BioCompute Conference and Workshop

2024 BioCompute Conference and Workshop

This workshop provides an opportunity to learn about interoperability between platforms that host BioCompute Objects (BCOs), the FDA’s "Portal" for reading, writing and sharing BCOs, real world applications including intended use in regulatory submissions, and potential implementations, including the use of AI/ML. This conference will have perspectives from both FDA and industry around the use of BioCompute in regulatory submissions, challenges for implementation, and tools designed to work with the standard. Join us for an in-depth discussion of BCO standards and an opportunity to view other projects during our poster session.

Event Registration: Eventbrite

Date & Time: May 10th, 2024, 9 AM - 5 PM

Location: Great Room, White Oak Campus, FDA center

Getting there

We encourage all attendees to arrive around 8:30 AM, thirty minutes prior to the start of the event at 9:00 AM, to ensure they have enough time to settle in and make the most of the day.

By Car

Enter "FDA Building 1" in Google Maps, then follow NW Loop Rd to the closest visitor parking, once reach the T-section turn right onto Michelson Rd to vehicle security screening before parking on the White Oak campus. FDA circulator shuttles are available between visitor parking and Building 1. Refer to the FDA website for the campus map.

By metro Public buses:

Ride On Bus Route 22 - stops at the Silver Spring (Red line) Metro station
BRT Flash Orange Route - stops at the Silver Spring (Red line) Metro station
Metro Bus C8 - stops at the White Flint (Red line), Glenmont (Red Line), and College Park (Green line) Metro stations
Metro Bus K6 - stops at the Fort Totten (Red, Green, and Yellow lines) Metro station
Metro Bus K9 - stops at the Fort Totten (Red, Green, and Yellow lines) Metro station

FDA shuttle to metro:

College Park Metro (Green line)
Glenmont Metro (Red line)
Medical Center Metro (Red line)
Shady Grove Metro (Red line)
Silver Spring Metro (Red line)
Twinbrook Metro (Red line)
More FDA shuttle info can be found here

Please go to the FDA site for more transportation options.

Accommodation

If you require accommodation during your stay, please refer to the following document for a list of hotels near the FDA White Oak Campus: Hotels by the FDA White Oak Campus. This resource provides information on nearby accommodations to ensure your comfort and convenience during the event.

Abstract

Documenting computational workflows has historically been a challenge for researchers and reviewers. Often, there are no clear guidelines for describing these workflows, leading to ad hoc definitions and descriptions that are insufficient for comprehension and reproducibility. BioCompute, an abbreviation for the IEEE standard 2791-2020, offers a structured framework for precisely yet adaptable describing workflows. A BioCompute Object (BCO) represents a workflow that adheres to this standard. This standard breaks down a workflow into its core components, aligning them with established data standards and ontologies. BioCompute significantly aids in streamlining workflow data, managing complexity across multiple versions, and enhancing reproducibility.

As the field evolves, addressing regulatory concerns becomes paramount, necessitating a common language for all stakeholders. The BioCompute community, comprised of over 400 individuals, has crafted a framework that offers guidance without imposing undue burdens on sponsors or reviewers. To date, BCO has been embraced by three FDA centers, effectively reducing delays in regulatory submissions. An ongoing pilot project involving the BioCompute team, FDA, and industry sponsors aims to improve communication efficiency and fluidity during submission and review processes.

The BioCompute Portal, the central platform for creating and submitting BCOs, has seen significant improvements in the past year, including Google and ORCID integration and an enhanced user interface. Furthermore, a new Portal instance at the FDA, along with four tools for authoring BCOs directly from various platforms (BCOnexus, Seven Bridges APP, the HIVE, and Galaxy platforms), and expanded documentation, make it easier to create, share, edit, and publish BCOs online. In addition, the FDA can retrieve BCOs from the public BCODB upon request, and with APIs and authentication tokens, users can directly submit to FDA portal API endpoints and interact with BCOs and their implementations seamlessly. All of the most up-to-date information about BioCompute will be introduced at the workshop and followed by a roundtable discussion to further improve the applicability of BioCompute from the FDA point of view and industry-wise.

