Early Stages Of Drug Discovery Research Articles

MolModa: accessible and secure molecular docking in a web browser.

Molecular docking advances early-stage drug discovery by predicting the geometries and affinities of small-molecule compounds bound to drug-target receptors, predictions that researchers can leverage in prioritizing drug candidates for experimental testing. Unfortunately, existing docking tools often suffer from poor usability, data security, and maintainability, limiting broader adoption. Additionally, the complexity of the docking process, which requires users to execute a series of specialized steps, often poses a substantial barrier for non-expert users. Here, we introduce MolModa, a secure, accessible environment where users can perform molecular docking entirely in their web browsers. We provide two case studies that illustrate how MolModa provides valuable biological insights. We further compare MolModa to other docking tools to highlight its strengths and limitations. MolModa is available free of charge for academic and commercial use, without login or registration, at https://durrantlab.com/molmoda.

PARTITIONING BASED PHYSICOCHEMICAL MODELS FOR ASSESSING INTESTINAL PERMEABILITY AND ABSORPTION OF DRUGS

Oral administration of drugs is highly preferred for almost all human beings than any other route of drug delivery except during some health challenges. Therefore, permeability assessment of drugs across intestinal membrane is essential in the early stages of drug discovery for time and cost reasons. Animals, including humans, have been used for decades as in vivo models for determining intestinal drug permeability and absorption. However, in vivo models are very invasive, time-consuming, and not cost-effective methods. Numerous in vitro models have been used to screen drug permeability and absorption through intestinal membranes. In this article partitioning based physicochemical models that can predict a compound/drug permeability potential across intestinal membrane will be elaborated upon.

Machine learning framework to predict pharmacokinetic profile of small molecule drugs based on chemical structure.

Accurate prediction of a new compound's pharmacokinetic (PK) profile is pivotal for the success of drug discovery programs. An initial assessment of PK in preclinical species and humans is typically performed through allometric scaling and mathematical modeling. These methods use parameters estimated from invitro or invivo experiments, which although helpful for an initial estimation, require extensive animal experiments. Furthermore, mathematical models are limited by the mechanistic underpinning of the drugs' absorption, distribution, metabolism, and elimination (ADME) which are largely unknown in the early stages of drug discovery. In this work, we propose a novel methodology in which concentration versus time profile of small molecules in rats is directly predicted by machine learning (ML) using structure-driven molecular properties as input and thus mitigating the need for animal experimentation. The proposed framework initially predicts ADME properties based on molecular structure and then uses them as input to a ML model to predict the PK profile. For the compounds tested, our results demonstrate that PK profiles can be adequately predicted using the proposed algorithm, especially for compounds with Tanimoto score greater than 0.5, the average mean absolute percentage error between predicted PK profile and observed PK profile data was found to be less than 150%. The suggested framework aims to facilitate PK predictions and thus support molecular screening and design earlier in the drug discovery process.

Discovery of Strong 3-Nitro-2-Phenyl-2H-Chromene Analogues as Antitrypanosomal Agents and Inhibitors of Trypanosoma cruzi Glucokinase.

