Improving the generalizability of protein-ligand binding predictions with AI-Bind

Accurate prediction of drug-target interactions in Chinese and western medicine by the CWI-DTI model

Accurate prediction of drug-target interactions (DTIs) is crucial for advancing drug discovery and repurposing. Computational methods have significantly improved the efficiency of experimental predictions for drug-target interactions in Western medicine. However, accurately predicting the complex relationships between Chinese medicine ingredients and targets remains a formidable challenge due to the vast number and high heterogeneity of these ingredients. In this study, we introduce the CWI-DTI method, which achieves high-accuracy prediction of DTIs using a large dataset of interactive relationships of drug ingredients or candidate targets. Moreover, we present a novel dataset to evaluate the prediction accuracy of both Chinese and Western medicine. Through meticulous collection and preprocessing of data on ingredients and targets, we employ an innovative autoencoder framework to fuse multiple drug (target) topological similarity matrices. Additionally, we employ denoising blocks, sparse blocks, and stacked blocks to extract crucial features from the similarity matrix, reducing noise and enhancing accuracy across diverse datasets. Our results indicate that the CWI-DTI model shows improved performance compared to several existing state-of-the-art methods on the datasets tested in both Western and Chinese medicine databases. The findings of this study hold immense promise for advancing DTI prediction in Chinese and Western medicine, thus fostering more efficient drug discovery and repurposing endeavors. Our model is available at https://github.com/WANG-BIN-LAB/CWIDTI .

Identifying compound-protein interactions with knowledge graph embedding of perturbation transcriptomics

The emergence of perturbation transcriptomics provides a new perspective for drug discovery, but existing analysis methods suffer from inadequate performance and limited applicability. In this work, we present PertKGE, a method designed to deconvolute compound-protein interactions from perturbation transcriptomics with knowledge graph embedding. By considering multi-level regulatory events within biological systems that share the same semantic context, PertKGE significantly improves deconvoluting accuracy in two critical "cold-start" settings: inferring targets for new compounds and conducting virtual screening for new targets. We further demonstrate the pivotal role of incorporating multi-level regulatory events in alleviating representational biases. Notably, it enables the identification of ectonucleotide pyrophosphatase/phosphodiesterase-1 as the target responsible for the unique anti-tumor immunotherapy effect of tankyrase inhibitor K-756 and the discovery of five novel hits targeting the emerging cancer therapeutic target aldehyde dehydrogenase 1B1 with a remarkable hit rate of 10.2%. These findings highlight the potential of PertKGE to accelerate drug discovery.

A chemical platform for the efficient screening of arylazopyrazole-based photoswitchable CENP-E inhibitors using mild cyclization reactions

A set of arylazopyrazole-based inhibitors targeting the mitotic motor protein CENP-E was discovered through the chemical platform using the quantitative cyclization of 1,3-diketone intermediate with various hydrazines under mild conditions. Through this efficient platform, the structure-activity relationship pertaining to the pyrazole photoswitch in photoswitchable CENP-E inhibitors not only in vitro but also in cells was successfully clarified.

Accelerating drug discovery, development, and clinical trials by artificial intelligence

Artificial intelligence (AI) has profoundly advanced the field of biomedical research, which also demonstrates transformative capacity for innovation in drug development. This paper aims to deliver a comprehensive analysis of the progress in AI-assisted drug development, particularly focusing on small molecules, RNA, and antibodies. Moreover, this paper elucidates the current integration of AI methodologies within the industrial drug development framework. This encompasses a detailed examination of the industry-standard drug development process, supplemented by a review of medications presently undergoing clinical trials. Conclusively, the paper tackles a predominant obstacle within the AI pharmaceutical sector: the absence of AI-conceived drugs receiving approval. This paper also advocates for the adoption of large language models and diffusion models as a viable strategy to surmount this challenge. This review not only underscores the significant potential of AI in drug discovery but also deliberates on the challenges and prospects within this dynamically progressing field.

