Accurate prediction of the binding affinity of a protein-ligand complex is essential for efficient and successful rational drug design. Therefore, many binding affinity prediction methods have been developed. In recent years, since deep learning technology has become powerful, it is also implemented to predict affinity. In this work, a new neural network model that predicts the binding affinity of a protein-ligand complex structure is developed. Our model predicts the binding affinity of a complex using the ensemble of multiple independently trained networks that consist of multiple channels of 3-D convolutional neural network layers. Our model was trained using the 3772 protein-ligand complexes from the refined set of the PDBbind-2016 database and tested using the core set of 285 complexes. The benchmark results show that the Pearson correlation coefficient between the predicted binding affinities by our model and the experimental data is 0.827, which is higher than the state-of-the-art binding affinity prediction scoring functions. Additionally, our method ranks the relative binding affinities of possible multiple binders of a protein quite accurately, comparable to the other scoring functions. Last, we measured which structural information is critical for predicting binding affinity and found that the complementarity between the protein and ligand is most important.

It has been noticed that the efficiency of drug development has been decreasing in the past few decades. To overcome the situation, protein–protein interactions (PPIs) have been identified as new drug targets as early as 2000. PPIs are more abundant in human cells than single proteins and play numerous important roles in cellular processes including diseases. However, PPIs have very different physicochemical features from the conventional drug targets, which make targeting PPIs challenging. Therefore, as of now, only a small number of PPI inhibitors have been approved or progressed to a stage of clinical trial. In this article, we first overview previous works that analyzed differences between PPIs with PPI targeting ligands and conventional drugs with their binding pockets. Then, we constructed an up-to-date list of PPI targeting drugs that have been approved or are currently under clinical trial and have bound drug–target structures available. Using the dataset, we analyzed the PPIs and their ligands using several scores of druggability. Druggability scores showed that PPI sites and their drugs targeting PPIs are less druggable than conventional binding pockets and drugs, which also indicates that PPI drugs do not follow the conventional rules for drug design, such as Lipinski’s rule of five. Our analyses suggest that developing a new rule would be beneficial for guiding PPI-drug discovery.

Potential inhibitors of a target biomolecule, NAD-dependent deacetylase Sirtuin 1, were identified by a contest-based approach, in which participants were asked to propose a prioritized list of 400 compounds from a designated compound library containing 2.5 million compounds usingin silicomethods and scoring. Our aim was to identify target enzyme inhibitors and to benchmark computer-aided drug discovery methods under the same experimental conditions. Collecting compound lists derived from various methods is advantageous for aggregating compounds with structurally diversified properties compared with the use of a single method. The inhibitory action on Sirtuin 1 of approximately half of the proposed compounds was experimentally accessed. Ultimately, seven structurally diverse compounds were identified.

We introduce an advanced model for predicting protein–ligand interactions. Our approach combines the strengths of graph neural networks with physics-based scoring methods. Existing structure-based machine-learning models for protein–ligand binding prediction often fall short in practical virtual screening scenarios, hindered by the intricacies of binding poses, the chemical diversity of drug-like molecules, and the scarcity of crystallographic data for protein–ligand complexes. To overcome the limitations of existing machine learning-based prediction models, we propose a novel approach that fuses three independent neural network models. One classification model is designed to perform binary prediction of a given protein–ligand complex pose. The other two regression models are trained to predict the binding affinity and root-mean-square deviation of a ligand conformation from an input complex structure. We trained the model to account for both deviations in experimental and predicted binding affinities and pose prediction uncertainties. By effectively integrating the outputs of the triplet neural networks with a physics-based scoring function, our model showed a significantly improved performance in hit identification. The benchmark results with three independent decoy sets demonstrate that our model outperformed existing models in forward screening. Our model achieved top 1% enrichment factors of 32.7 and 23.1 with the CASF2016 and DUD-E benchmark sets, respectively. The benchmark results using the LIT-PCBA set further confirmed its higher average enrichment factors, emphasizing the model’s efficiency and generalizability. The model’s efficiency was further validated by identifying 23 active compounds from 63 candidates in experimental screening for autotaxin inhibitors, demonstrating its practical applicability in hit discovery.Scientific contributionOur work introduces a novel training strategy for a protein–ligand binding affinity prediction model by integrating the outputs of three independent sub-models and utilizing expertly crafted decoy sets. The model showcases exceptional performance across multiple benchmarks. The high enrichment factors in the LIT-PCBA benchmark demonstrate its potential to accelerate hit discovery.

Fragment-based drug discovery is a technique that finds potent binding fragments to the binding hotspots and makes them a hit compound. The combination of fragments allows us to explore the large chemical space. Thus, it becomes an effective methodology for identifying lead compounds. Three concepts have been introduced to make the fragments into the compound: growing, merging, and linking. Recently, growing and merging techniques using AI have significantly improved the accuracy and efficiency of molecular design. In this review, recent techniques such as VAE, reinforcement learning, and SE(3)-equivariant models will be discussed. These methods enable precise molecular structure exploration and optimization. Additionally, we address techniques utilizing diffusion models, language models, and deep evolutionary learning. We also introduce linker optimization methods using reinforcement learning and deep learning-based models. This progress of fragment-based drug discovery methods with AI opens the possibility of discovering the vast chemical space with high efficiency.