Organization Committee

Jonathon Keeney (GWU/Chair); Raja Mazumder (GWU); Mark Walderhaug (FDA); Luis Santana-Quintero(FDA); Charles Hadley King (GWU); Omar Serang (DNAnexus); Sam Westreich (DNAnexus); Dennis A. Dean, II (Velsera); Phil Webster (Velsera); Kostis Karagiannis (Moderna)

Workshop Topics

Updates and future for BioCompute
Need for BCO at FDA
Considerations for Adoption in Industry
Progress Towards Usability
BCODB and Portal's use at the FDA
BCO platform interoperability
Exemplar workflows and real-world applications
BCO-Nextflow
Potential AI/ML-related implementations
FAQ session & Roundtable discussion

Poster Session

Participants are encouraged to submit a poster abstract by April 15th, 11:59 PM (EDT). Posters should describe a computational pipeline or topic related to BioCompute. Up to 5 posters will have their workflows converted into a BCO by the BCO team and published in the BCO Portal, and will be offered a 5-minute lightning talk at the end of the conference. All posters must be printed before the workshop and set up at the start of the event.

Abstract: Any computational workflow of interest, or topic related to BCOs

Poster Content: Posters should highlight or describe a workflow (either in progress or one already in use) and the way it is used.

Size: poster sizes can be up to 4' tall x 6' wide.

Submission Portal: click here.

Conference Webinar

BioCompute Conference 2024 at FDA

Presented by the BioCompute team with plenary address by CBER Director Dr. Peter Marks

Published May 10, 2024

Watch Here

00:00 Intro - Jonathon Keeney, PhD

11:45 Vision and Future of Biocompute - Raja Mazumder, PhD, MS, MSC

28:48 FDA Alignment With The Vision of Biocompute - Peter W. Marks, MD, PhD

43:23 Biocompute Vision From an FDA-Industry Perspective - Vahan Simonyan, MD, PhD

1:00:26 A computational framework for end-to-end clinical omics analysis - Jun Luo, PhD

1:10:47 Applications of Biocompute Objects For Viral Metagenomics Analysis - Julia Maritz, PhD

1:21:23 Question Panel 1

1:42:14 Biocompute Objects and Workflows - Mark Walderhaug, PhD

1:59:40 Computational Reviews at CBER - Luis V Santanz-Quintero, PhD

2:12:23 High Throughput Sequencing Data Challenges in FDA Regulatory Review - Eric Donaldson, PhD

2:30:26 Question Panel 2:

2:43:43 AI and Data Science For Health and Research: Exploring NIH Strategies, Challenges, and Visions for the Future - Susan Gregurick, PhD

3:02:32 Biocompute Object Submissions to FDA - Virginia Hussong

3:15:17 Extending BCO Specification into a Generic Metamodel For Computable Data - Guangzhi Sun, PhD

3:37:17 Functioning Biocompute Objects As AI Agents - Jonas S Almeida, PhD

3:52:20 DNAnexus: Democratizing access to BCOs with the BCOnexus Platform Free Editor - Sam Westreich, PhD

4:06:09 Facilitating BCO Use From Standard to Best Practices: The Neoantigen Use Case - Dennis A. Dean, PhD

4:28:52 BioCompute Portal, BioCompute Database, and User Database: Why So Many? - Charles Hadley King, MS

4:50:00 BCO Provenance Models and FAIR Computational Workflows - Stain Soiland-Reyes, MSc & Meznah Aloqalaa, MS

5:10:51 Generating BCO’s For Nextflow Pipelines - Ben Sherman, PhD

05:36:32 Large Language Models - Konstantinos Krampis, PhD

05:51:10 Retrieval-Augmented Generation For BioCompute Object Creation - Sean Kim, MS

06:07:28 BioCompute Library Project - Sean Keeney

06:11:54 Visualization For Published BCOs - Tianyi Wang, MS

06:16:16 BCOs to Unite Data and Workflow into Analysis - Patrick McNeely, MSc, PhD

