Open gene-finding model Carbon-A released with database of 566 million candidate genes

Carbon-A’s creators announced an open model that locates genes in DNA, alongside a database of 566 million candidate genes across 22,617 species, from fungi to mammals. They say the model does not need a close relative to read a new genome. Working with Active Site and UCSD, they also found RNA evidence for 239 genes missing from reference annotations for cats, chickens, hamsters and a lab plant. Carbon-A does not design DNA or predict what genes do.