← bioTriage-Open

AI-Driven Drug Discovery

Virtual screening, molecular generation, ADMET prediction, and computational drug repurposing from open-access literature.

56 papers curated · High: 1 · Medium: 24 · Low: 31 · Updated: 2026-11-19

Chemico-biological interactions 2026-12-01 Low

Integrative network pharmacology and experimental evaluation reveal TNBC-related biological activity and toxicity of Euphorbia umbellata crude latex.

This study combined computational network pharmacology with experimental testing in 4T1 triple-negative breast cancer cells to characterize the anti-cancer activity and toxicity of Euphorbia umbellata crude latex, a plant used in Brazilian traditional medicine. The authors curated 18 known latex compounds, predicted their molecular targets and pathways, and then validated biological effects experimentally. The work provides an integrative framework linking phytochemical composition to TNBC-relevant mechanisms and toxicity.

cancernetwork-analysisdrug-discovery
ACS infectious diseases 2026-11-13 Medium

A Machine Learning-Enhanced Ligand-Based Workflow Identifies Novel Cryptosporidium parvum Inhibitors.

The authors developed a ligand-based virtual screening workflow combining Thompson Sampling, a pharmacophore model, and a machine learning classifier to search 51 million virtual compounds from Enamine's REAL database for new Cryptosporidium parvum inhibitors. The approach identified novel candidate compounds, addressing the urgent need for alternatives to nitazoxanide, which is largely ineffective in severely immunocompromised patients.

drug-discoverymachine-learningmethodology
ChemMedChem 2026-11-12 Low

Drug Design and Discovery: The Long Road From Hansch Analysis to Artificial Intelligence.

This review traces the historical evolution of QSAR (quantitative structure-activity relationships) from Hansch analysis to modern machine learning and AI-driven drug discovery. It highlights how changes in molecular representation led to an explosion of descriptors and how growing data volumes necessitated advanced statistical techniques, with ML/AI now integral to pharmaceutical research.

drug-discoverymachine-learningreview
ChemMedChem 2026-11-12 Medium

From Generative Molecular Design to Agentic Artificial Intelligence in Drug Discovery.

This review traces the evolution of AI in early-stage drug discovery, from deep-learning generative models for designing small molecules and biologics to emerging agentic AI systems that can autonomously plan and execute multi-step tasks. It highlights that while state-of-the-art generative models exist, their behavior remains poorly understood, and rapid advances in agentic computational systems are reshaping human-machine interaction in target evaluation and molecule design.

drug-discoverymachine-learningreview
PLoS computational biology 2026-11-11 Medium

Pepti-DrugPhAI: Multimodal AI-evaluator for functional pharmacokinetics and toxicity of peptides.

Pepti-DrugPhAI is a multimodal AI tool that predicts pharmacokinetic properties and toxicity for peptide drug candidates, addressing a major bottleneck in peptide therapeutic development. It combines multiple data modalities to evaluate functional PK and safety profiles, aiming to reduce late-stage failures of promising peptide compounds.

machine-learningdrug-discoveryproteomics
RSC advances 2026-11-11 Low

Structure-function relationships in a square-planar Cu(ii) hybrid halometalate: experimental, DFT, molecular docking, optical, and biological studies.

The authors synthesized a new organic-inorganic copper(II) hybrid halometalate, (C6H8BrN2)2[CuBr4], and characterized it using X-ray crystallography, optical spectroscopy, DFT calculations, molecular docking, and biological assays. The crystal structure features discrete square-planar [CuBr4]2- anions held together by hydrogen bonds, π-π stacking, and halogen contacts, and the material shows semiconducting behavior with a ~2.5 eV band gap and broad visible emission. The work combines structural, electronic, and biological evaluations of a single hybrid material.

structural-biologymethodologydrug-discovery
Nanoscale advances 2026-11-10 Low

Integrated experimental and computational study of a novel indenonaphthodiazepine[5.3.3]propellane as a potential inhibitor of Mcl-1.

The authors synthesized a novel indenonaphthodiazepine[5.3.3]propellane compound via a one-pot multicomponent reaction and characterized it spectroscopically. DFT calculations probed its electronic properties, and molecular docking against the anti-apoptotic protein Mcl-1 gave a binding score of -5.678 kcal/mol, suggesting potential as an Mcl-1 inhibitor. The work is preliminary, relying on docking rather than experimental validation of binding or biological activity.

drug-discoverycancermethodology
Journal of advanced research 2026-11-09 Medium

GPCRAgent: an intelligent multi-agent platform integrating knowledge graphs and large language models for GPCR-targeted drug discovery.

