Predicting protein–ligand binding affinities (PLAs) is a core problem in drug discovery. Recent advances have shown great potential in applying machine learning (ML) for PLA prediction. However, most of them omit the 3D structures of complexes and physical interactions between proteins and ligands, which are considered essential to understanding the binding mechanism. This paper proposes a geometric interaction graph neural network (GIGN) that incorporates 3D structures and physical interactions for predicting protein–ligand binding affinities. Specifically, we design a heterogeneous interaction layer that unifies covalent and noncovalent interactions into the message passing phase to learn node representations more effectively. The heterogeneous interaction layer also follows fundamental biological laws, including invariance to translations and rotations of the complexes, thus avoiding expensive data augmentation strategies. GIGN achieves state-of-the-art performance on three external test sets. Moreover, by visualizing learned representations of protein–ligand complexes, we show that the predictions of GIGN are biologically meaningful.

SA-DDI is designed to learn size-adaptive molecular substructures for drug–drug interaction prediction and can provide explanations that are consistent with pharmacologists.

With the rapid development of proteomics and the rapid increase of target molecules for drug action, computer-aided drug design (CADD) has become a basic task in drug discovery. One of the key challenges in CADD is molecular representation. High-quality molecular expression with chemical intuition helps to promote many boundary problems of drug discovery. At present, molecular representation still faces several urgent problems, such as the polysemy of substructures and unsmooth information flow between atomic groups. In this research, we propose a deep contextualized Bi-LSTM architecture, Mol2Context-vec, which can integrate different levels of internal states to bring dynamic representations of molecular substructures. And the obtained molecular context representation can capture the interactions between any atomic groups, especially a pair of atomic groups that are topologically distant. Experiments show that Mol2Context-vec achieves state-of-the-art performance on multiple benchmark datasets. In addition, the visual interpretation of Mol2Context-vec is very close to the structural properties of chemical molecules as understood by humans. These advantages indicate that Mol2Context-vec can be used as a reliable and effective tool for molecular expression. Availability: The source code is available for download in https://github.com/lol88/Mol2Context-vec.

It has been demonstrated that MMP13 enzyme is related to most cancer cell tumors. The world’s largest traditional Chinese medicine database was applied to screen for structure-based drug design and ligand-based drug design. To predict drug activity, machine learning models (Random Forest (RF), AdaBoost Regressor (ABR), Gradient Boosting Regressor (GBR)), and Deep Learning models were utilized to validate the Docking results, and we obtained an R 2 of 0.922 on the training set and 0.804 on the test set in the RF algorithm. For the Deep Learning algorithm, R 2 of the training set is 0.90, and R 2 of the test set is 0.810. However, these TCM compounds fly away during the molecular dynamics (MD) simulation. We seek another method: peptide design. All peptide database were screened by the Docking process. Modification peptides were optimized the interaction modes, and the affinities were assessed with ZDOCK protocol and Refine Docked protein protocol. The 300 ns MD simulation evaluated the stability of receptor–peptide complexes. The double-site effect appeared on S2, a designed peptide based on a known inhibitor, when complexed with BCL2. S3, a designed peptide referred from endogenous inhibitor P16, competed against cyclin when binding with CDK6. The MDM2 inhibitors S5 and S6 were derived from the P53 structure and stable binding with MDM2. A flexible region of peptides S5 and S6 may enhance the binding ability by changing its own conformation, which was unforeseen. These peptides (S2, S3, S5, and S6) are potentially interesting to treat cancer; however, these findings need to be affirmed by biological testing, which will be conducted in the near future.

No abstract