Multi-modal deep learning enables efficient and accurate annotation of enzymatic active sites

Annotating active sites in enzymes is crucial for advancing multiple fields including drug discovery, disease research, enzyme engineering, and synthetic biology. Despite the development of numerous automated annotation algorithms, a significant trade-off between speed and accuracy limits their large-scale practical applications. We introduce EasIFA, an enzyme active site annotation algorithm that fuses latent enzyme representations from the Protein Language Model and 3D structural encoder, and then aligns protein-level information with the knowledge of enzymatic reactions using a multi-modal cross-attention framework. EasIFA outperforms BLASTp with a 10-fold speed increase and improved recall, precision, f1 score, and MCC by 7.57%, 13.08%, 9.68%, and 0.1012, respectively. It also surpasses empirical-rule-based algorithm and other state-of-the-art deep learning annotation method based on PSSM features, achieving a speed increase ranging from 650 to 1400 times while enhancing annotation quality. This makes EasIFA a suitable replacement for conventional tools in both industrial and academic settings. EasIFA can also effectively transfer knowledge gained from coarsely annotated enzyme databases to smaller, high-precision datasets, highlighting its ability to model sparse and high-quality databases. Additionally, EasIFA shows potential as a catalytic site monitoring tool for designing enzymes with desired functions beyond their natural distribution.

Effectiveness and Efficiency: Label-Aware Hierarchical Subgraph Learning for Protein-Protein Interaction

The study of protein-protein interactions (PPIs) holds immense significance in understanding various biological activities, as well as in drug discovery and disease diagnosis. Existing deep learning methods for PPI prediction, including graph neural networks (GNNs), have been widely employed as the solutions, while they often experience a decline in performance in the real world. We claim that the topological shortcut is one of the key problems contributing negatively to the performance, according to our analysis. By modeling the PPIs as a graph with protein as nodes and interactions as edge types, the prevailing models tend to learn the pattern of nodes' degrees rather than intrinsic sequence-structure profiles, leading to the problem termed topological shortcut. The huge data growth of PPI leads to intensive computational costs and challenges computing devices, causing infeasibility in practice. To address the discussed problems, we propose a label-aware hierarchical subgraph learning method (laruGL-PPI) that can effectively infer PPIs while being interpretable. Specifically, we introduced edge-based subgraph sampling to effectively alleviate the problems of topological shortcuts and high computing costs. Besides, the inner-outer connections of PPIs are modeled as a hierarchical graph, together with the dependencies between interaction types constructed by a label graph. Extensive experiments conducted across various scales of PPI datasets have conclusively demonstrated that the laruGL-PPI method surpasses the most advanced PPI prediction techniques currently available, particularly in the testing of unseen proteins. Also, our model can recognize crucial sites of proteins, such as surface sites for binding and active sites for catalysis.

Decoding the Foodome: Molecular Networks Connecting Diet and Health

Diet, a modifiable risk factor, plays a pivotal role in most diseases, from cardiovascular disease to type 2 diabetes mellitus, cancer, and obesity. However, our understanding of the mechanistic role of the chemical compounds found in food remains incomplete. In this review, we explore the "dark matter" of nutrition, going beyond the macro- and micronutrients documented by national databases to unveil the exceptional chemical diversity of food composition. We also discuss the need to explore the impact of each compound in the presence of associated chemicals and relevant food sources and describe the tools that will allow us to do so. Finally, we discuss the role of network medicine in understanding the mechanism of action of each food molecule. Overall, we illustrate the important role of network science and artificial intelligence in our ability to reveal nutrition's multifaceted role in health and disease.

Guiding questions to avoid data leakage in biological machine learning applications

Machine learning methods for extracting patterns from high-dimensional data are very important in the biological sciences. However, in certain cases, real-world applications cannot confirm the reported prediction performance. One of the main reasons for this is data leakage, which can be seen as the illicit sharing of information between the training data and the test data, resulting in performance estimates that are far better than the performance observed in the intended application scenario. Data leakage can be difficult to detect in biological datasets due to their complex dependencies. With this in mind, we present seven questions that should be asked to prevent data leakage when constructing machine learning models in biological domains. We illustrate the usefulness of our questions by applying them to nontrivial examples. Our goal is to raise awareness of potential data leakage problems and to promote robust and reproducible machine learning-based research in biology.

