AFsample: Improving Multimer Prediction with AlphaFold using Aggressive Sampling
Wallner, B.
Show abstract
The AlphaFold neural network model has revolutionized structural molecular biology with unprecedented performance. We demonstrate that by stochastically perturbing the neural network by enabling dropout at inference combined with massive sampling, it is possible to improve the quality of the generated models. We generated around 6,000 models per target compared to 25 default for AF2-multimer, with v1 and v2 multimer network models, with and without templates, and increased the number of recycles within the network. The method was benchmarked in CASP15, and compared to AF2-multimer it improved the average DockQ from 0.41 to 0.55 using identical input and was ranked at the very top in the protein assembly category when compared to all other groups participating in CASP15. The simplicity of the method should facilitate the adaptation by the field, and the method should be useful for anyone interested in modelling multimeric structures, alternate conformations or flexible structures. AvailabilityAFsample is available online at http://wallnerlab.org/AFsample.
Matching journals
The top 3 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- InterPepScore: A Deep Learning Score for Improving the FlexPepDock Refinement Protocol 96%
- QDeep: distance-based protein model quality estimation by residue-level ensemble error classifications using stacked deep residual neural networks 96%
- Evaluation of AlphaFold-Multimer prediction on multi-chain protein complexes 95%
Similar papers in this journal
- High-Accuracy Protein Structures By Combining Machine-Learning With Physics-Based Refinement 96%
- Improving protein tertiary structure prediction by deep learning and distance prediction in CASP14 95%
- DisCovER: distance- and orientation-based covariational threading for weakly homologous proteins 95%
Similar papers in this journal
- Protein language model embeddings for fast, accurate, alignment-free protein structure prediction 96%
- Combining Information from Crosslinks and Monolinks in the Modelling of Protein Structures 94%
- Assessing PDB Macromolecular Crystal Structure Confidence at the Individual Amino Acid Residue Level 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.