Particle.news
Download on the App Store

Illumina Launches Billion Cell Atlas to Train AI in Drug Discovery

The release marks BioInsight’s first commercial offering, signaling Illumina’s push to sell proprietary datasets beyond sequencing instruments.

Overview

  • Unveiled Jan. 13 at the J.P. Morgan Healthcare Conference, the dataset maps cellular responses to CRISPR perturbations across more than 200 disease-relevant cell lines covering roughly 20,000 genes.
  • AstraZeneca, Merck/MSD and Eli Lilly are initial collaborators, with access beginning as Illumina continues to expand the resource.
  • Illumina reports about 150 million single-cell profiles generated so far and projects reaching one billion cells by year-end.
  • The Atlas is designed to yield about 20 petabytes of single-cell transcriptomic data within a year using the company’s Single Cell 3’ RNA prep platform.
  • Illumina says the resource is intended to train large AI models, characterize disease and drug mechanisms, validate genetic targets and explore potential new indications.