Synthetic lethal connectivity and graph transformer improve synthetic lethality prediction

Synthetic lethality (SL) has shown great promise for the discovery of novel targets in cancer. CRISPR double-knockout (CDKO) technologies can only screen several hundred genes and their combinations, but not genome-wide. Therefore, good SL prediction models are highly needed for genes and gene pairs selection in CDKO experiments. However, lack of scalable SL properties prevents generalizability of SL interactions to out-of-sample data, thereby hindering modeling efforts. In this paper, we recognize that SL connectivity is a scalable and generalizable SL property. We develop a novel two-step multilayer encoder for individual sample-specific SL prediction model (MLEC-iSL), which predicts SL connectivity first and SL interactions subsequently. MLEC-iSL has three encoders, namely, gene, graph, and transformer encoders. MLEC-iSL achieves high SL prediction performance in K562 (AUPR, 0.73; AUC, 0.72) and Jurkat (AUPR, 0.73; AUC, 0.71) cells, while no existing methods exceed 0.62 AUPR and AUC. The prediction performance of MLEC-iSL is validated in a CDKO experiment in 22Rv1 cells, yielding a 46.8% SL rate among 987 selected gene pairs. The screen also reveals SL dependency between apoptosis and mitosis cell death pathways.

Versatile Framework for Drug-Target Interaction Prediction by Considering Domain-Specific Features

Predicting drug-target interactions (DTIs) is one of the crucial tasks in drug discovery, but traditional wet-lab experiments are costly and time-consuming. Recently, deep learning has emerged as a promising tool for accelerating DTI prediction due to its powerful performance. However, the models trained on limited known DTI data struggle to generalize effectively to novel drug-target pairs. In this work, we propose a strategy to train an ensemble of models by capturing both domain-generic and domain-specific features (E-DIS) to learn diverse domain features and adapt them to out-of-distribution data. Multiple experts were trained on different domains to capture and align domain-specific information from various distributions without accessing any data from unseen domains. E-DIS provides a comprehensive representation of proteins and ligands by capturing diverse features. Experimental results on four benchmark data sets in both in-domain and cross-domain settings demonstrated that E-DIS significantly improved model performance and domain generalization compared to existing methods. Our approach presents a significant advancement in DTI prediction by combining domain-generic and domain-specific features, enhancing the generalization ability of the DTI prediction model.

CPIExtract: A software package to collect and harmonize small molecule and protein interactions

The binding interactions between small molecules and proteins are the basis of cellular functions. Yet, experimental data available regarding compound-protein interaction is not harmonized into a single entity but rather scattered across multiple institutions, each maintaining databases with different formats. Extracting information from these multiple sources remains challenging due to data heterogeneity. Here, we present CPIExtract (Compound-Protein Interaction Extract), a tool to interactively extract experimental binding interaction data from multiple databases, perform filtering, and harmonize the resulting information, thus providing a gain of compound-protein interaction data. When compared to a single source, DrugBank, we show that it can collect more than 10 times the amount of annotations. The end-user can apply custom filtering to the aggregated output data and save it in any generic tabular file suitable for further downstream tasks such as network medicine analyses for drug repurposing and cross-validation of deep learning models.

Multi-target drugs for Alzheimer's disease

Alzheimer's disease (AD), a leading cause of dementia, increasingly challenges our healthcare systems and society. Traditional therapies aimed at single targets have fallen short owing to the complex, multifactorial nature of AD that necessitates simultaneous targeting of various disease mechanisms for clinical success. Therefore, targeting multiple pathologies at the same time could provide a synergistic therapeutic effect. The identification of new disease targets beyond the classical hallmarks of AD offers a fertile ground for the design of new multi-target drugs (MTDs), and building on existing compounds have the potential to yield in successful disease modifying therapies. This review discusses the evolving landscape of MTDs, focusing on their potential as AD therapeutics. Analysis of past and current trials of compounds with multi-target activity underscores the capacity of MTDs to offer synergistic therapeutic effects, and the flourishing genetic understanding of AD will inform and inspire the development of MTD-based AD therapies.

GroceryDB: Prevalence of Processed Food in Grocery Stores

The offering of grocery stores is a strong driver of consumer decisions, shaping their diet and long-term health. While highly processed food like packaged products, processed meat, and sweetened soft drinks have been increasingly associated with unhealthy diet, information on the degree of processing characterizing an item in a store is not straightforward to obtain, limiting the ability of individuals to make informed choices. Here we introduce GroceryDB, a database with over 50,000 food items sold by Walmart, Target, and Wholefoods, unveiling how big data can be harnessed to empower consumers and policymakers with systematic access to the degree of processing of the foods they select, and the potential alternatives in the surrounding food environment. The wealth of data collected on ingredient lists and nutrition facts allows a large scale analysis of ingredient patterns and degree of processing stratified by store, food category, and price range. We find that the nutritional choices of the consumers, translated as the degree of food processing, strongly depend on the food categories and grocery stores. Moreover, the data allows us to quantify the individual contribution of over 1,000 ingredients to ultra-processing. GroceryDB and the associated http://TrueFood.Tech/ website make this information accessible, guiding consumers toward less processed food choices while assisting policymakers in reforming the food supply.

