A gated graph transformer for protein complex structure quality assessment and its performance in CASP15

Abstract

Motivation: Proteins interact to form complexes to carry out essential biological functions. Computational methods such as AlphaFold-multimer have been developed to predict the quaternary structures of protein complexes. An important yet largely unsolved challenge in protein complex structure prediction is to accurately estimate the quality of predicted protein complex structures without any knowledge of the corresponding native structures. Such estimations can then be used to select high-quality predicted complex structures to facilitate biomedical research such as protein function analysis and drug discovery.

Results: In this work, we introduce a new gated neighborhood-modulating graph transformer to predict the quality of 3D protein complex structures. It incorporates node and edge gates within a graph transformer framework to control information flow during graph message passing. We trained, evaluated and tested the method (called DProQA) on newly-curated protein complex datasets before the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) and then blindly tested it in the 2022 CASP15 experiment. The method was ranked 3rd among the single-model quality assessment methods in CASP15 in terms of the ranking loss of TM-score on 36 complex targets. The rigorous internal and external experiments demonstrate that DProQA is effective in ranking protein complex structures.

Availability and implementation: The source code, data, and pre-trained models are available at https://github.com/jianlin-cheng/DProQA.

Figure 1.

An overview of the DProQA pipeline for quality assessment of protein complexes. The input is a protein complex structure. The output includes the predicted quality score of the input structure (e.g. DockQ score) and the probability of the quality class (i.e. incorrect topology, acceptable quality, medium quality, and high quality) which the structure is classified into.

Figure 2.

The gated graph transformer (GGT) model architecture for updating the features of nodes of a protein complex graph.

Figure 3.

The average TM-score ranking loss for all single-model methods. MULTICOM_egnn ranked 3rd among all single-model methods.

Figure 4.

MULTICOM_egnn’s loss histogram for CASP15 targets. The black dashed vertical line represents the position of the mean value.

Figure 5.

Target H1111. (a) TM-score distribution of CASP15 models of H1111. (b). From left to right, the three protein complex structures shown are the corresponding native structure, the true TOP-1 model, and the MULTICOM_egnn top selected model, respectively. Here, MULTICOM_egnn achieved a 0.0014 TM-score ranking loss.