GraphCPLMQA: Assessing protein model quality based on deep graph coupled networks using protein language model

Model quality evaluation is crucial part of protein structural biology. How to distinguish high-quality models from low-quality models, and to assess which high-quality models have relatively incorrect regions for improvement, are remain challenge. More importantly, the quality assessment of multimer models is a hot topic for structure predicton.In this work, we present GraphCPLMQA, a novel graph-coupled network that uses embeddings from protein language models to assess residue-level protein model quality. The GraphCPLMQA consists of a graph encoding module and a transform-based convolutional decoding module. In encoding module, the underlying relational representations of sequence and high-dimensional geometry structure are extracted by protein language models with Evolutionary Scale Modeling. In decoding module, the mapping connection between structure and quality are inferred by the representations and low-dimensional features. Specifically, the triangular location and residue level contact order features are designed to enhance the association between the local structure and the overall topology. Experimental results demonstrate that GraphCPLMQA using single-sequence embedding achieves the best performance compared to the CASP15 interface evaluation method in 9108 models of CASP15 multimer test set. In CAMEO blind test (2022-05-20[~]2022-08-13), GraphCPLMQA ranked first compared to other servers. GraphCPLMQA also outperforms state-of-the-art methods on 19,035 models in CASP13 and CASP14 monomer test set. Finally, on AlphaFold2 datasets, GraphCPLMQA was superior to self-assessment of AlphaFold2 in MAE metric, and it was able to screen out better models than AlphaFold2.