/AI6h ago

ESMFold2 Struggles on De Novo Protein Sequences Without Homology

--0--
Comments
Andre Watson 馃К
Reposts
Anshul Kundaje#1675
Original postAnshul Kundaje#1675
Andre Watson 馃К@nanogenomic

@proteinrosh Hey Roshan, any thoughts on ESMFold2 with respect to fully de novo sequences that have no evolutionary priors or homology? Finding that ESMFold2 works very well on homologues/conserved seqs (both with and without MSA), but fails to find poses for de novo sequences reliably.

8:28 AM 路 Jun 3, 2026 路 1.1K Views
Sentiment
Sentiment building, check back later.
Cluster Engagement
-
Views
-
Comments
-
Reposts
-
Bookmarks
Expand data
Posts from X
Most Activity
Most ActivityTimeline
No ranked X posts are available for this story yet.