Chagas disease is one of the world's neglected tropical diseases, caused by the human pathogenic protozoan parasite Trypanosoma cruzi. There is currently a lack of effective and tolerable clinically available therapeutics to treat this life-threatening illness and the discovery of modern alternative options is an urgent matter. T. cruzi glucokinase (TcGlcK) is a potential drug target because its product, d-glucose-6-phosphate, serves as a key metabolite in the pentose phosphate pathway, glycolysis, and gluconeogenesis. In 2019, we identified a novel cluster of TcGlcK inhibitors that also exhibited anti-T. cruzi efficacy called the 3-nitro-2-phenyl-2H-chromene analogues. This was achieved by performing a target-based high-throughput screening (HTS) campaign of 13,040 compounds. The selection criteria were based on first determining which compounds strongly inhibited TcGlcK in a primary screen, followed by establishing on-target confirmed hits from a confirmatory assay. Compounds that exhibited notable in vitro trypanocidal activity over the T. cruzi infective form (trypomastigotes and intracellular amastigotes) co-cultured in NIH-3T3 mammalian host cells, as well as having revealed low NIH-3T3 cytotoxicity, were further considered. Compounds GLK2-003 and GLK2-004 were determined to inhibit TcGlcK quite well with IC50 values of 6.1 µM and 4.8 µM, respectively. Illuminated by these findings, we herein screened a small compound library consisting of thirteen commercially available 3-nitro-2-phenyl-2H-chromene analogues, two of which were GLK2-003 and GLK2-004 (compounds 1 and 9, respectively). Twelve of these compounds had a one-point change from the chemical structure of GLK2-003. The analogues were run through a similar primary screening and confirmatory assay protocol to our previous HTS campaign. Subsequently, three in vitro biological assays were performed where compounds were screened against (a) T. cruzi (Tulahuen strain) infective form co-cultured within NIH-3T3 cells, (b) T. brucei brucei (427 strain) bloodstream form, and (c) NIH-3T3 host cells alone. We report on the TcGlcK inhibitor constant determinations, mode of enzyme inhibition, in vitro antitrypanosomal IC50 determinations, and an assessment of structure-activity relationships. Our results reveal that the 3-nitro-2-phenyl-2H-chromene scaffold holds promise and can be further optimized for both Chagas disease and human African trypanosomiasis early-stage drug discovery research.

Abstract LB388: Super potent and efficacious ALK/ROS1/TRK inhibitor that overcomes multiple drug-resistant mutations

Abstract Abnormal activation of ALK, ROS1 and TRK is implicated in tumorigenesis, progression, invasion, and metastasis of malignancies through various mechanisms. The first generation TRK inhibitors including larotrectinib and entrectinib have limited applications due to acquired resistance in clinical settings. Multiple on-target mutations occur at the solvent-front, gatekeeper, and xDFG motif inside the ATP binding pocket of the TRK kinase domain. While some candidates under current clinical development aim to address the solvent-front mutation, they are ineffective against the xDFG mutations (e.g.NTRK1.G667X). There is strong unmet medical needs for next generation TRK inhibitors that can address resistance mutations. We employ the synergistic advantages of domain expertise and modern technology to accelerate early stage drug discovery. On our deep learning (DL) centric AI platform, extensive scientific computing supports large scale virtual screening, generative design, drug property prediction and multi-objective lead optimization. Only the optimized candidate compounds are passed along for synthesis and testing to save time and cost. In Ba/F3-TPM3 cellular assays, LB-2159 showed superior potencies that are hundreds to thousands of fold higher than larotrectinib and/or entrectinib, with IC50 values of 0.02 nM for NTRK1.WT, 0.3 nM for NTRK1.G595R and 0.88 nM for NTRK1.G667C. Similar cellular potencies were demonstrated for ALK and ROS1. In xenograft models LB-2159 at 10 mg/kg bid resulted in complete tumor regression in Ba/F3-TPM3-NTRK1-WT tumors, 96% TGI in Ba/F3-TPM3-NTRK1-G595R tumors, 84% TGI and 100% survival in mice bearing Ba/F3-TPM3-NTRK1-G667C tumors. The pharmacokinetic profiles of LB-2159 are overall favorable, demonstrating extended stability and half life. In vitro safety screens yielded clean profiles without any DDI signal or hERG liability. Preclinical safety evaluations exhibited good tolerance in rats. In multiple dose escalation studies LB-2159 was well tolerated at 50 mg/kg/day and beyond. Empowered by our proprietary AI platform, we have discovered the novel ALK/ROS1/TRK inhibitors that overcome clinically known resistance mutations. Our drug candidate exhibited good tolerance in animals in preclinical safety evaluations. In xenograft models that are resistant to current treatments the compound demonstrates strong tumor suppression efficacy. LB-2159 is advancing through IND enabling studies, and its preclinical profiles support the potential for best-in-class treatment for NTRK fusion-positive solid tumors. Citation Format: Li Xing, Jingbing Wang, Wei Zhu. Super potent and efficacious ALK/ROS1/TRK inhibitor that overcomes multiple drug-resistant mutations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB388.