An end-to-end method for predicting compound-protein interactions based on simplified homogeneous graph convolutional network and pre-trained language model

Identification of interactions between chemical compounds and proteins is crucial for various applications, including drug discovery, target identification, network pharmacology, and elucidation of protein functions. Deep neural network-based approaches are becoming increasingly popular in efficiently identifying compound-protein interactions with high-throughput capabilities, narrowing down the scope of candidates for traditional labor-intensive, time-consuming and expensive experimental techniques. In this study, we proposed an end-to-end approach termed SPVec-SGCN-CPI, which utilized simplified graph convolutional network (SGCN) model with low-dimensional and continuous features generated from our previously developed model SPVec and graph topology information to predict compound-protein interactions. The SGCN technique, dividing the local neighborhood aggregation and nonlinearity layer-wise propagation steps, effectively aggregates K-order neighbor information while avoiding neighbor explosion and expediting training. The performance of the SPVec-SGCN-CPI method was assessed across three datasets and compared against four machine learning- and deep learning-based methods, as well as six state-of-the-art methods. Experimental results revealed that SPVec-SGCN-CPI outperformed all these competing methods, particularly excelling in unbalanced data scenarios. By propagating node features and topological information to the feature space, SPVec-SGCN-CPI effectively incorporates interactions between compounds and proteins, enabling the fusion of heterogeneity. Furthermore, our method scored all unlabeled data in ChEMBL, confirming the top five ranked compound-protein interactions through molecular docking and existing evidence. These findings suggest that our model can reliably uncover compound-protein interactions within unlabeled compound-protein pairs, carrying substantial implications for drug re-profiling and discovery. In summary, SPVec-SGCN demonstrates its efficacy in accurately predicting compound-protein interactions, showcasing potential to enhance target identification and streamline drug discovery processes.Scientific contributionsThe methodology presented in this work not only enables the comparatively accurate prediction of compound-protein interactions but also, for the first time, take sample imbalance which is very common in real world and computation efficiency into consideration simultaneously, accelerating the target identification and drug discovery process.

Decoding the molecular interplay in the central dogma: An overview of mass spectrometry-based methods to investigate protein-metabolite interactions

With the emergence of next-generation nucleotide sequencing and mass spectrometry-based proteomics and metabolomics tools, we have comprehensive and scalable methods to analyze the genes, transcripts, proteins, and metabolites of a multitude of biological systems. Despite the fascinating new molecular insights at the genome, transcriptome, proteome and metabolome scale, we are still far from fully understanding cellular organization, cell cycles and biology at the molecular level. Significant advances in sensitivity and depth for both sequencing as well as mass spectrometry-based methods allow the analysis at the single cell and single molecule level. At the same time, new tools are emerging that enable the investigation of molecular interactions throughout the central dogma of molecular biology. In this review, we provide an overview of established and recently developed mass spectrometry-based tools to probe metabolite-protein interactions-from individual interaction pairs to interactions at the proteome-metabolome scale.

From petals to healing: consolidated network pharmacology and molecular docking investigations of the mechanisms underpinning Rhododendron arboreum flower's anti-NAFLD effects

Rhododendron arboreum: Sm., also known as Burans is traditionally used as an anti-inflammatory, anti-diabetic, hepatoprotective, adaptogenic, and anti-oxidative agent. It has been used since ancient times in Indian traditional medicine for various liver disorders. However, the exact mechanism behind its activity against NAFLD is not known. The aim of the present study is to investigate the molecular mechanism of Rhododendron arboreum flower (RAF) in the treatment of NAFLD using network pharmacology and molecular docking methods. Bioactives were also predicted for their drug-likeness score, probable side effects and ADMET profile. Protein-protein interaction (PPI) data was obtained using the STRING platform. For the visualisation of GO analysis, a bioinformatics server was employed. Through molecular docking, the binding affinity between potential targets and active compounds were assessed. A total of five active compounds of RAF and 30 target proteins were selected. The targets with higher degrees were identified through the PPI network. GO analysis indicated that the NAFLD treatment with RAF primarily entails a response to the fatty acid biosynthetic process, lipid metabolic process, regulation of cell death, regulation of stress response, and cellular response to a chemical stimulus. Molecular docking and molecular dynamic simulation exhibited that rutin has best binding affinity among active compounds and selected targets as indicated by the binding energy, RMSD, and RMSF data. The findings comprehensively elucidated toxicity data, potential targets of bioactives and molecular mechanisms of RAF against NAFLD, providing a promising novel strategy for future research on NAFLD treatment.

