Protein Interaction Network Reconstruction with a Structural Gated Attention Deep Model by Incorporating Network Structure Information

Protein-protein interactions (PPIs) provide a physical basis of molecular communications for a wide range of biological processes in living cells. Establishing the PPI network has become a fundamental but essential task for a better understanding of biological events and disease pathogenesis. Although many machine learning algorithms have been employed to predict PPIs, with only protein sequence information as the training features, these models suffer from low robustness and prediction accuracy. In this study, a new deep-learning-based framework named the Structural Gated Attention Deep (SGAD) model was proposed to improve the performance of PPI network reconstruction (PINR). The improved predictive performances were achieved by augmenting multiple protein sequence descriptors, the topological features and information flow of the PPI network, which were further implemented with a gating mechanism to improve its robustness to noise. On 11 independent test data sets and one combined data set, SGAD yielded area under the curve values of approximately 0.83-0.93, outperforming other models. Furthermore, the SGAD ensemble can learn more characteristics information on protein pairs through a two-layer neural network, serving as a powerful tool in the exploration of PPI biological space.

Automated summary

Protein-protein interactions (PPIs) play a crucial role in biological processes, and establishing the PPI network is important for understanding biological events and disease pathogenesis. Existing machine learning models for predicting PPIs suffer from low robustness and accuracy. In this study, a new deep-learning-based framework called the SGAD model was proposed, which improves the performance of PPI network reconstruction by incorporating protein sequence descriptors, topological features, and information flow. The SGAD model achieved better predictive performance compared to other models, and its ensemble can learn more characteristics information on protein pairs for exploring the biological space of PPIs.