Gram matrix: an efficient representation of molecular conformation and learning objective for molecular pretraining.

Accurate prediction of molecular properties is fundamental in drug discovery and development, providing crucial guidance for effective drug design. A critical factor in achieving accurate molecular property prediction lies in the appropriate representation of molecular structures. Presently, prevalent deep learning-based molecular representations rely on 2D structure information as the primary molecular representation, often overlooking essential three-dimensional (3D) conformational information due to the inherent limitations of 2D structures in conveying atomic spatial relationships. In this study, we propose employing the Gram matrix as a condensed representation of 3D molecular structures and for efficient pretraining objectives. Subsequently, we leverage this matrix to construct a novel molecular representation model, Pre-GTM, which inherently encapsulates 3D information. The model accurately predicts the 3D structure of a molecule by estimating the Gram matrix. Our findings demonstrate that Pre-GTM model outperforms the baseline Graphormer model and other pretrained models in the QM9 and MoleculeNet quantitative property prediction task. The integration of the Gram matrix as a condensed representation of 3D molecular structure, incorporated into the Pre-GTM model, opens up promising avenues for its potential application across various domains of molecular research, including drug design, materials science, and chemical engineering.

MolFilterGAN: a progressively augmented generative adversarial network for triaging AI-designed molecules.

Artificial intelligence (AI)-based molecular design methods, especially deep generative models for generating novel molecule structures, have gratified our imagination to explore unknown chemical space without relying on brute-force exploration. However, whether designed by AI or human experts, the molecules need to be accessibly synthesized and biologically evaluated, and the trial-and-error process remains a resources-intensive endeavor. Therefore, AI-based drug design methods face a major challenge of how to prioritize the molecular structures with potential for subsequent drug development. This study indicates that common filtering approaches based on traditional screening metrics fail to differentiate AI-designed molecules. To address this issue, we propose a novel molecular filtering method, MolFilterGAN, based on a progressively augmented generative adversarial network. Comparative analysis shows that MolFilterGAN outperforms conventional screening approaches based on drug-likeness or synthetic ability metrics. Retrospective analysis of AI-designed discoidin domain receptor 1 (DDR1) inhibitors shows that MolFilterGAN significantly increases the efficiency of molecular triaging. Further evaluation of MolFilterGAN on eight external ligand sets suggests that MolFilterGAN is useful in triaging or enriching bioactive compounds across a wide range of target types. These results highlighted the importance of MolFilterGAN in evaluating molecules integrally and further accelerating molecular discovery especially combined with advanced AI generative models.

Graph neural network approaches for drug-target interactions.

Developing new drugs remains prohibitively expensive, time-consuming, and often involves safety issues. Accurate prediction of drug-target interactions (DTIs) can guide the drug discovery process and thus facilitate drug development. Non-Euclidian data such as drug-like molecule structures, key pocket residue structures, and protein interaction networks can be represented effectively using graphs. Therefore, the emerging graph neural network has been rapidly applied to predict DTIs, and proved effective in finding repositioning drugs and accelerating drug discovery. In this review, we provide a brief overview of deep neural networks used in DTI models. Then, we summarize the database required for DTI prediction, followed by a comprehensive introduction of applications of graph neural networks for DTI prediction. We also highlight current challenges and future directions to guide the further development of this field.

Drug target inference by mining transcriptional data using a novel graph convolutional network framework.

A fundamental challenge that arises in biomedicine is the need to characterize compounds in a relevant cellular context in order to reveal potential on-target or off-target effects. Recently, the fast accumulation of gene transcriptional profiling data provides us an unprecedented opportunity to explore the protein targets of chemical compounds from the perspective of cell transcriptomics and RNA biology. Here, we propose a novel Siamese spectral-based graph convolutional network (SSGCN) model for inferring the protein targets of chemical compounds from gene transcriptional profiles. Although the gene signature of a compound perturbation only provides indirect clues of the interacting targets, and the biological networks under different experiment conditions further complicate the situation, the SSGCN model was successfully trained to learn from known compound-target pairs by uncovering the hidden correlations between compound perturbation profiles and gene knockdown profiles. On a benchmark set and a large time-split validation dataset, the model achieved higher target inference accuracy as compared to previous methods such as Connectivity Map. Further experimental validations of prediction results highlight the practical usefulness of SSGCN in either inferring the interacting targets of compound, or reversely, in finding novel inhibitors of a given target of interest.