Abstract LB389: Highly potent and selective RET inhibitor with robust anti-tumor activities against wild-type and acquired drug-resistant mutations

Abstract RET gene mutations and fusions are the driving events for the generation and deterioration of many cancer types. The first generation RET inhibitors, selpercatinib (Loxo-292) and pralsetinib (Blu-667), demonstrated clinically acquired resistance (e.g., solvent frontier mutations G810X, double mutations V804X-G810X etc.). Novel treatment that overcomes multiple resistance mutations is needed to improve patient outcomes. Leveraging our in-house AI technology platform, we have discovered the next generation RET inhibitors that overcome multiple resistance mutations including solvent front, gatekeeper and various double mutations. We employ the synergistic advantages of domain expertise and modern technology to accelerate early stage drug discovery. On our deep learning (DL) centric AI platform, extensive scientific computing supports large scale virtual screening, generative design, drug property prediction and multi-objective lead optimization. Only the optimized candidate compounds are passed along for synthesis and testing to save time and cost. In cell proliferation assays, our compound effectively inhibited KIF5B-RET Ba/F3 WT, V804M, G810S, G810R, V804M-G810S with IC50 values of 0.4 nM, 1.3 nM, 1.1 nM, 3.4 nM, 6.3 nM, respectively, demonstrating significant improvement over selpercatinib in the range of several hundred fold in activities. Importantly, the compound is highly selective against VEGFR2, an off-target responsible for cardiovascular adverse effects in the approved RET inhibitors. The superior cellular potencies translated consistently into animal models. In Ba/F3 KIF5B-RET-G810R xenograft model, treatment of 25 mg/kg bid resulted in >90% TGI with a100% survival rate. The compound showed favorable pharmacokinetic properties, lack of drug-drug interaction (DDI), and devoid of hERG liability (IC50 > 10uM). Preclinical safety evaluations in rats exhibited good tolerance margin. We have discovered the second generation compound that potently inhibits a diverse range of RET alterations. The compound exhibits preferable pharmaceutical profiles and is advancing toward IND application. Citation Format: Li Xing, Wei Zhu, Jingbing Wang. Highly potent and selective RET inhibitor with robust anti-tumor activities against wild-type and acquired drug-resistant mutations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB389.

MolProphet: A One-Stop, General Purpose, and AI-Based Platform for the Early Stages of Drug Discovery.

Artificial intelligence (AI) is an effective tool to accelerate drug discovery and cut costs in discovery processes. Many successful AI applications are reported in the early stages of small molecule drug discovery. However, most of those applications require a deep understanding of software and hardware, and focus on a single field that implies data normalization and transfer between those applications is still a challenge for normal users. It usually limits the application of AI in drug discovery. Here, based on a series of robust models, we formed a one-stop, general purpose, and AI-based drug discovery platform, MolProphet, to provide complete functionalities in the early stages of small molecule drug discovery, including AI-based target pocket prediction, hit discovery and lead optimization, and compound targeting, as well as abundant analyzing tools to check the results. MolProphet is an accessible and user-friendly web-based platform that is fully designed according to the practices in the drug discovery industry. The molecule screened, generated, or optimized by the MolProphet is purchasable and synthesizable at low cost but with good drug-likeness. More than 400 users from industry and academia have used MolProphet in their work. We hope this platform can provide a powerful solution to assist each normal researcher in drug design and related research areas. It is available for everyone at https://www.molprophet.com/.

Deciphering the Molecular Mechanisms of Reactive Metabolite Formation in the Mechanism-Based Inactivation of Cytochrome p450 1B1 by 8-Methoxypsoralen and Assessing the Driving Effect of phe268.

