Protein complex prediction with AlphaFold-Multimer

Abstract

While the vast majority of well-structured single protein chains can now be predicted to high accuracy due to the recent AlphaFold [1] model, the prediction of multi-chain protein complexes remains a challenge in many cases. In this work, we demonstrate that an AlphaFold model trained specifically for multimeric inputs of known stoichiometry, which we call AlphaFold-Multimer, significantly increases accuracy of predicted multimeric interfaces over input-adapted single-chain AlphaFold while maintaining high intra-chain accuracy. On a benchmark dataset of 17 heterodimer proteins without templates (introduced in [2]) we achieve at least medium accuracy (DockQ [3]≥0.49) on 14 targets and high accuracy (DockQ≥0.8) on 6 targets, compared to 9 targets of at least medium accuracy and 4 of high accuracy for the previous state of the art system (an AlphaFold-based system from [2]). We also predict structures for a large dataset of 4,433 recent protein complexes, from which we score all non-redundant interfaces with low template identity. For heteromeric interfaces we successfully predict the interface (DockQ≥0.23) in 67% of cases, and produce high accuracy predictions (DockQ≥0.8) in 23% of cases, an improvement of +25 and +11 percentage points over the flexible linker modification of AlphaFold [4] respectively. For homomeric interfaces we successfully predict the interface in 69% of cases, and produce high accuracy predictions in 34% of cases, an improvement of +5 percentage points in both instances.

Assessing and enhancing foldability in designed proteins

“…Recently, methods that use deep learning have been very successful in ab initio structure prediction [36][37][38][39]. The most successful of these, AlphaFold2, has shown atomic-accuracy prediction in community blind-prediction experiments and other structure-prediction challenges [40].…”

Identification of Iron-Sulfur (Fe-S) and Zn-binding Sites Within Proteomes Predicted by DeepMind’s AlphaFold2 Program Dramatically Expands the Metalloproteome

“…As this manuscript is being completed rapid advances are being made in adapting and extending AlphaFold2 to allow it to model homomeric and heteromeric protein complexes (e.g. [35][36][37]). It seems likely, therefore, that in the near future the same strategy reported here might be used to identify ligand binding sites that lie at the interface between separate polypeptide chains.…”

Protein Design with Deep Learning

“…Following the success of DL in the CASP13 competition (Critical Assessment of protein Structure Prediction) [7,22], many of the top methods integrated with success some DL components for CASP14 [23], including trRosetta [24], D-I-TASSER, and D-Quark [25]. Currently, top methods [8,26] can sometimes reach atomic accuracy on single chains, and can even deal with multimers [27]. For more information on DL for protein structure prediction, the reader is referred to the recent reviews in [28,29].…”