Computational Drug Discovery Methods

vToxiNet: a biologically constrained deep learning framework for interpretable prediction of drug-induced hepatotoxicity

vToxiNet predicts drug-induced hepatotoxicity using a biologically constrained deep learning framework, achieving robust performance and mechanistic interpretability.

GTA-5: A Unified Graph Transformer Framework for Ligands and Protein Binding Sites - Part I: Constructing the PDB Pocket and Ligand Space

GTA-5 embeds protein binding pockets and ligands into latent spaces to enable structural reasoning, showing functional clustering and physicochemical property capture.

Artemis: Harnessing Knowledge Graphs for Next-Generation Drug Target Prioritization

Artemis integrates public biomedical knowledge graphs with clinical data to prioritize drug targets, showing effectiveness in breast cancer applications.

Generative Chemistry Platform for Small Molecules Targeting RNA: A Case Study for Chemical Optimization

The paper introduces a generative chemistry platform for designing RNA-targeting small molecules and compares its performance to existing models and human-designed chemistry.

Molecular dynamics simulations demonstrate reduced antibiotic affinity to mirror bacterial targets

The paper uses molecular dynamics simulations to assess antibiotic binding to natural and mirror bacterial targets, finding reduced affinity for mirror targets.

The redesign of the molecular scaffold of viral ion channel blockers

This study redesigns molecular scaffolds to create more effective viral ion channel blockers by optimizing hydrophobic and hydrophilic patterns and simplifying structures.

IRIS: A Machine Learning-Based Pose Re-Ranking Tool for RNA-Ligand Docking

IRIS improves RNA-ligand docking pose ranking by integrating a regression model trained on experimental complexes, increasing accuracy in selecting near-native poses.

Critical Assessment of ML models for ADMET Prediction in TDC leaderboards

The paper evaluates ADMET prediction models on TDC leaderboards, finding most top models lack reproducibility and exhibit data leakage, with only three methods passing rigorous checks.

MAP: A Knowledge-driven Framework for Predicting Single-cell Responses for Unprofiled Drugs

MAP predicts single-cell responses to unprofiled drugs using a knowledge-driven framework that integrates biological knowledge and improves performance over baselines.

scDock: Streamlining drug discovery targeting cell-cell communication via scRNA-seq analy-sis and molecular docking

scDock connects scRNA-seq analysis with molecular docking to identify drug candidates targeting cell-cell communication.

Transcriptome-based lead generation, ligand- and structure-based prioritization and experimental validation of TLR5-activating molecules

This paper generates TLR5-activating molecules using a systems-level approach combining transcriptome data and computational methods, validated experimentally.

GARD: Genomic Data based Drug Repurposing in Head and Neck Cancer with Large Language Model Validation

The paper develops GARD, a pipeline using genomic data and large language models to identify drug repurposing candidates for head and neck cancer, finding both known and emerging therapeutic options.

Domain-adaptation deep learning models do not outperform simple baseline models in single-cell anti-cancer drug sensitivity prediction

This study shows that complex domain adaptation models do not outperform simple gradient boosting models in predicting single-cell anti-cancer drug sensitivity.

Geometric-Evolutionary Deep Learning Decodes the Human GPCR-Metabolome Interactome and Enables Systematic De-Orphanization

G-LEAP uses deep learning to predict GPCR-ligand interactions, identifying endogenous ligands and orphan ligands with high accuracy.

LinkDTI: Drug-Target Interactionsprediction through a Link Predictionframework on Biomedical KnowledgeGraph

LinkDTI predicts drug-target interactions using a graph-based framework on a biomedical knowledge graph, outperforming baselines and identifying new interactions.

Chemical Probes in Scientific Literature: Expanding and Validating Target-Disease Evidence

This study systematically analyzes chemical probe literature to identify target-disease associations, revealing their utility in early therapeutic discovery and highlighting opportunities for repurposing.

Therapeutic Monoclonal Antibodies Repurposing in Oncology via IMGT/mAb-KG Embeddings

The paper develops a knowledge graph embedding tool for mAb repurposing in oncology, identifying two mAbs for clinical trials.

A Non-Alchemical Absolute Binding Free Energy Framework for Small Molecule Drugs

This paper introduces a non-alchemical ABFE framework for protein-ligand binding free energy calculations, showing improved accuracy and stability over existing methods.

Benchmarking Large Language Models for Predicting Therapeutic Antisense Oligonucleotide Efficacy

This study benchmarks large language models to predict antisense oligonucleotide efficacy using DNA sequences and target gene information, finding that GPT-3.5-Turbo performs best with few-shot prompting.

Drug Repurposing: A Potential Therapeutic Strategy for the Treatment of Chikugunya Virus

This study identifies Indinavir as a potential inhibitor of CHIKV nsP2 using HIV/HCV protease inhibitors through molecular dynamics simulations.

A Multi-Modal AI/ML-based Framework for Protein Conformation Selection and Prediction in Drug Discovery Applications

This paper presents a multi-modal AI/ML framework using GCNs to predict protein conformations for drug discovery by integrating global and local descriptors.

Reliable Evaluation and Learning in Multi-input Biological Association Prediction

The paper introduces a framework and model to evaluate and learn multi-input biological associations by eliminating shortcut biases, improving fairness and robustness in predictions.

Drug-Target Interaction Prediction with PIGLET

This paper introduces PIGLET, a graph transformer method for drug-target interaction prediction that outperforms existing models on a drug-based split of the Human dataset.

Atom-level generative foundation model for molecular interaction with pockets

This paper introduces PocketXMol, a generative foundation model for molecular interactions with protein pockets, achieving state-of-the-art performance in 11 tasks and demonstrating success in designing inhibitors and peptides.

Systematic Evaluation of Transfer Learning Strategies for Clinical Chemotherapy Response Prediction

This study evaluates transfer learning strategies for predicting chemotherapy response using clinical data, finding that hybrid and fine-tuning approaches perform better than direct model transfer or biomarker-based methods.

A Pan-Cancer Ex Vivo Drug Screen Atlas for Functional Precision Oncology

The paper presents a pan-cancer ex vivo drug screen atlas with 2.1M experiments, identifying tissue-specific drug sensitivities and demonstrating their predictive value for clinical profiles.

AI-guided competitive docking for virtual screening and compound efficacy prediction

This study uses deep learning tools to rank drug candidates by their binding affinity to protein targets, improving antibiotic design and drug discovery efficiency.

Deciphering DEL Pocket Patterns Through Contrastive Learning

This paper introduces ErePOC, a contrastive learning model for representing protein binding pockets, which improves DEL target selection by identifying shared characteristics and functional similarities.

EVOSYNTH: Enabling Multi-Target Drug Discovery through Latent Evolutionary Optimization and Synthesis-Aware Prioritization

This paper introduces a framework for multi-target drug discovery that combines latent evolution and synthesis-aware prioritization to generate and prioritize drug candidates with high translational potential.

Discovery of TDP-43 aggregation inhibitors via a hybrid machine learning framework

The paper develops a hybrid machine learning framework to identify TDP-43 aggregation inhibitors, discovering two compounds effective in reducing aggregation in cell and model organism assays.