HybridGCN for protein solubility prediction with adaptive weighting of multiple features

Abstract

The solubility of proteins stands as a pivotal factor in the realm of pharmaceutical research and production. Addressing the imperative to enhance production efficiency and curtail experimental costs, the demand arises for computational models adept at accurately predicting solubility based on provided datasets. Prior investigations have leveraged deep learning models and feature engineering techniques to distill features from raw protein sequences for solubility prediction. However, these methodologies have not thoroughly delved into the interdependencies among features or their respective magnitudes of significance. This study introduces HybridGCN, a pioneering Hybrid Graph Convolutional Network that elevates solubility prediction accuracy through the combination of diverse features, encompassing sophisticated deep-learning features and classical biophysical features. An exploration into the intricate interplay between deep-learning features and biophysical features revealed that specific biophysical attributes, notably evolutionary features, complement features extracted by advanced deep-learning models. Augmenting the model’s capability for feature representation, we employed ESM, a substantial protein language model, to derive a zero-shot learning feature capturing comprehensive and pertinent information concerning protein functions and structures. Furthermore, we proposed a novel feature fusion module termed Adaptive Feature Re-weighting (AFR) to integrate multiple features, thereby enabling the fine-tuning of feature importance. Ablation experiments and comparative analyses attest to the efficacy of the HybridGCN approach, culminating in state-of-the-art performances on the public eSOL and S. cerevisiae datasets.

Cerebra: a computationally efficient framework for accurate protein structure prediction

“…Many computational methods adopt the graph representation to describe a protein, where the nodes correspond to amino acid residues and the edges quantify the inter-residue interactions. The graph representation of proteins achieves successes in many practical applications, including the protein structure prediction 13,14 , the protein functional analysis 23–26 , the protein-ligand interaction 27–29 , etc ., due to the adequate consideration of the pairwise residue-residue interaction in addition to the 1D sequence information. Historically, theories or models purely based on pairwise descriptions have confronted great challenges, since they are unable to provide a perfect description for the complex physical, chemical or biological systems, particularly when the three-body and higher-order many-body effects are salient.…”

Cerebra: A Computationally Efficient Framework for Accurate Protein Structure Prediction

“…Many computational methods adopt the graph representation to describe a protein, where the nodes correspond to amino acid residues and the edges quantify the inter-residue interactions. The graph representation of proteins achieves successes in many practical applications, including the protein structure prediction 13,14 , the protein functional analysis [25][26][27][28] , the protein-ligand interaction [29][30][31] , etc., due to the adequate consideration of the pairwise residue-residue interaction in addition to the 1D sequence information. Historically, theories or models purely based on pairwise descriptions have confronted great challenges, since they are unable to provide a perfect description for the complex physical, chemical or biological systems, particularly when the three-body and higher-order many-body effects are salient.…”