MoLPC2

MoLPC2 predicts the structures of large protein complexes and their stoichiometry using Monte Carlo Tree Search and AlphaFold2.

Key Features:

• Stoichiometry-free prediction: MoLPC2 assembles protein complexes while simultaneously predicting their stoichiometry without requiring prior knowledge of stoichiometry.
• Alternative AlphaFold2 prediction sampling: MoLPC2 improves predictions by allowing sampling of alternative AlphaFold predictions.
• Large complex assembly: MoLPC2 accurately predicted the structures of 50 out of 175 nonredundant protein complexes with TM-score ≥ 0.8 without knowing the stoichiometry.

Scientific Applications:

• Large protein complex structure prediction: MoLPC2 predicts structures of large protein complexes solely from sequence information.
• Stoichiometry determination: MoLPC2 determines the stoichiometry of protein complexes during the assembly process.

Methodology:

MoLPC2 uses Monte Carlo Tree Search algorithms enhanced from MoLPC to enable assembly of protein complexes while simultaneously predicting stoichiometry, with improved prediction through sampling alternative AlphaFold predictions.