This study provides a comprehensive computational exploration of the inhibitory activity and metabolic pathways of 8-methoxypsoralen (8-MP), a furocoumarin derivative used for treating various skin disorders, on cytochrome P450 (P450). Employing quantum chemical DFT calculations, molecular docking, and molecular dynamics (MD) simulations analyses, the biotransformation mechanisms and the active site binding profile of 8-MP in CYP1B1 were investigated. Three plausible inactivation mechanisms were minutely scrutinized. Further analysis explored the formation of reactive metabolites in subsequent P450 metabolic processes, including covalent adduct formation through nucleophilic addition to the epoxide, 8-MP epoxide hydrolysis, and non-CYP-catalyzed epoxide ring opening. Special attention was paid to the catalytic effect of residue Phe268 on the mechanism-based inactivation (MBI) of P450 by 8-MP. Energetic profiles and facilitating conditions revealed a slight preference for the C4'=C5' epoxidation pathway, while recognizing a potential kinetic competition with the 8-OMe demethylation pathway due to comparable energy demands. The formation of covalent adducts via nucleophilic addition, particularly by phenylalanine, and the generation of potentially harmful reactive metabolites through autocatalyzed ring cleavage are likely to contribute significantly to P450 metabolism of 8-MP. Our findings highlight the key role of Phe268 in retaining 8-MP within the active site of CYP1B1, thereby facilitating initial oxygen addition transition states. This research offers crucial molecular-level insights that may guide the early stages of drug discovery and risk assessment related to the use of 8-MP.

AI-guided pipeline for protein–protein interaction drug discovery identifies a SARS-CoV-2 inhibitor

Protein–protein interactions (PPIs) offer great opportunities to expand the druggable proteome and therapeutically tackle various diseases, but remain challenging targets for drug discovery. Here, we provide a comprehensive pipeline that combines experimental and computational tools to identify and validate PPI targets and perform early-stage drug discovery. We have developed a machine learning approach that prioritizes interactions by analyzing quantitative data from binary PPI assays or AlphaFold-Multimer predictions. Using the quantitative assay LuTHy together with our machine learning algorithm, we identified high-confidence interactions among SARS-CoV-2 proteins for which we predicted three-dimensional structures using AlphaFold-Multimer. We employed VirtualFlow to target the contact interface of the NSP10-NSP16 SARS-CoV-2 methyltransferase complex by ultra-large virtual drug screening. Thereby, we identified a compound that binds to NSP10 and inhibits its interaction with NSP16, while also disrupting the methyltransferase activity of the complex, and SARS-CoV-2 replication. Overall, this pipeline will help to prioritize PPI targets to accelerate the discovery of early-stage drug candidates targeting protein complexes and pathways.

Expanding FTMap for Fragment-Based Identification of Pharmacophore Regions in Ligand Binding Sites.

The knowledge of ligand binding hot spots and of the important interactions within such hot spots is crucial for the design of lead compounds in the early stages of structure-based drug discovery. The computational solvent mapping server FTMap can reliably identify binding hot spots as consensus clusters, free energy minima that bind a variety of organic probe molecules. However, in its current implementation, FTMap provides limited information on regions within the hot spots that tend to interact with specific pharmacophoric features of potential ligands. E-FTMap is a new server that expands on the original FTMap protocol. E-FTMap uses 119 organic probes, rather than the 16 in the original FTMap, to exhaustively map binding sites, and identifies pharmacophore features as atomic consensus sites where similar chemical groups bind. We validate E-FTMap against a set of 109 experimentally derived structures of fragment-lead pairs, finding that highly ranked pharmacophore features overlap with the corresponding atoms in both fragments and lead compounds. Additionally, comparisons of mapping results to ensembles of bound ligands reveal that pharmacophores generated with E-FTMap tend to sample highly conserved protein-ligand interactions. E-FTMap is available as a web server at https://eftmap.bu.edu.

Protein lipidation in cancer: mechanisms, dysregulation and emerging drug targets.

