Want to help shape guidelines for pandemic prep, schema consensus, sequencing metadata, and categorical variants? Join four new GA4GH groups!

16 Nov 2023

Anyone interested is invited to join the Categorical Variation Study Group, the Data Model and Schema Consensus (DaMaSC) Study Group, and groups developing the Experiments Metadata Standard and Ethical Preparedness for Pandemics and Epidemics Framework.

How do you figure out if your CRAM file came from a single cell assay or bulk sequencing experiment?

Have you noticed that key evidence linking genes and disease is often attached to broad categories of variants, which are hard to find and compare?

In the next pandemic, how do we make it easier for human-pathogen genomics researchers to speed up their work while maintaining the same level of ethical conduct?

And why doesn’t every GA4GH Work Stream represent data using a shared language?

Four new groups at the Global Alliance for Genomics and Health (GA4GH) will address those questions, and more.

Sign up to join the conversation if you want to shape the way we assemble genomic data, use data to fight disease (in a single patient all the way up to a pandemic), and break down silos that make data ineffective.

The four groups are:

• Experiments Metadata Standard, drafting a checklist for capturing crucial information about genomics experiments to make finding and comparing data easier, and to reduce potential biases during data analysis;
• Ethical Preparedness for Pandemics and Epidemics (EPPE) Framework, writing a practical guide for use during global disease outbreaks to ensure research is conducted ethically, despite occurring at a faster pace;
• Categorical Variation (CatVar) Study Group, exploring how to model the categories of variation that carry key evidence about whether a certain gene variant leads to disease;
• Data Model and Schema Consensus (DaMaSC) Study Group (newly relaunched), exploring how to better align GA4GH Work Streams to avoid making biomedical data silos worse.

Learn more about each group below.

Two groups have begun building products — the EPPE Framework and the Experiments Metadata Standard — after the GA4GH Standards Steering Committee voted in September 2023 to green-light their development.

The other two are Study Groups: teams that survey the landscape of the genomics and health community, define needs, and determine whether a GA4GH product could help.

Anyone in the genomics and health community is welcome to participate in the groups. You will work alongside contributors from around the world to explore and build GA4GH products.

Experiments Metadata Standard

Researchers often obtain genomic data as CRAM or VCF documents, but these contain little information on the experiment that produced the results. For example: are the data from whole genome sequencing, transcriptomics, or another kind of epigenomic experiment?

That lack of clarity is an issue.

“It is important to get details about the experiment that produced the genomic data you are dealing with,” said David Bujold, Lead of the standard development team and Bioinformatics Manager at McGill University.

“Instruments and sequencing techniques will  affect the results. You need to understand the purpose of an experiment, like whether it targeted only certain regions of the genome, or a specific epigenetic feature, which will lead you to interpret the data in different ways. The GA4GH Experiments Metadata Standard aims to help clear up these ambiguities,” Bujold said.

The team, which operates within the GA4GH Discovery Work Stream, has begun developing a checklist of the minimum experimental information that should be captured about every high-throughput sequencing assay.

EPPE Framework

“We have seen time and again that when researchers move quickly during a global health emergency, they have sometimes completely disregarded basic ethical principles. And yet, speed is of the essence to get research results into the hands of decision-makers,” said Anja Bedeker, EPPE Lead, Co-Chair of the Public Health Alliance for Genomic Epidemiology (PHA4GE), and a Research Associate at the South African National Bioinformatics Institute. 

“The GA4GH Ethical Preparedness for Pandemics and Epidemics Framework will present a path for balancing these two crucial obligations in human-pathogen genomics research,” Bedeker said.

The EPPE Framework will be a practical tool that researchers can use when studying human-pathogen genomics during pandemics and epidemics.

To build the new framework, the GA4GH Regulatory & Ethics Work Stream is collaborating with the PHA4GE Ethics and Data Sharing Working Group.

CatVar Study Group

Say you’re reading a study on breast cancer, and you see the phrase “TP53 R248 mutations” (Berns et al. 1998). This statement doesn’t refer to a single change on a single gene. It refers to all possible variants that lead to changes at the amino acid position R248 on the protein TP53.

The phrase “TP53 R248 mutations” is an example of a categorical variant — a way for researchers to share knowledge about an entire category of genomic variation in one statement. Knowledge-bases like CIViC, ClinVar, OncoKB, and the JaxCKB use categorical variants to describe genomic evidence.

The problem? When a patient or research participant comes in with a very specific variant, you have to manually search many of these categorical variants to find the crucial evidence you need — whether or not that particular person’s variant leads to disease.

“In the GA4GH Categorical Variation Study Group, we’re going to tackle the complexity and confusion of categorical variants head-on. We hope to build a data model that helps researchers quickly find and connect categorical variants — and thus get better answers for patients and research participants,” said Daniel Puthawala, CatVar Lead and a Postdoctoral Scientist at Nationwide Children’s Hospital.

DaMaSC Study Group

For many reasons — political, ethical, commercial, circumstantial, and otherwise — biomedical data often get stuck in silos. GA4GH was founded to expand responsible use of data in order to benefit human health. The new DaMaSC Study Group will help make that mission a reality by ramping up interoperability in the way GA4GH products represent data.

“GA4GH standards power data search and analysis worldwide. So we are deeply invested in ensuring standards work together to disassemble silos, rather than reinforce them accidentally,” said Kathy Reinold, an independent consultant who leads DaMaSC. “The GA4GH Data Model and Schema Consensus Study Group is exploring how to create an easy starting point for representing genomic and related data in an interoperable way.”

The team wants to establish a resource for anyone creating new genomic data representations, or schemas. DaMaSC members will study what conventions, guidelines, and existing schemas can simplify data representation. The Study Group also plans to explore what a common vocabulary of data representation across Work Streams would look like — in service of making GA4GH products easier to use and combine.