GEMS: Molecular Structure Elucidation as Learned Search over Isomer Graphs
Start here
GEMS improves small-molecule identification from tandem mass spectra by refining candidate structures through learned 2-switch bond edits that preserve the molecular formula and atom valences. It reaches Hit@1/Hit@10 of 29.64%/50.81% on NPLIB1 and 30.22%/41.33% on MassSpecGym. On NPLIB1 it runs end to end in about 3.3 minutes versus 26.2 hours for FRIGID, and fingerprint prediction error is identified as the main bottleneck in target recovery.
Excerpted from arxiv.org
Coverage timeline
Follow the coverage to see the different angles.
- arxiv.orgPickGEMS: Molecular Structure Elucidation as Learned Search over Isomer Graphs
GEMS improves small-molecule identification from tandem mass spectra by refining candidate structures through learned 2-switch bond edits that preserve the molecular formula and atom valences. It reaches Hit@1/Hit@10 of 29.64%/50.81% on NPLIB1 and 30.22%/41.33% on MassSpecGym. On NPLIB1 it runs end to end in about 3.3 minutes versus 26.2 hours for FRIGID, and fingerprint prediction error is identified as the main bottleneck in target recovery.
Coverage and heat for this story
Reports so far 1·Heat now 9·Peak heat 10 (Oct 19 20:00)·Reports, 24h +1
Two panels, same clock (pointer stays in sync): top is cumulative editorial reports — only rises; last point = All reports; slope = new reports that hour. Bottom is Live-board heatIndex — 7-day participants with a 24-hour half-life, so it can fall when talk cools.