Carbon-A’s 566 Million Gene Map Outsizes RefSeq Database by 16x

Researchers released Carbon-A, an AI tool and database containing 566.34 million candidate genes across 22,617 species, which is approximately 16x larger than the RefSeq dataset. The project validated 239 genes previously missing from RefSeq through wet lab experiments with Active Site and UCSD. Carbon-A identifies gene coordinates rather than predicting function or designing new DNA to address safety concerns.