MSA-Regularized Protein Sequence Transformer toward Predicting Genome-Wide Chemical-Protein Interactions: Application to GPCRome Deorphanization

Abstract

Small molecules play a critical role in modulating biological systems. Knowledge of chemical–protein interactions helps address fundamental and practical questions in biology and medicine. However, with the rapid emergence of newly sequenced genes, the endogenous or surrogate ligands of a vast number of proteins remain unknown. Homology modeling and machine learning are two major methods for assigning new ligands to a protein but mostly fail when sequence homology between an unannotated protein and those with known functions or structures is low. In this study, we develop a new deep learning framework to predict chemical binding to evolutionary divergent unannotated proteins, whose ligand cannot be reliably predicted by existing methods. By incorporating evolutionary information into self-supervised learning of unlabeled protein sequences, we develop a novel method, distilled sequence alignment embedding (DISAE), for the protein sequence representation. DISAE can utilize all protein sequences and their multiple sequence alignment (MSA) to capture functional relationships between proteins without the knowledge of their structure and function. Followed by the DISAE pretraining, we devise a module-based fine-tuning strategy for the supervised learning of chemical–protein interactions. In the benchmark studies, DISAE significantly improves the generalizability of machine learning models and outperforms the state-of-the-art methods by a large margin. Comprehensive ablation studies suggest that the use of MSA, sequence distillation, and triplet pretraining critically contributes to the success of DISAE. The interpretability analysis of DISAE suggests that it learns biologically meaningful information. We further use DISAE to assign ligands to human orphan G-protein coupled receptors (GPCRs) and to cluster the human GPCRome by integrating their phylogenetic and ligand relationships. The promising results of DISAE open an avenue for exploring the chemical landscape of entire sequenced genomes.

Exploration of Dark Chemical Genomics Space via Portal Learning: Applied to Targeting the Undruggable Genome and COVID-19 Anti-Infective Polypharmacology

“…When compared with the state-of-the-art method DISAE [1], which already was shown to outperform other leading methods for predicting CPIs of orphan receptors, PortalCG demonstrates superior performance in terms of both Receiver Operating Characteristic (ROC) and Precision-Recall (PR) curves, as shown in Figure 4(a). Because the ratio of positive and negative cases is imbalanced, the PR curve is more informative than the ROC curve.…”

“…In PortalCG, protein structure information is used as a portal to connect a source protein sequence and a corresponding target protein function (Figure 1A). We begin by performing self-supervised training to map tens of millions of sequences into a universal embedding space, using our recent distilled sequence alignment embedding (DISAE) algorithm [1]. Then, 3D structural information about the ligand-binding site is used to fine-tune the sequence embedding.…”

“…Experiments are first organized to test PortalCG performance against baseline models, DISAE [1] and AlphFold2 [5]. DISAE is a protein language which predicts protein function based on protein sequence information alone.…”

“…Such efforts could also provide novel approaches for identifying new druggable targets and discovering effective therapeutic strategies for currently incurable diseases; for instance, in Alzheimer's disease (AD) many disease-associated genes have been identified from multiple omics studies, but are currently considered un-druggable [14]. Accurately predicting chemical-protein interactions (CPIs) on a genome-wide scale is a challenging OOD problem [1]. If one considers only the reported area under the receiver operating characteristic curve (AUROC), which has achieved 0.9 in many state-of-theart methods [15][16], it may seem the problem has been solved.…”

Sequence-based prediction of protein binding regions and drug–target interactions

“…To model interdependency, MolTrans [ 24 ] leverages transformers built on frequent consecutive protein subsequences and SMILES notations to construct interaction maps for DTI pairs. DISAE [ 25 ] utilizes evolutionarily distilled sequence representations as inputs to ALBERT [ 26 ]. Some attention-based models train physical interactions between the substructures of ligands and binding sites of proteins to give better performance and interpretability.…”