Proper network randomization is key to assessing social balance

Social ties, either positive or negative, lead to signed network patterns, the subject of balance theory. For example, strong balance introduces cycles with even numbers of negative edges. The statistical significance of such patterns is routinely assessed by comparisons to null models. Yet, results in signed networks remain controversial. Here, we show that even if a network exhibits strong balance by construction, current null models can fail to identify it. Our results indicate that matching the signed degree preferences of the nodes is a critical step and so is the preservation of network topology in the null model. As a solution, we propose the STP null model, which integrates both constraints within a maximum entropy framework. STP randomization leads to qualitatively different results, with most social networks consistently demonstrating strong balance in three- and four-node patterns. On the basis our results, we present a potential wiring mechanism behind the observed signed patterns and outline further applications of STP randomization.

Heterogeneous sampled subgraph neural networks with knowledge distillation to enhance double-blind compound-protein interaction prediction

Identifying binding compounds against a target protein is crucial for large-scale virtual screening in drug development. Recently, network-based methods have been developed for compound-protein interaction (CPI) prediction. However, they are difficult to be applied to unseen (i.e., never-seen-before) proteins and compounds. In this study, we propose SgCPI to incorporate local known interacting networks to predict CPI interactions. SgCPI randomly samples the local CPI network of the query compound-protein pair as a subgraph and applies a heterogeneous graph neural network (HGNN) to embed the active/inactive message of the subgraph. For unseen compounds and proteins, SgCPI-KD takes SgCPI as the teacher model to distillate its knowledge by estimating the potential neighbors. Experimental results indicate: (1) the sampled subgraphs of the CPI network introduce efficient knowledge for unseen molecular prediction with the HGNNs, and (2) the knowledge distillation strategy is beneficial to the double-blind interaction prediction by estimating molecular neighbors and distilling knowledge.

Multi-level bioinformatics resources support drug target discovery of protein-protein interactions

Drug discovery often begins with a new target. Protein-protein interactions (PPIs) are crucial to multitudinous cellular processes and offer a promising avenue for drug-target discovery. PPIs are characterized by multi-level complexity: at the protein level, interaction networks can be used to identify potential targets, whereas at the residue level, the details of the interactions of individual PPIs can be used to examine a target's druggability. Much great progress has been made in target discovery through multi-level PPI-related computational approaches, but these resources have not been fully discussed. Here, we systematically survey bioinformatics tools for identifying and assessing potential drug targets, examining their characteristics, limitations and applications. This work will aid the integration of the broader protein-to-network context with the analysis of detailed binding mechanisms to support the discovery of drug targets.

A bidirectional interpretable compound-protein interaction prediction framework based on cross attention

The identification of compound-protein interactions (CPIs) plays a vital role in drug discovery. However, the huge cost and labor-intensive nature in vitro and vivo experiments make it urgent for researchers to develop novel CPI prediction methods. Despite emerging deep learning methods have achieved promising performance in CPI prediction, they also face ongoing challenges: (i) providing bidirectional interpretability from both the chemical and biological perspective for the prediction results; (ii) comprehensively evaluating model generalization performance; (iii) demonstrating the practical applicability of these models. To overcome the challenges posed by current deep learning methods, we propose a cross multi-head attention oriented bidirectional interpretable CPI prediction model (CmhAttCPI). First, CmhAttCPI takes molecular graphs and protein sequences as inputs, utilizing the GCW module to learn atom features and the CNN module to learn residue features, respectively. Second, the model applies cross multi-head attention module to compute attention weights for atoms and residues. Finally, CmhAttCPI employs a fully connected neural network to predict scores for CPIs. We evaluated the performance of CmhAttCPI on balanced datasets and imbalanced datasets. The results consistently show that CmhAttCPI outperforms multiple state-of-the-art methods. We constructed three scenarios based on compound and protein clustering and comprehensively evaluated the model generalization ability within these scenarios. The results demonstrate that the generalization ability of CmhAttCPI surpasses that of other models. Besides, the visualizations of attention weights reveal that CmhAttCPI provides chemical and biological interpretation for CPI prediction. Moreover, case studies confirm the practical applicability of CmhAttCPI in discovering anticancer candidates.