Multi-instance learning of graph neural networks for aqueous pKa prediction.

MOTIVATION: The acid dissociation constant (pKa) is a critical parameter to reflect the ionization ability of chemical compounds and is widely applied in a variety of industries. However, the experimental determination of pKa is intricate and time-consuming, especially for the exact determination of micro-pKa information at the atomic level. Hence, a fast and accurate prediction of pKa values of chemical compounds is of broad interest. RESULTS: Here, we compiled a large-scale pKa dataset containing 16 595 compounds with 17 489 pKa values. Based on this dataset, a novel pKa prediction model, named Graph-pKa, was established using graph neural networks. Graph-pKa performed well on the prediction of macro-pKa values, with a mean absolute error around 0.55 and a coefficient of determination around 0.92 on the test dataset. Furthermore, combining multi-instance learning, Graph-pKa was also able to automatically deconvolute the predicted macro-pKa into discrete micro-pKa values. AVAILABILITY AND IMPLEMENTATION: The Graph-pKa model is now freely accessible via a web-based interface (https://pka.simm.ac.cn/). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Regulating Glucose Metabolism with Prodrug Nanoparticles for Promoting Photoimmunotherapy of Pancreatic Cancer.

The low immunogenicity, insufficient infiltration of T lymphocytes, and dismal response to immune checkpoint blockade therapy pose major difficulties in immunotherapy of pancreatic cancer. Photoimmunotherapy by photodynamic therapy (PDT) can induce an antitumor immune response by triggering immunogenic cell death in the tumor cells. Notwithstanding, PDT-driven oxygen consumption and microvascular damage can further aggravate hypoxia to exaggerates glycolysis, leading to lactate accumulation and immunosuppressive tumor microenvironment. Herein, a supramolecular prodrug nanoplatform codelivering a photosensitizer and a prodrug of bromodomain-containing protein 4 inhibitor (BRD4i) JQ1 for combinatory photoimmunotherapy of pancreatic cancer are demonstrated. The nanoparticles are fabricated by host-guest complexation between cyclodextrin-grafted hyaluronic acid (HA-CD) and adamantine-conjugated heterodimers of pyropheophorbide a (PPa) and JQ1, respectively. HA can achieve active tumor targeting by recognizing highly expressed CD44 on the surface of pancreatic tumors. PPa-mediated PDT can enhance the immunogenicity of the tumor cells and promote intratumoral infiltration of the cytotoxic T lymphocytes. Meanwhile, JQ1 combats PDT-mediated immune evasion through inhibiting expression of c-Myc and PD-L1, which are key regulators of tumor glycolysis and immune evasion. Collectively, this study presents a novel strategy to enhance photoimmunotherapy of the pancreatic cancer by provoking T cells activation and overcoming adaptive immune resistance.

Automated design and optimization of multitarget schizophrenia drug candidates by deep learning.

Complex neuropsychiatric diseases such as schizophrenia require drugs that can target multiple G protein-coupled receptors (GPCRs) to modulate complex neuropsychiatric functions. Here, we report an automated system comprising a deep recurrent neural network (RNN) and a multitask deep neural network (MTDNN) to design and optimize multitarget antipsychotic drugs. The system has successfully generated novel molecule structures with desired multiple target activities, among which high-ranking compound 3 was synthesized, and demonstrated potent activities against dopamine D2, serotonin 5-HT1A and 5-HT2A receptors. Hit expansion based on the MTDNN was performed, 6 analogs of compound 3 were evaluated experimentally, among which compound 8 not only exhibited specific polypharmacology profiles but also showed antipsychotic effect in animal models with low potential for sedation and catalepsy, highlighting their suitability for further preclinical studies. The approach can be an efficient tool for designing lead compounds with multitarget profiles to achieve the desired efficacy in the treatment of complex neuropsychiatric diseases.

TransformerCPI: improving compound-protein interaction prediction by sequence-based deep learning with self-attention mechanism and label reversal experiments.

