Computational Drug Discovery Methods

PhysDock: A Physics-Guided All-Atom Diffusion Model for Protein-Ligand Complex Prediction

PhysDock predicts protein-ligand complexes using a physics-guided diffusion model, achieving state-of-the-art performance in redocking and cross-docking benchmarks.

Small molecule bioactivity benchmarks are often well-predicted by counting cells

The paper shows that cell counting can predict compound bioactivity as well as Cell Painting or gene expression, but Cell Painting adds unique value in protein-target-specific benchmarks.

A database for large-scale docking and experimental results

The paper develops a database of docking results and experimental data to improve benchmarking and chemical space exploration.

A thermodynamic cycle to predict the competitive inhibition outcomes of an evolving enzyme

This study uses thermodynamic cycles and FEP simulations to predict the effects of resistance mutations on DHFR's competitive inhibition with TMP, revealing structural and dynamical impacts on enzyme function.

In silico discovery of potential inhibitors targeting the MEIG1-PACRG complex for male contraceptive development

This study identifies small molecule inhibitors targeting the MEIG1-PACRG complex using virtual screening, docking, and MD simulations, highlighting distinct binding determinants for each protein.

Attention-Based Solution for Synergistic Virus Combination Therapy

The paper predicts synergistic drug combinations for viral diseases using attention-based models and validates two combinations against herpes simplex virus type 1.

INSIGHT: In Silico Drug Screening Platform using Interpretable Deep Learning Network

The paper introduces INSIGHT, a deep learning framework that predicts drug response using multiomics data and drug molecular structure, outperforming existing methods and enabling biological interpretation.

SieveAI: Development of an Automated extensible and customisable drug discovery pipeline and its validation

SieveAI automates molecular docking workflows for drug discovery, reducing manual intervention and integrating multiple open-source tools.

TARRAGON: Therapeutic Target Applicability Ranking and Retrieval-Augmented Generation Over Networks

TARRAGON prioritizes and validates therapeutic targets for non-muscle invasive urinary bladder cancer using a computational framework integrating knowledge graphs and large language models.

Discovery of Novel Trypanothione Reductase Inhibitors through Pharmacophore Modeling, virtual Screening and Molecular Dynamics Simulations. New insights for Human African Trypanosomiasis

The paper identifies two novel inhibitors of Trypanothione Reductase through pharmacophore modeling, virtual screening, and molecular dynamics simulations, showing better efficacy than Melarsoprol.

Drug repurposing reveals Posaconazole as a CYP11A1 inhibitor enhancing anti-tumour immunity

The paper identifies Posaconazole as a CYP11A1 inhibitor that enhances anti-tumour immunity in a mouse model of lung metastasis.

MolSnapper: Conditioning Diffusion for Structure Based Drug Design

MolSnapper improves structure-based drug design by conditioning diffusion models with 3D pharmacophores, generating more valid molecules with higher structural and chemical similarity.

Similarity of drug targets to human microbiome metaproteome promotes pharmacological promiscuity

The paper assesses similarity between drug targets and human microbiome metaproteomes to identify potential off-target effects and pharmacological promiscuity.

DTMol: Pocket-based Molecular Docking using Diffusion Transformers

DTMol is a diffusion model for pocket-based molecular docking that outperforms existing methods on benchmark datasets and identifies new JAK2 inhibitors.

An analysis of free energy methods to quantify and displace the KRasG12D Thr58 water

The paper evaluates free energy methods for quantifying and displacing a conserved water molecule in KRasG12D inhibitors.

Computational Evaluation of Phytochemicals as Potential Anti-HIV Drugs Targeting CCR5 and CXCR4 Receptors

This study computationally evaluates 53 phytochemicals targeting CCR5 and CXCR4 receptors for anti-HIV potential, identifying six with favorable druglikeness and ADMET properties.

OBE-DB: A Computational Tool and Web Server for the Prediction of Antiobesity Drugs

OBE-DB predicts anti-obesity drugs using shape similarity and inverse virtual screening, offering a user-friendly tool that outperforms conventional methods in accuracy and usability.

AlphaFold3 in Drug Discovery: A Comprehensive Assessment of Capabilities, Limitations, and Applications

This study evaluates AlphaFold3's performance in drug discovery tasks, finding it excels at static protein-ligand interactions but struggles with conformational changes, GPCR bias, and affinity ranking.

A Folding-Docking-Affinity framework for protein-ligand binding affinity prediction

This study introduces a framework that predicts protein-ligand binding affinity by folding proteins, determining binding conformations, and using 3D structures, achieving performance comparable to existing docking-free methods.

Learning Universal Representations of Intermolecular Interactions with ATOMICA

ATOMICA learns transferable representations of intermolecular interactions across diverse molecular types, enabling cross-modal comparisons and experimentally testable hypotheses.

Chemical representation standardization needed to generalize metabolic pathway involvement prediction across the Kyoto Encyclopedia of Genes and Genomes, Reactome, and MetaCyc knowledgebases

The paper trains a model to predict compound pathway involvement across KEGG, Reactome, and MetaCyc using standardized chemical representations, achieving high accuracy.

Deep Modeling of Gain of Function Mutations on Androgen Receptor

This study uses deep modeling to predict gain-of-function mutations in the androgen receptor and their drug responses, achieving 80% accuracy in identifying agonist/antagonist relationships.

Fleming: An AI Agent for Antibiotic Discovery in Mycobacterium Tuberculosis

Fleming is an AI agent that identifies lead compounds for Mycobacterium tuberculosis by combining discriminative and generative models, achieving higher discrimination and hit rates than alternative approaches.

Attention Guided Mechanism Interpretable Drug-Gene Interaction (MIDI) Modeling for Cancer Drug Response Prediction and Target Effect Explanation

This paper introduces MIDI, a model that predicts cancer drug response and explains drug-target interactions using an attention mechanism, outperforming existing methods in prediction and interpretability.

NeSyDPP4-QSAR: A Neuro-Symbolic AI Approach for Potent DPP-4-Inhibitor Discovery in Diabetes Treatment

This paper develops a neuro-symbolic AI model, NeSyDPP4-QSAR, to predict DPP-4 inhibitor activity, achieving high accuracy in IC50 prediction for drug discovery.

Computer-aided drug design and pharmacophore modeling towards the discovery of novel anti-Ebola agents

This study uses computational methods to design novel anti-Ebola agents targeting the virus polymerase.

Employing Steered MD Simulations for Effective Virtual Screening: Active Pharmacophore Search by Dynamic Corrections to target MKK3-MYC Interactions

The paper uses steered MD simulations to screen small molecules targeting MKK3-MYC interactions, identifying compounds with promising binding affinities.

AlphaFold3 and RoseTTAFold All-Atom structures enable radiosensitizers discovery by targeting multiple DNA damage repair proteins

The paper identifies radiosensitizers by docking natural products against AI-predicted and experimental DNA repair protein structures, finding compounds that effectively target tumor radioresistance.

Comparative study of protein X-ray and NMR structures: molecular docking-based virtual screening

This study compares X-ray and NMR protein structures in virtual ligand screening, finding X-ray structures generally perform better in ligand enrichment and binding-site configuration.

A genetics-guided integrative framework for drug repurposing: identifying antihypertensive drug telmisartan for type 2 diabetes

This study develops a genetics-informed framework to repurpose drugs for type 2 diabetes, identifying telmisartan as a potential treatment based on GWAS data and EHR validation.