Computational Drug Discovery Methods

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.

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.

The paper analyzes a hypothetical protein from C. difficile to identify potential ligands for therapeutic development through physicochemical properties, structural prediction, and molecular docking studies.

This paper introduces Ouroboros, a framework that improves drug discovery by integrating molecular representation learning with generative modeling, outperforming traditional methods in multiple tasks.

The paper identifies two potential MCL1 inhibitors through computational methods including pharmacophore modeling, QSAR, docking, and DFT analysis, showing favorable ADMET properties and binding stability.

The paper introduces a hybrid machine learning framework that improves protein-ligand interaction prediction by combining biophysicochemical insights with large-scale representations, achieving faster and more accurate virtual screening.

The paper introduces PREDR, a model that predicts drug-disease relationships by reconstructing meta-paths from heterogeneous biological knowledge graphs, outperforming existing methods in drug repurposing.

This paper identifies flaws in current Drug Response Prediction evaluation methods and proposes new strategies to improve their reliability and generalization.

This study introduces a context-aware hierarchical fusion model for drug relational learning, showing improved performance in predicting drug-drug interactions and side effects across different contexts.

This study benchmarks HelixFold3's ability to predict holo structures for free energy perturbation calculations, showing improved performance over AlphaFold2 and existing methods.

This paper uses uniform design with AI to predict physicochemical properties of short peptides, identifying key factors affecting aggregation and providing predictive models and datasets.

The paper develops fine-tuned deep learning models for predicting safer drugs targeting class A GPCRs by leveraging transfer learning and sequence-based analysis.

The paper evaluates structure-based methods to predict how kinase mutations affect inhibitor binding, finding they accurately classify mutations as resistant or sensitizing.

This paper introduces a tool that integrates novel chemical reactions into retrosynthesis tools, improving performance in 70% of cases and reducing synthesis costs by 20%.

This paper develops a high-throughput screening method to identify drugs targeting polyploid giant cancer cells in breast cancer, combining empirical assays with a machine learning model for virtual screening.

The paper proposes a Graph Transformer-Convolution Network (GTCN) to predict drug-cancer associations by integrating heterogeneous biological data and enhancing interpretability.

This study benchmarks ML-integrated approaches for predicting PK profiles in rats, showing comparable accuracy across multiple methods for prioritizing drug candidates.

The paper introduces IFPP, a molecular representation for identifying isofunctional molecules, and proposes Metric Learning to improve scalability for large libraries.

This paper addresses data imbalance in drug response prediction by introducing a multi-objective optimization approach using a custom loss function to improve model performance.

This paper identifies D-Talopyranose as a potential drug candidate targeting Brucella melitensis DBP to combat antibiotic resistance through computational methods.

GPCRVS predicts compound activity against GPCRs using machine learning, validated on ChEMBL datasets for secretin-like and chemokine receptors.

The paper introduces SDDF, a platform using volunteer computing and active learning to efficiently generate molecular conformation datasets for drug discovery.

This paper evaluates the use of GPT for drug sensitivity prediction using pharmacogenomics data, showing improved performance through prompt engineering and fine-tuning.

BaCNet uses deep learning to predict antibiotic-target interactions and discover novel compounds against drug-resistant pathogens.

This paper develops an improved machine learning model for predicting drug combination dose-response surfaces, showing better performance than scalar-valued synergy score methods.

DrugDiff generates novel small molecules with desired properties using a diffusion model guided by predictors.

The paper uses machine learning to predict drug responses in colorectal cancer cells, identifying novel compounds with cytotoxic or cytostatic effects.

The paper develops ML models to predict antimicrobial activity of small molecules using a knowledge graph, achieving 75.9% accuracy in identifying active compounds.

MolGene-E generates high-quality drug molecules from single-cell transcriptomics data, outperforming baselines in zero-shot molecule generation.

This study identifies small molecule binders to hnRNPA1 using MD simulations and fragment screening, revealing cryptic pockets and conformational changes.