A new paradigm for applying deep learning to protein-ligand interaction prediction

Protein-ligand interaction prediction presents a significant challenge in drug design. Numerous machine learning and deep learning (DL) models have been developed to accurately identify docking poses of ligands and active compounds against specific targets. However, current models often suffer from inadequate accuracy or lack practical physical significance in their scoring systems. In this research paper, we introduce IGModel, a novel approach that utilizes the geometric information of protein-ligand complexes as input for predicting the root mean square deviation of docking poses and the binding strength (pKd, the negative value of the logarithm of binding affinity) within the same prediction framework. This ensures that the output scores carry intuitive meaning. We extensively evaluate the performance of IGModel on various docking power test sets, including the CASF-2016 benchmark, PDBbind-CrossDocked-Core and DISCO set, consistently achieving state-of-the-art accuracies. Furthermore, we assess IGModel's generalizability and robustness by evaluating it on unbiased test sets and sets containing target structures generated by AlphaFold2. The exceptional performance of IGModel on these sets demonstrates its efficacy. Additionally, we visualize the latent space of protein-ligand interactions encoded by IGModel and conduct interpretability analysis, providing valuable insights. This study presents a novel framework for DL-based prediction of protein-ligand interactions, contributing to the advancement of this field. The IGModel is available at GitHub repository https://github.com/zchwang/IGModel.

Towards explainable interaction prediction: Embedding biological hierarchies into hyperbolic interaction space

Given the prolonged timelines and high costs associated with traditional approaches, accelerating drug development is crucial. Computational methods, particularly drug-target interaction prediction, have emerged as efficient tools, yet the explainability of machine learning models remains a challenge. Our work aims to provide more interpretable interaction prediction models using similarity-based prediction in a latent space aligned to biological hierarchies. We investigated integrating drug and protein hierarchies into a joint-embedding drug-target latent space via embedding regularization by conducting a comparative analysis between models employing traditional flat Euclidean vector spaces and those utilizing hyperbolic embeddings. Besides, we provided a latent space analysis as an example to show how we can gain visual insights into the trained model with the help of dimensionality reduction. Our results demonstrate that hierarchy regularization improves interpretability without compromising predictive performance. Furthermore, integrating hyperbolic embeddings, coupled with regularization, enhances the quality of the embedded hierarchy trees. Our approach enables a more informed and insightful application of interaction prediction models in drug discovery by constructing an interpretable hyperbolic latent space, simultaneously incorporating drug and target hierarchies and pairing them with available interaction information. Moreover, compatible with pairwise methods, the approach allows for additional transparency through existing explainable AI solutions.

Evaluating generalizability of artificial intelligence models for molecular datasets

Deep learning has made rapid advances in modeling molecular sequencing data. Despite achieving high performance on benchmarks, it remains unclear to what extent deep learning models learn general principles and generalize to previously unseen sequences. Benchmarks traditionally interrogate model generalizability by generating metadata based (MB) or sequence-similarity based (SB) train and test splits of input data before assessing model performance. Here, we show that this approach mischaracterizes model generalizability by failing to consider the full spectrum of cross-split overlap, i.e., similarity between train and test splits. We introduce Spectra, a spectral framework for comprehensive model evaluation. For a given model and input data, Spectra plots model performance as a function of decreasing cross-split overlap and reports the area under this curve as a measure of generalizability. We apply Spectra to 18 sequencing datasets with associated phenotypes ranging from antibiotic resistance in tuberculosis to protein-ligand binding to evaluate the generalizability of 19 state-of-the-art deep learning models, including large language models, graph neural networks, diffusion models, and convolutional neural networks. We show that SB and MB splits provide an incomplete assessment of model generalizability. With Spectra, we find as cross-split overlap decreases, deep learning models consistently exhibit a reduction in performance in a task- and model-dependent manner. Although no model consistently achieved the highest performance across all tasks, we show that deep learning models can generalize to previously unseen sequences on specific tasks. Spectra paves the way toward a better understanding of how foundation models generalize in biology.

