22 November 2023 in Journal of Chemical Information and Modeling
Molecular generation is crucial for advancing drug discovery, materials science, and chemical exploration. It expedites the search for new drug candidates, facilitates tailored material creation, and enhances our understanding of molecular diversity. By employing artificial intelligence techniques such as molecular generative models based on molecular graphs, researchers have tackled the challenge of identifying efficient molecules with desired properties. Here, we propose a new molecular generative model combining a graph-based deep neural network and a reinforcement learning technique. We evaluated the validity, novelty, and optimized physicochemical properties of the generated molecules. Importantly, the model explored uncharted regions of chemical space, allowing for the efficient discovery and design of new molecules. This innovative approach has considerable potential to revolutionize drug discovery, materials science, and chemical research for accelerating scientific innovation. By leveraging advanced techniques and exploring previously unexplored chemical spaces, this study offers promising prospects for the efficient discovery and design of new molecules in the field of drug development.
17 July 2023 in Journal of Chemical Information and Modeling
Identifying compound-protein interactions (CPIs) is crucial for drug discovery. Since experimentally validating CPIs is often time-consuming and costly, computational approaches are expected to facilitate the process. Rapid growths of available CPI databases have accelerated the development of many machine-learning methods for CPI predictions. However, their performance, particularly their generalizability against external data, often suffers from a data imbalance attributed to the lack of experimentally validated inactive (negative) samples. In this study, we developed a self-training method for augmenting both credible and informative negative samples to improve the performance of models impaired by data imbalances. The constructed model demonstrated higher performance than those constructed with other conventional methods for solving data imbalances, and the improvement was prominent for external datasets. Moreover, examination of the prediction score thresholds for pseudo-labeling during self-training revealed that augmenting the samples with ambiguous prediction scores is beneficial for constructing a model with high generalizability. The present study provides guidelines for improving CPI predictions on real-world data, thus facilitating drug discovery.
22 October 2025 in Journal of Chemical Information and Modeling
Early evaluation of absorption, distribution, metabolism, and excretion (ADME) properties is crucial for streamlining drug development. Traditional in vivo and in vitro approaches are often expensive. Moreover, during lead optimization, these methods rely heavily on the expertise of specialists, leading to efficiency challenges. Consequently, in silico methods for ADME prediction are attracting increasing attention. However, existing in silico methods face two major issues: a decline in predictive performance caused by limited ADME data and a lack of clarity regarding the rationale for lead optimization to improve ADME properties. In this study, we built an AI model capable of predicting ten different ADME parameters to overcome these challenges. Our training approach was based on a graph neural network combining multitask learning, which shares information across multiple tasks to increase the number of usable samples, with fine-tuning that adapts to each task. In addition, we applied the integrated gradients method to compound data collected before and after lead optimization to quantify and interpret each input feature's contribution to the predicted ADME values. Our proposed model achieved the highest performance for seven of the ten ADME parameters compared with conventional methods. Furthermore, visualization of the changes in chemical structures before and after lead optimization revealed that the model's explanations aligned well with established chemical insights. These results suggest that data-driven approaches may assist molecular design by providing complementary insights into empirical rules.
15 July 2026 in Journal of Chemical Information and Modeling
Chemically modified and macrocyclic peptides are increasingly important therapeutics, yet current molecular representation models do not natively represent chemical modification and covalent topology in a unified way. Atom-level strings obscure macrocyclic connectivity, whereas protein sequence models cannot encode noncanonical residues and explicit cross-links. Here we pretrain an encoder-only transformer directly on Hierarchical Editing Language for Macromolecules (HELM) notation, which specifies monomer identity and connectivity. In this work, we show that the resulting representations achieve best mean performance in cyclic peptide membrane permeability prediction (random split R2 = 0.668; retaining best mean performance under a Murcko scaffold split), exceeding external pretrained SMILES-based encoders. An architecture-matched SMILES control narrowed the HELM–SMILES gap under full finetuning, whereas HELM-BERT retained clearer advantages in frozen-representation settings. HELM-BERT also preserves HELM-specified macrocyclic topology in a linearly accessible form and supports competitive peptide–protein interaction prediction across complementary Propedia and ChEMBL benchmarks. More broadly, these findings suggest that pretraining directly on notation that makes structural constraints explicit offers a transferable strategy for biomolecular modalities that fall between small-molecule chemistry and protein sequence.
15 February 2024 in bioRxiv
Deep learning-based compound-protein interaction (CPI) prediction models are promising in the field of molecular biology, particularly for facilitating the drug discovery process. In practical applications, CPI models should achieve a high generalization performance, quantify prediction confidence, and ensure explainability. Here, we propose ChemGLaM, a chemical genomics language model for reliable and explainable CPI predictions, by addressing these three crucial aspects. ChemGLaM integrates independently pre-trained chemical and protein language models through an interaction block with a cross-attention mechanism, achieving state-of-the-art performance in predicting novel CPIs. Incorporating uncertainty estimation and attention visualization enables ChemGLaM to enhance the success rate of virtual screening and to provide molecular insights into CPIs. Furthermore, we demonstrate its practical applicability by constructing a public database for large-scale CPI predictions and enabling drug/target exploration for candidate treatment of amyotrophic lateral sclerosis (ALS). ChemGLaM represents a significant step toward overcoming the challenges of AI-driven drug discovery and addressing unmet medical needs.
23 September 2025 in ChemRxiv
Accurately predicting novel compound-protein interactions (CPIs) is essential for accelerating drug discovery. The generalizability of machine learning-based CPI prediction models relies significantly on the availability and diversity of CPI datasets. To maximize data utility, particularly for highly confidential datasets maintained by industries, federated learning (FL)—which integrates multi-site data while preserving privacy—has emerged as a promising approach. Nonetheless, its effectiveness when applied to heterogeneous data from diverse molecular domains, a common real-world scenario, remains unclear, thereby limiting its broader adoption. This study evaluates FL for CPI prediction using datasets spanning multiple chemical and protein domains, providing practical guidance for optimizing the FL approach. Results indicate that the FL model enhanced out-of-domain prediction performance but was surpassed by local models for in-domain data under data heterogeneity. Drawing on these findings, a new strategy was developed to achieve robust performance for in- and out-of-domain tasks: a similarity-guided ensemble that combines the global FL model with fine-tuned models based on each client's local data. This method demonstrated effectiveness with real-world industry data, including samples from the public database and 13 pharmaceutical companies. Cumulatively, these findings offer practical guidance for implementing FL in contemporary drug discovery processes.