacome.scai.fraunhofer.de), in-situ and in-vitro experiments will be used to test drug repositioning hypotheses for Alzheimer's disease (AD) and in the AETIONOMY project (https://www.aetionomy.eu), a cohort of patients with AD and Parkinson's disease will be tested to validate candidate mechanisms corresponding to the progression of the disease.
Co-author: Sumit Madan

12:10 pm Enjoy Lunch on Your Own

1:20 Dessert Refreshment Break in the Exhibit Hall with Poster Viewing (Commonwealth Hall)

Cityview 2

1:55 Chairperson’s Remarks

Alice Starr, Bioinformatics Product Manager, Genentech

2:00 Sequence. Store. Sign Out: Clinical NGS Platform for Variant Interpretation and Sign Out

Alexis Carter, MD, Physician Informaticist, Department of Pathology and Laboratory Medicine, Children’s Healthcare of Atlanta

In 2017, Children’s Healthcare of Atlanta undertook Next Generation Sequencing (NGS) as a new initiative. Using open source tools and web technologies, Children’s built a robust, scalable and secure Clinical NGS Application for use by physicians for variant annotation and interpretation. The application has many unique features to ease and speed the interpretation and annotation of variants. The Clinical NGS application presents all information needed to interpret variants in a single user-friendly web form. In the presentation, we will demonstrate and discuss many key features. For instance, how pathologists can view run-level and sample-level QC as well as sample variants in HGVS nomenclature along with tumor types, clinical context and prior patients with the same variant; how sample variants are categorized by a flexible, rules-based auto-classifier which the physician can accept or override with a few clicks; and finally, how variants can be annotated and commented, and then the pathologist can generate a report sorted by clinical significance with a single click that complies with various standards from the Association of Molecular Pathology, ASCO and CAP. The open-source storage and pipeline will be described in a separate talk by Ramesh Sringeri, Senior Applications Developer, Mobile Solutions, Children’s Healthcare of Atlanta, in Track 2 Data Computing.

2:30 Facilitating the Interpretation and Communication of in vivo Studies Results through Statistical Analysis and Visualization in R Shiny

Alice Starr, Bioinformatics Product Manager, Genentech

Our new in vivo data analysis tool is a fruit of a close collaboration between translational scientists, biostatisticians and software engineers. We are providing fit-for-purpose analyses for our scientists in the oncology and neuroscience groups, to facilitate the interpretation, comparison and communication of in vivo studies results, which helps in turn optimize the design and execution of our studies, and overall the health of our research portfolio. It has been built using the R Shiny framework, and using data from our existing pre-clinical data collection platform.

3:00 Designing Intuitive Software Applications for Life Scientists

Nikiforos Karamanis, PhD, Lead User Experience Designer, Web Development, European Bioinformatics Institute, Genome Campus

Although a software application that is intuitive can help scientists utilize data more easily and advance their research, there is little guidance on how to design such an application specifically for life scientists. We will discuss how we are designing intuitive applications for wet lab scientists at the European Bioinformatics Institute (EMBL-EBI) by engaging with our target users from the very earliest stages of design [1]. This has helped us understand the informational needs of wet lab scientists, and how these needs differ from the questions asked by domain and data specialists (such as bioinformaticians and software developers). We will outline how we are building on this understanding by including our target users in collaborative design and evaluation activities throughout software development. This ensures that our data are presented in ways that wet lab scientists find intuitive and useful. [1] Karamanis, N. et al. Designing an intuitive web application for drug discovery scientists. Drug Discovery Today https://doi.org/10.1016/j.drudis.2018.01.032

3:30 30M+ Papers to Read? Query AI with Points of Interest to Receive Distilled Hypotheses

Roman Gurinovich, Systems Architect, R&D, sci.AI

Traditional reading of all new papers published daily (~5k) can take a long time. To address this challenge, sci.AI algorithms extract facts from all the research communication worldwide every single day. Then you can ask AI: “Is there meaningful connection between GPR120 and Alzheimer’s disease?” and the platform will mimic human reasoning through the semantic network of extracted facts to return discovered pathways and support your research.

4:00 Conference Adjourns

Register Early and Save

Data Platforms and Storage Infrastructure