Protein lipidation describes a diverse class of post-translational modifications (PTMs) that is regulated by over 40 enzymes, targeting more than 1,000 substrates at over 3,000 sites. Lipidated proteins include more than 150 oncoproteins, including mediators of cancer initiation, progression and immunity, receptor kinases, transcription factors, G protein-coupled receptors and extracellular signalling proteins. Lipidation regulates the physical interactions of its protein substrates with cell membranes, regulating protein signalling and trafficking, and has a key role in metabolism and immunity. Targeting protein lipidation, therefore, offers a unique approach to modulate otherwise undruggable oncoproteins; however, the full spectrum of opportunities to target the dysregulation of these PTMs in cancer remains to be explored. This is attributable in part to the technological challenges of identifying the targets and the roles of protein lipidation. The early stage of drug discovery for many enzymes in the pathway contrasts with efforts for drugging similarly common PTMs such as phosphorylation and acetylation, which are routinely studied and targeted in relevant cancer contexts. Here, we review recent advances in identifying targetable protein lipidation pathways in cancer, the current state-of-the-art in drug discovery, and the status of ongoing clinical trials, which have the potential to deliver novel oncology therapeutics targeting protein lipidation.

Predicting pharmacodynamic effects through early drug discovery with artificial intelligence-physiologically based pharmacokinetic (AI-PBPK) modelling.

A mechanism-based pharmacokinetic/pharmacodynamic (PK/PD) model links the concentration-time profile of a drug with its therapeutic effects based on the underlying biological or physiological processes. Clinical endpoints play a pivotal role in drug development. Despite the substantial time and effort invested in screening drugs for favourable pharmacokinetic (PK) properties, they may not consistently yield optimal clinical outcomes. Furthermore, in the virtual compound screening phase, researchers cannot observe clinical outcomes in humans directly. These uncertainties prolong the process of drug development. As incorporation of Artificial Intelligence (AI) into the physiologically based pharmacokinetic/pharmacodynamic (PBPK) model can assist in forecasting pharmacodynamic (PD) effects within the human body, we introduce a methodology for utilizing the AI-PBPK platform to predict the PK and PD outcomes of target compounds in the early drug discovery stage. In this integrated platform, machine learning is used to predict the parameters for the model, and the mechanism-based PD model is used to predict the PD outcome through the PK results. This platform enables researchers to align the PK profile of a drug with desired PD effects at the early drug discovery stage. Case studies are presented to assess and compare five potassium-competitive acid blocker (P-CAB) compounds, after calibration and verification using vonoprazan and revaprazan.

Biocatalytic strategy for the construction of sp3-rich polycyclic compounds from directed evolution and computational modelling.

Catalysis with engineered enzymes has provided more efficient routes for the production of active pharmaceutical agents. However, the potential of biocatalysis to assist in early-stage drug discovery campaigns remains largely untapped. In this study, we have developed a biocatalytic strategy for the construction of sp3-rich polycyclic compounds via the intramolecular cyclopropanation of benzothiophenes and related heterocycles. Two carbene transferases with complementary regioisomer selectivity were evolved to catalyse the stereoselective cyclization of benzothiophene substrates bearing diazo ester groups at the C2 or C3 position of the heterocycle. The detailed mechanisms of these reactions were elucidated by a combination of crystallographic and computational analyses. Leveraging these insights, the substrate scope of one of the biocatalysts could be expanded to include previously unreactive substrates, highlighting the value of integrating evolutionary and rational strategies to develop enzymes for new-to-nature transformations. The molecular scaffolds accessed here feature a combination of three-dimensional and stereochemical complexity with 'rule-of-three' properties, which should make them highly valuable for fragment-based drug discovery campaigns.

Development of a high-throughput screening platform to identify new therapeutic agents for Medulloblastoma Group 3

Pediatric brain tumors (PBTs) represent about 25 % of all pediatric cancers and are the most common solid tumors in children and adolescents. Medulloblastoma (MB) is the most frequently occurring malignant PBT, accounting for almost 10 % of all pediatric cancer deaths. MB Group 3 (MB G3) accounts for 25–30 % of all MB cases and has the worst outcome, particularly when associated with MYC amplification. However, no targeted treatments for this group have been developed so far.Here we describe a unique high throughput screening (HTS) platform specifically designed to identify new therapies for MB G3. The platform incorporates optimized and validated 2D and 3D efficacy and toxicity models, that account for tumor heterogenicity, limited efficacy and unacceptable toxicity from the very early stage of drug discovery. The platform has been validated by conducting a pilot HTS campaign with a 1280 lead-like compound library. Results showed 8 active compounds, targeting MB reported targets and several are currently approved or in clinical trials for pediatric patients with PBTs, including MB. Moreover, hits were combined to avoid tumor resistance, identifying 3 synergistic pairs, one of which is currently under clinical study for recurrent MB and other PBTs.