06:19:20 Closing Remarks - Raja Mazumder, PhD, MS, MSc & Mark Walderhaug, PhD

Schedule

Morning Session	Time	Speaker	Content
	9:00 - 10:00AM	Introduction and Future for BioCompute
	9:00 - 9:15	Jonathon Keeney Research Assistant Professor, George Washington University	Welcome and Introduction
	9:15 - 9:30	Raja Mazumder Professor, George Washington University	Vision and Future of BioCompute
	9:30 - 9:45	Peter Marks Director, CBER, FDA	Plenary Address
	9:45 - 10:00AM	Break
	10:00 - 10:55AM	Considerations for Adoption in Industry
	10:00 - 10:15	Vahan Simonyan CSO, Embleema, DNA HIVE	BioCompute Vision from an FDA-Industry Perspective
	10:15 - 10:30	Jun Luo Data Scientist, Novo Nordisk	A Computational Framework for end-to-end Clinical Omics Analysis: Single Cell Mechanism of action study in chronic diseases
	10:30 - 10:45	Julia Maritz Senior Scientist, Merck	Applications of BioCompute Objects for Viral Metagenomics Analysis
	10:45 - 10:55	Discussion
	10:55 - 11:10AM	Break and poster review
	11:10 - 12:05PM	Need for BCO at FDA
	11:10 - 11:25	Mark Walderhaug Associate Director for Risk Assessment, CBER, FDA	BioCompute Objects and Workflows
	11:25 - 11:40	Luis Santana-Quintero HIVE lead, CBER, FDA	Computational Reviews at CBER
	11:40 - 11:55	Eric Donaldson Clinical Virology Reviewer	Perspectives from a Reviewer: High Throughput Sequencing Data Challenges in FDA Regulatory Review
	11:55 - 12:05	Discussion
	12:05 - 1:10PM	Lunch & Poster Session
Afternoon Session	1:10 - 4:30PM	Technical, Administrative, and Usability Session
	1:10 - 1:25	Susan Gregurick Director of Data Science, NIH	AI for Health and Research: Exploring NIH’s Strategies, Challenges, and Vision for the Future
	1:25 - 1:40	Virginia Hussong Miscellaneous Administration and Program, FDA	Forthcoming Guidance for BCO Submission
	1:40 - 1:55	Guangzhi Sun Associate Researcher, China National Institute of Standardization (CNIS), ISO	Extending BCO specification into a Generic Metamodel for Computable Data
	1:55 - 2:10PM	Break and Poster review
	2:10 - 2:25	Jonas Almeida Director of Data Science, Division of Cancer Epidemiology and Genetics, NCI	Functionalization of Biocompute Objects using Generative AI
	2:25 - 2:40	Sam Westreich Sr. Scientific Program Manager, DNAnexus	DNAnexus: Democratizing access to BCOs with the BCOnexus Platform-Free Editor
	2:40 - 3:00	Dennis Dean Director of Translational Sciences, Velsera	Facilitating BCO Use from Standard to Best Practices: The Neoantigen Workflow Use Case
	3:00 - 3:15	Hadley King Research Scientist, HIVE Lab, George Washington University	BioCompute Portal, BioCompute Database, and User Database: Why so many?
	3:15 - 3:30PM	Break and Poster review
	3:30 - 3:45	Stian Soiland-Reyes and Meznah Aloqalaa Senior Lecturer, University of Manchester/Ph.D. Student	BCO, Provenance Models (PRIMAD, RO-Crate) and FAIR Computational Workflows
	3:45 - 4:00	Ben Sherman Developer, Seqera Labs	Generating BioCompute Objects for Nextflow Pipelines
	4:00 - 4:15	Konstantinos Krampis Associate Professor of Biology and Computer Science, CUNY	BCO in AI: Leveraging Large Language Models (aka ChatGPT) for Biocompute Object Generation in Bioinformatics
	4:15 - 4:30	Sean Kim Bioinformatics Specialist, George Washington University	Enhanced BioCompute Object Creation through Retrieval-Augmented Generation from Scientific Papers
	4:30 - 4:40PM	Break and Poster review
	4:40 - 5:00PM	Lightning talk and closing
	4:40 - 4:55	Chosen speakers	Poster lightning talks
	4:55 - 5:00PM	Closing Mark Walderhaug and Raja Mazumder

Confirmed Presenters


Peter Marks	Susan K. Gregurick	Raja Mazumder	Jonas Almeida	Jun Luo
Vahan Simonyan	Sam Westreich	Konstantinos Krampis	Eric Donaldson	Dennis A. Dean, II
Virginia Hussong	Jonathon Keeney	Luis Santana-Quintero	Ben Sherman	Sean Kim
Guangzhi Sun	Hadley King	Stian Soiland-Reyes	Mark Walderhaug	Julia Maritz