Accurately predicting the binding affinity between proteins and ligands is crucial in drug screening and optimization, but it is still a challenge in computer-aided drug design. The recent success of AlphaFold2 in predicting protein structures has brought new hope for deep learning (DL) models to accurately predict protein–ligand binding affinity. However, the current DL models still face limitations due to the low-quality database, inaccurate input representation and inappropriate model architecture. In this work, we review the computational methods, specifically DL-based models, used to predict protein–ligand binding affinity. We start with a brief introduction to protein–ligand binding affinity and the traditional computational methods used to calculate them. We then introduce the basic principles of DL models for predicting protein–ligand binding affinity. Next, we review the commonly used databases, input representations and DL models in this field. Finally, we discuss the potential challenges and future work in accurately predicting protein–ligand binding affinity via DL models.

Evaluating the protein–ligand binding affinity is a substantial part of the computer-aided drug discovery process. Most of the proposed computational methods predict protein–ligand binding affinity using either limited full-length protein 3D structures or simple full-length protein sequences as the input features. Thus, protein–ligand binding affinity prediction remains a fundamental challenge in drug discovery. In this study, we proposed a novel deep learning-based approach, DLSSAffinity, to accurately predict the protein–ligand binding affinity. Unlike the existing methods, DLSSAffinity uses the pocket–ligand structural pairs as the local information to predict short-range direct interactions. Besides, DLSSAffinity also uses the full-length protein sequence and ligand SMILES as the global information to predict long-range indirect interactions. We tested DLSSAffinity on the PDBbind benchmark. The results showed that DLSSAffinity achieves Pearson's R = 0.79, RMSE = 1.40, and SD = 1.35 on the test set. Comparing DLSSAffinity with the existing state-of-the-art deep learning-based binding affinity prediction methods, the DLSSAffinity model outperforms other models. These results demonstrate that combining global sequence and local structure information as the input features of a deep learning model can improve the accuracy of protein–ligand binding affinity prediction.

The diversity and complexity of RNA include sequence, secondary structure, and tertiary structure characteristics. These elements are crucial for RNA’s specific recognition of other molecules. With advancements in biotechnology, RNA–ligand structures allow researchers to utilize experimental data to uncover the mechanisms of complex interactions. However, determining the structures of these complexes experimentally can be technically challenging and often results in low-resolution data. Many machine learning computational approaches have recently emerged to learn multiscale-level RNA features to predict the interactions. Predicting interactions remains an unexplored area. Therefore, studying RNA–ligand interactions is essential for understanding biological processes. In this review, we analyze the interaction characteristics of RNA–ligand complexes by examining RNA’s sequence, secondary structure, and tertiary structure. Our goal is to clarify how RNA specifically recognizes ligands. Additionally, we systematically discuss advancements in computational methods for predicting interactions and to guide future research directions. We aim to inspire the creation of more reliable RNA–ligand interaction prediction tools.

The role of the HIV-1 Tat protein in viral transcription is linked to variations among subtypes.

Background: Intrinsically disordered regions (IDRs) within proteins often act as pivotal linkage units for the interaction of functional domains. The p53 tumor suppressor protein contains intrinsically disordered N-terminal and C-terminal domains (NTD and CTD), playing crucial regulatory roles in cellular processes. Furthermore, experimental approaches have encountered challenges in elucidating the structural regulation by the IDRs. Methods: In this work, we employed microsecond-scale molecular dynamics simulations to explore the allosteric regulation mechanism of the p53 DNA binding domain (DBD) induced by the CTD and the DNA binding. Subsequently, we integrated dynamic cross-correlation analysis with binding free energy calculations to evaluate the interaction between the CTD and DNA. Results: The free energy landscapes (FELs) were utilized to identify the conformational ensemble of the p53 DBD. The FELs revealed that the CTD enhances the allosteric regulatory mechanisms. Conclusions: Firstly, the conformation of DBD changes on the S6-S7 loop and L1 upon DNA binding. Then the CTD directly interacts with DNA and further regulates the allosteric network (involving the S6-S7 loop, L1 loop, S4, S10, H1, and H3) to promote the binding of DBD to DNA. The allosteric mechanisms presented in this work will provide new insights into the functional mechanisms of the p53 CTD and inform the rational design of p53-targeted drugs.

The electrochemical hydrogen evolution reaction (HER) holds substantial promise for large-scale green hydrogen production due to its cost-effectiveness, high performance, and scalability for repeated implementation. A promising strategy that involves a novel transition-metal chalcogenide with a self-supported framework can improve the reaction kinetics in the HER. We synthesize a nano three-dimensional self-supported HER catalyst via a straightforward one-step chemical vapor deposition process. This method incorporates a dual co-reaction with sulfur (S) and selenium (Se) onto a commercially available Monel alloy (CNSSe). The dynamics ion exchange redox reactions promote the formation of heterogeneous catalyst structures. Density functional theory (DFT) calculations indicate that the CNSSe catalyst exhibits low hydrogen coverage, as evidenced by a thermoneutral free energy of adsorbed hydrogen (ΔG_H*) of 0.105 eV, which can be attributed to the dual introduction of S and Se. Consequently, the optimized CNSSe electrocatalyst achieves an enhancement in HER performance exceeding 100% compared to catalysts introduced exclusively with either S or Se. These results underscore the substantial potential of the optimized CNSSe electrocatalyst to improve both the performance and economic feasibility of HER technologies in alkaline water electrolysis.