In silico prediction of ocular toxicity of compounds using explainable machine learning and deep learning approaches.

The accurate identification of chemicals with ocular toxicity is of paramount importance in health hazard assessment. In contemporary chemical toxicology, there is a growing emphasis on refining, reducing, and replacing animal testing in safety evaluations. Therefore, the development of robust computational tools is crucial for regulatory applications. The performance of predictive models is heavily reliant on the quality and quantity of data. In this investigation, we amalgamated the most extensive dataset (4901 compounds) sourced from governmental GHS-compliant databases and literature to develop binary classification models of chemical ocular toxicity. We employed 12 molecular representations in conjunction with six machine learning algorithms and two deep learning algorithms to create a series of binary classification models. The findings indicated that the deep learning method GCN outperformed the machine learning models in cross-validation, achieving an impressive AUC of 0.915. However, the top-performing machine learning model (RF-Descriptor) demonstrated excellent performance with an AUC of 0.869 on the test set and was therefore selected as the best model. To enhance model interpretability, we conducted the SHAP method and attention weights analysis. The two approaches offered visual depictions of the relevance of key descriptors and substructures in predicting ocular toxicity of chemicals. Thus, we successfully struck a delicate balance between data quality and model interpretability, rendering our model valuable for predicting and comprehending potential ocular-toxic compounds in the early stages of drug discovery.

A Comprehensive Review of the Advancement in Omic Technologies in the Field of Drug Discovery and Development

Abstract: Discovering new drugs is time-consuming and expensive and involves many different tools from various domains. Numerous omic technologies, such as genomics, transcriptomics, proteomics, and metabolomics, have been created to speed up the process. Leveraging genetic and genomic insights, these methodologies play a pivotal role. Genetic insights aid in target identification, prioritization, and the prediction of drug outcomes. Gene expression data informs drug discovery, while proteomics uncovers targets and facilitates high-throughput profiling. Enhancing drug efficacy necessitates mechanistic insights into downstream effects, enabling side effects and resistance prediction. Early-stage drug discovery now extensively employs diverse metabolomics platforms. This review underscores the recent strides of omic technologies in drug discovery, affirming their role in enhancing drug viability and regulatory approval. The emphasis lies on the latest advancements in genomics, transcriptomics, proteomics, and metabolomics, collectively fortifying drug development.

High-throughput assay to simultaneously evaluate activation of CYP3A and the direct and time-dependent inhibition of CYP3A, CYP2C9, and CYP2D6 using liquid chromatography-tandem mass spectrometry

In the early stages of drug discovery, adequate evaluation of the potential drug–drug interactions (DDIs) of drug candidates is important. Several CYP3A activators are known to lead to underestimation of DDIs. These compounds affect midazolam 1'-hydroxylation but not midazolam 4-hydroxylation. We used both metabolic reactions of midazolam to evaluate the activation and inhibition of CYP3A activators simultaneously. For our CYP inhibition assay using cocktail probe substrates, simultaneous liquid chromatography-tandem mass spectrometry monitoring of 1'-hydroxymidazolam and 4-hydroxymidazolam for CYP3A was established in addition to monitoring of 4-hydroxydiclofenac and 1'-hydroxybufuralol for CYP2C9 and CYP2D6. The results of our cocktail inhibition assay were well correlated with those of a single inhibition assay, as were the estimated inhibition parameters for typical CYP3A inhibitors. In our assay, a proprietary compound that activated midazolam 1'-hydroxylation and tended to inhibit 4-hydroxylation was evaluated along with known CYP3A activators. All compounds were well characterised by comparison of the results of midazolam 1'- and 4-hydroxylation. In conclusion, our CYP cocktail inhibition assay can detect CYP3A activation and assess the direct and time-dependent inhibition potentials for CYP3A, CYP2C9, and CYP2D6. This method is expected to be very efficient in the early stages of drug discovery.