MOTIVATION: Identifying compound-protein interaction (CPI) is a crucial task in drug discovery and chemogenomics studies, and proteins without three-dimensional structure account for a large part of potential biological targets, which requires developing methods using only protein sequence information to predict CPI. However, sequence-based CPI models may face some specific pitfalls, including using inappropriate datasets, hidden ligand bias and splitting datasets inappropriately, resulting in overestimation of their prediction performance. RESULTS: To address these issues, we here constructed new datasets specific for CPI prediction, proposed a novel transformer neural network named TransformerCPI, and introduced a more rigorous label reversal experiment to test whether a model learns true interaction features. TransformerCPI achieved much improved performance on the new experiments, and it can be deconvolved to highlight important interacting regions of protein sequences and compound atoms, which may contribute chemical biology studies with useful guidance for further ligand structural optimization. AVAILABILITY AND IMPLEMENTATION: https://github.com/lifanchen-simm/transformerCPI.

Deep Learning Enhancing Kinome-Wide Polypharmacology Profiling: Model Construction and Experiment Validation.

The kinome-wide virtual profiling of small molecules with high-dimensional structure-activity data is a challenging task in drug discovery. Here, we present a virtual profiling model against a panel of 391 kinases based on large-scale bioactivity data and the multitask deep neural network algorithm. The obtained model yields excellent internal prediction capability with an auROC of 0.90 and consistently outperforms conventional single-task models on external tests, especially for kinases with insufficient activity data. Moreover, more rigorous experimental validations including 1410 kinase-compound pairs showed a high-quality average auROC of 0.75 and confirmed many novel predicted "off-target" activities. Given the verified generalizability, the model was further applied to various scenarios for depicting the kinome-wide selectivity and the association with certain diseases. Overall, the computational model enables us to create a comprehensive kinome interaction network for designing novel chemical modulators or drug repositioning and is of practical value for exploring previously less studied kinases.

Pushing the Boundaries of Molecular Representation for Drug Discovery with the Graph Attention Mechanism.

Hunting for chemicals with favorable pharmacological, toxicological, and pharmacokinetic properties remains a formidable challenge for drug discovery. Deep learning provides us with powerful tools to build predictive models that are appropriate for the rising amounts of data, but the gap between what these neural networks learn and what human beings can comprehend is growing. Moreover, this gap may induce distrust and restrict deep learning applications in practice. Here, we introduce a new graph neural network architecture called Attentive FP for molecular representation that uses a graph attention mechanism to learn from relevant drug discovery data sets. We demonstrate that Attentive FP achieves state-of-the-art predictive performances on a variety of data sets and that what it learns is interpretable. The feature visualization for Attentive FP suggests that it automatically learns nonlocal intramolecular interactions from specified tasks, which can help us gain chemical insights directly from data beyond human perception.

KinomeX: a web application for predicting kinome-wide polypharmacology effect of small molecules.

MOTIVATION: The large-scale kinome-wide virtual profiling for small molecules is a daunting task by experimental and traditional in silico drug design approaches. Recent advances in deep learning algorithms have brought about new opportunities in promoting this process. RESULTS: KinomeX is an online platform to predict kinome-wide polypharmacology effect of small molecules based solely on their chemical structures. The prediction is made by a multi-task deep neural network model trained with over 140 000 bioactivity data points for 391 kinases. Extensive computational and experimental validations have been performed. Overall, KinomeX enables users to create a comprehensive kinome interaction network for designing novel chemical modulators, and is of practical value on exploring the previously less studied or untargeted kinases. AVAILABILITY AND IMPLEMENTATION: KinomeX is available at: https://kinome.dddc.ac.cn. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Artificial intelligence in drug design.

Thanks to the fast improvement of the computing power and the rapid development of the computational chemistry and biology, the computer-aided drug design techniques have been successfully applied in almost every stage of the drug discovery and development pipeline to speed up the process of research and reduce the cost and risk related to preclinical and clinical trials. Owing to the development of machine learning theory and the accumulation of pharmacological data, the artificial intelligence (AI) technology, as a powerful data mining tool, has cut a figure in various fields of the drug design, such as virtual screening, activity scoring, quantitative structure-activity relationship (QSAR) analysis, de novo drug design, and in silico evaluation of absorption, distribution, metabolism, excretion and toxicity (ADME/T) properties. Although it is still challenging to provide a physical explanation of the AI-based models, it indeed has been acting as a great power to help manipulating the drug discovery through the versatile frameworks. Recently, due to the strong generalization ability and powerful feature extraction capability, deep learning methods have been employed in predicting the molecular properties as well as generating the desired molecules, which will further promote the application of AI technologies in the field of drug design.