Cracking the black box of deep sequence-based protein-protein interaction prediction

Identifying protein-protein interactions (PPIs) is crucial for deciphering biological pathways. Numerous prediction methods have been developed as cheap alternatives to biological experiments, reporting surprisingly high accuracy estimates. We systematically investigated how much reproducible deep learning models depend on data leakage, sequence similarities and node degree information, and compared them with basic machine learning models. We found that overlaps between training and test sets resulting from random splitting lead to strongly overestimated performances. In this setting, models learn solely from sequence similarities and node degrees. When data leakage is avoided by minimizing sequence similarities between training and test set, performances become random. Moreover, baseline models directly leveraging sequence similarity and network topology show good performances at a fraction of the computational cost. Thus, we advocate that any improvements should be reported relative to baseline methods in the future. Our findings suggest that predicting PPIs remains an unsolved task for proteins showing little sequence similarity to previously studied proteins, highlighting that further experimental research into the 'dark' protein interactome and better computational methods are needed.

Ligand Modification-Free Methods for the Profiling of Protein-Environmental Chemical Interactions

Adverse health outcomes caused by environmental chemicals are often initiated via their interactions with proteins. Essentially, one environmental chemical may interact with a number of proteins and/or a protein may interact with a multitude of environmental chemicals, forming an intricate interaction network. Omics-wide protein-environmental chemical interaction profiling (PECI) is of prominent importance for comprehensive understanding of these interaction networks, including the toxicity mechanisms of action (MoA), and for providing systematic chemical safety assessment. However, such information remains unknown for most environmental chemicals, partly due to their vast chemical diversity. In recent years, with the continuous efforts afforded, especially in mass spectrometry (MS) based omics technologies, several ligand modification-free methods have been developed, and new attention for systematic PECI profiling was gained. In this Review, we provide a comprehensive overview on these methodologies for the identification of ligand-protein interactions, including affinity interaction-based methods of affinity-driven purification, covalent modification profiling, and activity-based protein profiling (ABPP) in a competitive mode, physicochemical property changes assessment methods of ligand-directed nuclear magnetic resonance (ligand-directed NMR), MS integrated with equilibrium dialysis for the discovery of allostery systematically (MIDAS), thermal proteome profiling (TPP), limited proteolysis-coupled mass spectrometry (LiP-MS), stability of proteins from rates of oxidation (SPROX), and several intracellular downstream response characterization methods. We expect that the applications of these ligand modification-free technologies will drive a considerable increase in the number of PECI identified, facilitate unveiling the toxicological mechanisms, and ultimately contribute to systematic health risk assessment of environmental chemicals.

Advances in ligand-specific biosensing for structurally similar molecules

The specificity of biological systems makes it possible to develop biosensors targeting specific metabolites, toxins, and pollutants in complex medical or environmental samples without interference from structurally similar compounds. For the last two decades, great efforts have been devoted to creating proteins or nucleic acids with novel properties through synthetic biology strategies. Beyond augmenting biocatalytic activity, expanding target substrate scopes, and enhancing enzymes' enantioselectivity and stability, an increasing research area is the enhancement of molecular specificity for genetically encoded biosensors. Here, we summarize recent advances in the development of highly specific biosensor systems and their essential applications. First, we describe the rational design principles required to create libraries containing potential mutants with less promiscuity or better specificity. Next, we review the emerging high-throughput screening techniques to engineer biosensing specificity for the desired target. Finally, we examine the computer-aided evaluation and prediction methods to facilitate the construction of ligand-specific biosensors.

ZeroBind: a protein-specific zero-shot predictor with subgraph matching for drug-target interactions

Existing drug-target interaction (DTI) prediction methods generally fail to generalize well to novel (unseen) proteins and drugs. In this study, we propose a protein-specific meta-learning framework ZeroBind with subgraph matching for predicting protein-drug interactions from their structures. During the meta-training process, ZeroBind formulates training a protein-specific model, which is also considered a learning task, and each task uses graph neural networks (GNNs) to learn the protein graph embedding and the molecular graph embedding. Inspired by the fact that molecules bind to a binding pocket in proteins instead of the whole protein, ZeroBind introduces a weakly supervised subgraph information bottleneck (SIB) module to recognize the maximally informative and compressive subgraphs in protein graphs as potential binding pockets. In addition, ZeroBind trains the models of individual proteins as multiple tasks, whose importance is automatically learned with a task adaptive self-attention module to make final predictions. The results show that ZeroBind achieves superior performance on DTI prediction over existing methods, especially for those unseen proteins and drugs, and performs well after fine-tuning for those proteins or drugs with a few known binding partners.

