Computational Drug Discovery Methods

Designing High-Affinity Progesterone Binders: Pocket Analysis and Scaffold Selection

This study uses in-silico docking to design high-affinity progesterone binders by analyzing binding pockets and selecting scaffolds for biosensor development.

Large-scale quantum computing framework enhances drug discovery in multiple stages

This paper presents a quantum computing framework for drug discovery that improves stability and performance in multiple stages, identifying novel druggable sites and compounds.

Large Language Model Agent for Modular Task Execution in Drug Discovery

AgentD, a GPT-4o agent wired to retrieval, generative, ADMET and Boltz-2 tools, chains early drug-discovery steps for BCL-2 and raised molecules with QED > 0.6 from 34 to 55.

A Computational Workflow for Structure-Guided Design of Potent and Selective Kinase Peptide Substrates

This paper introduces Subtimizer, a computational pipeline for designing potent and selective kinase peptide substrates using AI tools, showing improved activity and selectivity in tested kinases.

Assessing Conformation Validity and Rationality of Deep Learning-Generated 3D Molecules

This paper introduces a two-stage method to evaluate 3D molecule conformations generated by AI, achieving high speed and quantum mechanical accuracy.

Ribociclib as a Potential Multi-Target Inhibitor of Pro-Inflammatory Cytokines: An In Silico Investigation

This paper investigates Ribociclib's interactions with pro-inflammatory cytokines using computational methods, finding it forms stable complexes with several cytokines and may act as a multi-target anti-inflammatory agent.

Model-Driven Hybrid AI Framework for End-to-End Autonomous Decision-Making in Drug Development

This paper introduces a hybrid AI framework for drug development that links clinical questions to pharmacokinetic simulations and decision support using PICO specifications and machine learning.

Macro-Equi-Diff (MED): Scaffold-based Macrocycles Generation Using Equivariant Diffusion

The paper introduces MED, a deep learning framework for generating macrocycles from acyclic molecules, achieving high validity and binding affinity in drug discovery.

Cellular Context Influences Kinase Inhibitor Selectivity

The paper compares cell-free and cellular assays for kinase inhibitor selectivity, finding divergent results and identifying new kinase interactions in cells.

Hierarchical Representation Learning for Drug Mechanism-of-Action Prediction from Gene Expression Data

This paper introduces a hierarchical representation learning framework for drug mechanism-of-action prediction from gene expression data, achieving improved F1 performance and interpretable latent space capturing MoA-level separation and compound-level substructure.

ABFormer: A Transformer-based Model to Enhance Antibody-Drug Conjugates Activity Prediction through Contextualized Antibody-Antigen Embedding

ABFormer improves ADC activity prediction by using contextualized antibody-antigen embeddings, outperforming baselines with 100% accuracy on a test set.

Beyond Lipschitz: Ranking Binding Affinity in Hyperbolic Space

This paper introduces AlphaRank, a deep learning model that ranks protein-ligand binding affinity using hyperbolic geometry to overcome limitations of Euclidean space models.

Fragment Screening and Structure-Guided Development of Heparanase Inhibitors Reveals Orthosteric and Allosteric Inhibition

This study uses fragment-based drug design to identify novel heparanase inhibitors, resulting in compounds with improved potency for potential therapeutic applications.

BEACON: predicting side effects and therapeutics outcomes to drugs by Bridging knowlEdge grAph with CONtextual language model

BEACON predicts drug side effects and therapeutic outcomes by integrating biomedical knowledge graphs with contextual language models, achieving high accuracy in drug sensitivity and drug-drug interaction predictions.

Chemical Genomics Language Model toward Reliable and Explainable Compound-Protein Interaction Exploration

This paper introduces ChemGLaM, a chemical genomics language model for reliable and explainable compound-protein interaction predictions, achieving state-of-the-art performance and enabling drug/target exploration for ALS.

ArtiDock: accurate Machine Learning approach to protein-ligand docking optimized for high-throughput virtual screening

ArtiDock is a machine learning-based protein-ligand docking method that improves accuracy over classical techniques while maintaining low computational cost for high-throughput virtual screening.

Deciphering the Therapeutic Potential of Resveratrol Against Pancreatic Cancer Through Network Pharmacology

This study identifies resveratrol's therapeutic targets in pancreatic cancer using network pharmacology and molecular docking, revealing key pathways and hub genes involved in cancer progression.

Fragment-Guided New Therapeutic Molecule Discovery and Mapping of Clinically Relevant Interactomes

This study uses a self-adjusting neural network to discover new therapeutic molecules and map interactomes for JAK2 and GLP-1R, revealing insights into drug-target binding and metabolic trajectories.

Selector: A General Python Library for Diverse Subset Selection

Selector is a Python library for subset selection that provides diverse sampling algorithms and integrates with data analysis tools, with applications in computational chemistry and drug discovery.

Mechanistic Language Modeling and Oxygenated 3D Screening Reveal Berberine and Enzalutamide Synergy in Resistant Prostate Cancer

Explainable Artificial Intelligence Reveals Potential Candidate Mechanism of Strain-Specific Drug Depletion

This paper uses AI to predict strain-specific drug depletion by analyzing drug properties and bacterial genomes, identifying potential enzymes involved in drug metabolism.

A chemical space model for the exploration of eco-toxicological data

This paper applies TMAP to ecotoxicological data to explore chemical modes of action and identify potential hazards in data-poor chemicals.

PCPpred: Prediction of Chemically Modified Peptide Permeability Across Multiple Assays for Oral Delivery

The paper develops PCPpred, a model predicting peptide permeability across assays, achieving high performance on Caco-2 and PAMPA tests using molecular features and ensemble methods.

One-Hot News: Drug Synergy Models Shortcut Molecular Features

This paper shows that using one-hot encodings for drugs and cell lines in synergy prediction models can perform as well as or better than using molecular features, suggesting models rely more on identifiers than features.

BioChemAIgent: An AI-driven Protein Modeling and Docking Framework for Structure-Based Drug Discovery

BioChemAIgent provides an AI-driven framework for protein modeling and docking in drug discovery, integrating multiple tools into a unified interface.

De novo protein ligand design including protein flexibility and conformational adaptation

This paper introduces AI-MCLig, a method for designing protein ligands that accounts for protein flexibility and conformational adaptation, achieving binding scores comparable to known binders.

Identification of Molecular Compounds Targeting Bacterial Propionate Metabolism with Topological Machine Learning

This paper uses topological machine learning to identify 15 compounds that inhibit methylcitrate dehydratase, a key enzyme in bacterial propionate metabolism, and validates their interactions through molecular docking.

Physical-Chemical Approach to Identify Local Structural Determinants of Molecular Mechanisms: Case Study of Antimalarial Drug Pyronaridine and Crystal-Growth Inhibition

The paper develops a physical-chemical framework to identify fragment-level structural determinants of molecular mechanisms, applying it to pyronaridine's crystal-growth inhibition and antimalarial activity.

BOLD-GPCRs: A Transformer-Powered App for Predicting Ligand Bioactivity and Mutational Effects Across Class A GPCRs

BOLD-GPCRs predicts ligand bioactivity and mutational effects for class A GPCRs using a transformer-based deep learning framework, achieving robust performance across understudied receptors.

A Case-Based Explainable Graph Neural Network Framework for Mechanistic Drug Repositioning

This paper introduces DBR-X, an explainable GNN framework for drug repositioning that outperforms existing models in predicting drug-disease associations and provides interpretable biological explanations.