A residue-level preference optimization method that uses AlphaFold pLDDT scores as rewards triples the in silico design success rate of LigandMPNN on enzyme benchmarks.
Learning from Protein Structure with Geometric Vector Perceptrons
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Improving Protein Sequence Design through Designability Preference Optimization
A residue-level preference optimization method that uses AlphaFold pLDDT scores as rewards triples the in silico design success rate of LigandMPNN on enzyme benchmarks.