Framingham and TCGA were pivotal without AI; models in the loop could boost data generation

anshulkundaje · x · 2026-09-21

A discussion among genomics researchers. viriditax argues that pivotal data generation efforts like the Framingham Heart Study and TCGA were launched without a specific mechanistic model in mind, functioning more as "north stars"—end goals that forced the building of a useful technology stack, and which couldn't realistically be redone today. anhulkundaje (Stanford epigenomics professor) agrees, adding that these projects predate generative AI, and small tweaks with models in the loop across phases of data generation would have substantially increased bang for buck.

Related event: Scholars debate how science should generate data in the AI era(3 posts)→

Original post →

More from Research

Research channel →