Ruth Timme on pathogen data standards at GMI13
On day three of the Global Microbial Identifier conference in Vancouver, Canada, Lee Katz and Andrew Page speak with Ruth Timme of the U.S. Food and Drug Administration. They discuss the Pathogen Data Object Model, differences between foodborne pathogen and virus genome submissions, and the difficulty of finding inconsistently structured metadata. The conversation also considers how GMI could help governments implement standards developed by technical groups.
Listen to Episode 113: Global Microbial Identifier conference with Ruth Timme on SoundCloud
In this episode
Coordinating foodborne pathogen data
Ruth runs the FDA’s genomic epidemiology programme for foodborne pathogens. The FDA funds a couple of dozen laboratories to generate sequence data, and much of her work involves ensuring that their submissions follow a standard format and arranging updates when they do not. She also works with CDC and USDA, and with PHA4GE on international adoption of standards.
Salmonella and Listeria provide examples of the programme’s public-data contribution. Andrew asks about hundreds of thousands of Salmonella submissions, but Ruth cautions that this is probably an overestimate. The conversation establishes that the submission volume is substantial without giving a confirmed total.
A consistent structure for pathogen genomes
The Pathogen Data Object Model, or DOM, proposes that whole-genome sequence data for any pathogen should be submitted and stored in the same structured format within INSDC. Ruth explains that there are currently many ways to submit a pathogen genome and many places to put its metadata, making both submission and retrieval confusing.
The aim is a consistent standard that third-party applications can use without having to accommodate different arrangements for Salmonella and viruses. Ruth says a standard had already been established for foodborne pathogens a decade earlier, with applications built around it. Her experience with COVID genome submissions showed her that those conventions were not universal.
Virus records and hard-to-find submissions
Ruth describes metadata being placed on flat files and pipelines being built to add or retrieve it. She estimates that roughly half of virus records queried in INSDC lack a BioSample; this is an informal estimate in the conversation, not a presented analysis.
Andrew describes another difficulty: samples and data can pass through several laboratories in different countries before sequencing and submission. After being challenged about apparently missing data, he checked and found that raw data, metadata, BioSample records and consensus genomes had been submitted. They still were not appearing as expected. The example illustrates how difficult it can be to trace a submission even when its components are present.
Technical standards and government uptake
Ruth suggests that GMI could provide a place to discuss standards developed elsewhere and work out how to implement them. She points to its intergovernmental presence and collaborations with WHO, FAO and ministries of health, while stressing that her thoughts about GMI’s future are still emerging.
Lee contrasts PHA4GE’s hands-on technical work with GMI’s government contacts and potential role in securing uptake. Andrew highlights PHA4GE’s papers, functional code and specifications, alongside GMI’s ability to bring people together to discuss them. This is a proposed distinction between complementary roles, not an announcement of a formal division of responsibilities.
Highlights
- 00:01:00 — Ruth Timme introduces her role in the FDA’s foodborne pathogen genomics programme.
- 00:01:36 — Standardising submissions from a couple of dozen FDA-funded laboratories.
- 00:02:21 — The Pathogen Data Object Model and a shared structure for INSDC submissions.
- 00:03:00 — How a common standard could simplify third-party applications.
- 00:03:33 — Established foodborne pathogen conventions meet different COVID genome submission practices.
- 00:04:17 — Ruth estimates that around half of virus records in INSDC lack a BioSample.
- 00:04:25 — Multi-country laboratory collaborations make submitted data difficult to trace.
- 00:05:49 — Ruth considers GMI’s potential role in implementing standards through government partnerships.
- 00:07:16 — PHA4GE’s technical outputs and GMI’s role as a forum for discussion.
In their own words
GMI is a good place to discuss the standards that are being set in other organizations.
— Ruth Timme, 00:05:49
Who is talking
- Lee Katz (host)
- Andrew Page (host)
- Ruth Timme (guest, U.S. Food and Drug Administration)
Tools and resources mentioned
INSDC, BioSample, GenBank, Pathogen Data Object Model (DOM).
Questions this episode answers
What is the Pathogen Data Object Model?
Ruth describes it as a common structured format for submitting and storing whole-genome sequence data for any pathogen within INSDC. Its purpose is to make submissions and queries more consistent, including for third-party applications.
What does Ruth Timme do at the FDA?
She runs the FDA’s genomic epidemiology programme for foodborne pathogens. Her work includes coordinating standardised submissions from a couple of dozen funded laboratories and collaborating with other U.S. agencies and international partners.
What problems does the episode identify with virus genome submissions?
Ruth describes inconsistent metadata placement and virus records without BioSample records. Andrew also recounts a case where submitted raw data, metadata, BioSample records and consensus genomes were not appearing as expected.
How might GMI and PHA4GE complement each other?
The speakers associate PHA4GE with technical standards, code and specifications, while suggesting that GMI could help governments discuss and implement that work. They present this as an emerging view rather than an agreed organisational arrangement.
This page was written from a machine transcript of the episode and checked against it. Listen to the episode for the whole conversation.
Show notes
Lee and Andrew are at the Global Microbial Identifier conference (GMI13) in Vancouver Canada. On day 3 we catch up with Dr Ruth Timme.