Biomolecular NMR spectroscopy in the era of artificial intelligence

Biomolecular nuclear magnetic resonance (NMR) spectroscopy and artificial intelligence (AI) have a burgeoning synergy. Deep learning-based structural predictors have forever changed structural biology, yet these tools currently face limitations in accurately characterizing protein dynamics, allostery, and conformational heterogeneity. We begin by highlighting the unique abilities of biomolecular NMR spectroscopy to complement AI-based structural predictions toward addressing these knowledge gaps. We then highlight the direct integration of deep learning approaches into biomolecular NMR methods. AI-based tools can dramatically improve the acquisition and analysis of NMR spectra, enhancing the accuracy and reliability of NMR measurements, thus streamlining experimental processes. Additionally, deep learning enables the development of novel types of NMR experiments that were previously unattainable, expanding the scope and potential of biomolecular NMR spectroscopy. Ultimately, a combination of AI and NMR promises to further revolutionize structural biology on several levels, advance our understanding of complex biomolecular systems, and accelerate drug discovery efforts.

Artificial intelligence methods in kinase target profiling: Advances and challenges

Kinases have a crucial role in regulating almost the full range of cellular processes, making them essential targets for therapeutic interventions against various diseases. Accurate kinase-profiling prediction is vital for addressing the selectivity/specificity challenges in kinase drug discovery, which is closely related to lead optimization, drug repurposing, and the understanding of potential drug side effects. In this review, we provide an overview of the latest advancements in machine learning (ML)-based and deep learning (DL)-based quantitative structure-activity relationship (QSAR) models for kinase profiling. We highlight current trends in this rapidly evolving field and discuss the existing challenges and future directions regarding experimental data set construction and model architecture design. Our aim is to offer practical insights and guidance for the development and utilization of these approaches.

Pmf-cpi: assessing drug selectivity with a pretrained multi-functional model for compound-protein interactions

Compound-protein interactions (CPI) play significant roles in drug development. To avoid side effects, it is also crucial to evaluate drug selectivity when binding to different targets. However, most selectivity prediction models are constructed for specific targets with limited data. In this study, we present a pretrained multi-functional model for compound-protein interaction prediction (PMF-CPI) and fine-tune it to assess drug selectivity. This model uses recurrent neural networks to process the protein embedding based on the pretrained language model TAPE, extracts molecular information from a graph encoder, and produces the output from dense layers. PMF-CPI obtained the best performance compared to outstanding approaches on both the binding affinity regression and CPI classification tasks. Meanwhile, we apply the model to analyzing drug selectivity after fine-tuning it on three datasets related to specific targets, including human cytochrome P450s. The study shows that PMF-CPI can accurately predict different drug affinities or opposite interactions toward similar targets, recognizing selective drugs for precise therapeutics.Kindly confirm if corresponding authors affiliations are identified correctly and amend if any.Yes, it is correct.

A Robust Drug-Target Interaction Prediction Framework with Capsule Network and Transfer Learning

Drug-target interactions (DTIs) are considered a crucial component of drug design and drug discovery. To date, many computational methods were developed for drug-target interactions, but they are insufficiently informative for accurately predicting DTIs due to the lack of experimentally verified negative datasets, inaccurate molecular feature representation, and ineffective DTI classifiers. Therefore, we address the limitations of randomly selecting negative DTI data from unknown drug-target pairs by establishing two experimentally validated datasets and propose a capsule network-based framework called CapBM-DTI to capture hierarchical relationships of drugs and targets, which adopts pre-trained bidirectional encoder representations from transformers (BERT) for contextual sequence feature extraction from target proteins through transfer learning and the message-passing neural network (MPNN) for the 2-D graph feature extraction of compounds to accurately and robustly identify drug-target interactions. We compared the performance of CapBM-DTI with state-of-the-art methods using four experimentally validated DTI datasets of different sizes, including human (Homo sapiens) and worm (Caenorhabditis elegans) species datasets, as well as three subsets (new compounds, new proteins, and new pairs). Our results demonstrate that the proposed model achieved robust performance and powerful generalization ability in all experiments. The case study on treating COVID-19 demonstrates the applicability of the model in virtual screening.