Yhn ScienceBiology first seen 1 d ago, last 13 h ago, peak #9
DNA annotation model uncovers 566 million new gene candidates
Original: Carbon-A: DNA annotation model used to discover 566M new gene candidates
Carbon-A, a DNA annotation model from Hugging Face's biology team, has been used to identify 566 million previously uncharacterized gene candidates across genomic data. The model is designed to speed up genome annotation, including on GenBank-scale datasets, where manual or traditional computational methods struggle to keep pace with the volume of sequenced DNA. The release highlights how machine learning tools are accelerating biological discovery.
Why now: The scale of newly discovered gene candidates suggests machine learning is opening up large untapped areas of genomic data.
Evidence
API: https://socialmediatrends-api.osmike.com/v1/trends/1664375