Canadian Platform for Genomics and Precision Health

How We Are Solving It

The technology developed throughout this project will enable privacy-preserving “federated insights”, where insights can be derived across distributed datasets without moving them. All data remains in place and so, instead of “moving data to the question” this new approach “moves questions to the data”. This is more efficient, secure, scalable, compliant and empowers custodians to retain control over data. The CP4GPH will be used to train AI-powered models across federated networks of data in neuroscience, oncology, rare disease, infectious disease, health care, agriculture and beyond.

The goal of the project is to democratize access to leading-edge technologies for customer groups across the genomics and precision health ecosystem to be able to create, grow and take advantage of federated learning networks. To help achieve these aims, the project will create a menu of new software and related services that can be licensed to data generators (such as sequencing companies, hospitals, precision health initiatives, patient advocacy groups and funders) who want to increase the impact of their data, and to data consumers (such as health ministries and pharmaceutical companies) who want to extract insights from shared data.

The project will cultivate an open innovation ecosystem that additional stakeholders can plug into, and will ultimately commercialize globally-leading solutions to manage, share, analyze, and translate insights from genomics and health data. This will put Canada at the forefront of precision health: driving innovation, creating jobs, and fostering high-caliber research and discovery.

*Funding redirected by Autism Speaks, PacBio and Roche to be used to fund DNAstack, McGill, Holland Bloorview, University of New Brunswick (UNB) and SickKids. DNAstack will be overseeing the management and distribution of this fund on behalf of Autism Speaks, PacBio and Roche.

The Result

The primary goal of the CP4GPH project was to democratize access to leading-edge technologies for customer groups across the genomics and precision health ecosystem to be able to create, grow and take advantage of federated learning networks. Fulfilling this goal manifested under multiple commercial releases of new products, services and partnerships by the consortium partners over the project’s course:

Omics AI Launch

In September 2023, DNAstack publicly launched their Omics AI software suite. Together, the software products included within Omics AI can be used to create standards compliant federated networks and allow for federated analyses to be run across connected data. The end-to-end software suite is compliant with open standards developed by the Global Alliance for Genomics and Health and can be deployed across multiple cloud platforms as well as on premises.

WGS Variant Pipeline and Workflows Marketplace Launch

In October 2023, PacBio announced the release of the WGS Variant Pipeline – a complete, standardized computational method for HiFi whole genome sequencing (WGS) data analysis, making it the most complete human WGS secondary analysis pipeline available today. In parallel, DNAstack launched its Workflows Marketplace to unify omics analysis and make it easier and more secure for scientists to discover and run trusted, high-quality, reproducible computational pipelines for analyzing omics data. The marketplace includes a curated collection of best practices pipelines for processing data from popular sequencing instrument manufacturers, including PacBio.

Partnerships

Due to the progress achieved by Integrate.ai, they were selected as the federated learning technology partner to power one of the world’s largest precision health consortia. Their partnership with the Terry Fox Research Institute to deliver the Digital Health and Discovery Platform (DHDP) will advance Canadian healthcare through cutting-edge AI technology and a world-leading data access and governance framework.

MSSNG DB7 Release

As a result of all the work performed by consortium partners —  DNAstack, Autism Speaks and SickKids Hospital expanded Autism Speaks’ MSSNG database, the world’s largest autism whole-genome dataset, with the release of DB7; adding 2,300 whole genome sequences along with associated phenotypic data, bringing the total number of samples in MSSNG to 13,500. In addition to the MSSNG samples, 700 samples from Qatar will be included in the release, which is critical for personalized healthcare for underrepresented communities. This advancement will help further the goal of providing researchers with the best resources to enable the identification of autism subtypes, which will pave the way for better diagnostics, as well as personalized and more accurate treatments.