Developing a GNN-based AI model to predict mitochondrial toxicity using the bagging method.

Mitochondrial toxicity has been implicated in the development of various toxicities, including hepatotoxicity. Therefore, mitochondrial toxicity has become a major screening factor in the early discovery phase of drug development. Several models have been developed to predict mitochondrial toxicity based on chemical structures. However, they only provide a binary classification of positive or negative results and do not provide the substructures that contribute to a positive decision. Therefore, we developed an artificial intelligence (AI) model to predict mitochondrial toxicity and visualize structural alerts. To construct the model, we used the open-source software library kMoL, which employs a graph neural network approach that allows learning from chemical structure data. We also utilized the integrated gradient method, which enables the visualization of substructures that contribute to positive results. The dataset used to construct the AI model exhibited a significant imbalance, with significantly more negative than positive data. To address this, we employed the bagging method, which resulted in a model with high predictive performance, as evidenced by an F1 score of 0.839. This model can also be used to visualize substructures that contribute to mitochondrial toxicity using the integrated gradient method. Our AI model predicts mitochondrial toxicity based on chemical structures and may contribute to screening mitochondrial toxicity in the early stages of drug discovery.

Recent developments and applications of ambient mass spectrometry imaging in pharmaceutical research: an overview.

The application of ambient mass spectrometry imaging "MSI" is expanding in the areas of fundamental research on drug delivery and multiple phases of the process of identifying and developing drugs. Precise monitoring of a drug's pharmacological workflows, such as intake, distribution, metabolism, and discharge, is made easier by MSI's ability to determine the concentrations of the initiating drug and its metabolites across dosed samples without losing spatial data. Lipids, glycans, and proteins are just a few of the many phenotypes that MSI may be used to concurrently examine. Each of these substances has a particular distribution pattern and biological function throughout the body. MSI offers the perfect analytical tool for examining a drug's pharmacological features, especially in vitro and in vivo effectiveness, security, probable toxic effects, and putative molecular pathways, because of its high responsiveness in chemical and physical environments. The utilization of MSI in the field of pharmacy has further extended from the traditional tissue examination to the early stages of drug discovery and development, including examining the structure-function connection, high-throughput capabilities in vitro examination, and ex vivo research on individual cells or tumor spheroids. Additionally, an enormous array of endogenous substances that may function as tissue diagnostics can be scanned simultaneously, giving the specimen a highly thorough characterization. Ambient MSI techniques are soft enough to allow for easy examination of the native sample to gather data on exterior chemical compositions. This paper provides a scientific and methodological overview of ambient MSI utilization in research on pharmaceuticals.

GraphCL-DTA: A Graph Contrastive Learning With Molecular Semantics for Drug-Target Binding Affinity Prediction.

Drug-target binding affinity prediction plays an important role in the early stages of drug discovery, which can infer the strength of interactions between new drugs and new targets. However, the performance of previous computational models is limited by the following drawbacks. The learning of drug representation relies only on supervised data without considering the information in the molecular graph itself. Moreover, most previous studies tended to design complicated representation learning modules, while uniformity used to measure representation quality is ignored. In this study, we propose GraphCL-DTA, a graph contrastive learning with molecular semantics for drug-target binding affinity prediction. This graph contrastive learning framework replaces the dropout-based data augmentation strategy by performing data augmentation in the embedding space, thereby better preserving the semantic information of the molecular graph. A more essential and effective drug representation can be learned through this graph contrastive framework without additional supervised data. Next, we design a new loss function that can be directly used to adjust the uniformity of drug and target representations. By directly optimizing the uniformity of representations, the representation quality of drugs and targets can be improved. The effectiveness of the above innovative elements is verified on two real datasets, KIBA and Davis. Compared with the GraphDTA model, the relative improvement of the GraphCL-DTA model on the two datasets is 2.7% and 4.5%. The graph contrastive learning framework and uniformity function in the GraphCL-DTA model can be embedded into other computational models as independent modules to improve their generalization capability.