GPCRAgent is a multi-agent platform that combines knowledge graphs with large language models to support GPCR-targeted drug discovery. It aims to overcome the limited integration of structured domain knowledge and poor interpretability of LLM-only approaches for GPCR research. The work presents an intelligent system for systematic characterization of GPCR functions and ligand interactions.

drug-discoveryLLMnetwork-analysis
Journal of chemical information and modeling 2026-11-09 Medium

QRxVision: Multimodal QR Encoding and Computer Vision for Domain-Specific Drug Discovery.

QRxVision introduces a multimodal framework that combines QR-code-style molecular encodings with computer vision to improve domain-specific drug discovery. The approach aims to make high-throughput virtual screening faster and more accessible by representing chemical structures in a machine-readable visual format that can be processed by vision models. This work proposes a novel encoding strategy for ligand-based screening rather than a validated new drug candidate.

drug-discoverymachine-learningmethodology
Journal of computer-aided molecular design 2026-11-06 Low

Anticancer activity of sodium propionate in hepatocellular carcinoma cells via downregulation of HDAC2 and HDAC6.

This study uses bioinformatics analyses of public HCC transcriptomic and chemical-gene interaction data to identify HDAC2 and HDAC6 as targets of sodium propionate, a short-chain fatty acid. In vitro experiments in hepatocellular carcinoma cells support that sodium propionate exerts anticancer activity by downregulating these histone deacetylases. The work positions a common dietary metabolite as a potential epigenetic modulator in liver cancer.

cancerepigenomicsdrug-discovery
Journal of biochemistry 2026-11-05 Low

Strengthening Japan's drug discovery and medical research ecosystem: AMED's innovation challenge.

This paper describes how AMED, Japan's main biomedical funding agency, is restructuring its programs under a 2025-2029 strategic plan to better translate academic research into medicines and healthcare technologies. It outlines five strategic priorities aimed at closing gaps between basic research and practical application, including stronger coordination across AMED's funding programs.

clinicaldrug-discoveryreview
Journal of microbiology & biology education 2026-11-05 Low

An educational activity tailored for undergraduate students on AI-driven nanobody design for targeted antiviral therapeutics.

This paper describes a short educational activity that teaches undergraduate students how to use generative AI tools to design nanobodies—small antibody fragments—against viral targets. It addresses the gap between the rapid adoption of AI in drug discovery and the lack of hands-on training available in typical undergraduate lab courses, where traditional nanobody discovery (requiring months of animal immunization and screening) is impractical.

LLMdrug-discoverymethodology
Bioorganic chemistry 2026-11-05 Low

Thymol-based derivatives as anticancer agents: design, synthesis, biological evaluation, and molecular docking studies.

Seventeen thymol-based ester derivatives were synthesized by coupling thymol with halogenated and non-halogenated cinnamoyl groups, then screened for anticancer activity against four cancer cell lines and normal fibroblasts. The derivative TM7 emerged as the most promising antiproliferative agent, and molecular docking was used to rationalize its binding to a likely target. This is an early-stage medicinal chemistry study identifying a new lead scaffold from a natural product.

drug-discoverycancermethodology
Royal Society open science 2026-11-04 Low

Discovery of potential ALK inhibitors from Glycosmis phytochemicals via integrated machine learning and umbrella sampling approaches.

This study used a computational pipeline combining machine learning, molecular docking, molecular dynamics, umbrella sampling, and toxicity prediction to screen 307 phytochemicals from the Glycosmis genus for potential inhibitors of anaplastic lymphoma kinase (ALK), a cancer drug target. A CatBoost regression model predicted binding free energies and narrowed the set to 26 compounds with strong predicted affinity, which were then evaluated with docking and free-energy methods. The work identifies candidate natural-product ALK inhibitors for further experimental validation.

drug-discoverymachine-learningcancer
RSC advances 2026-11-04 Low

Design, synthesis, and docking analysis of benzene sulfonamide-[1,4]benzoxazepine hybrids as acetylcholinesterase and butyrylcholinesterase inhibitors.

The authors synthesized a series of hybrid [1,4]benzoxazepine-sulfonamide compounds and tested them as inhibitors of acetylcholinesterase and butyrylcholinesterase, enzymes targeted in Alzheimer's disease therapy. Several compounds showed micromolar dual inhibition, with compound 12j being the most potent butyrylcholinesterase inhibitor (IC50 = 1.06 µM) and showing selectivity for BChE over AChE. Docking analysis was used to rationalize binding.

drug-discoverymethodologyneuroscience
RSC advances 2026-11-04 Low

Innovative design and development of novel thiazole-paracetamol hybrids as COX inhibitors and immunomodulatory activity: design, synthesis, SAR, in vitro biological evaluation, and molecular docking analysis.

This study designed and synthesized a series of thiazole-paracetamol hybrid molecules (compounds 4-11) linked via hydrazone or azo groups, aiming to combine anti-inflammatory COX inhibition with immunomodulatory activity in a single pharmacophore. The hybrids were chemically characterized and evaluated in vitro for biological activity, with structure-activity relationships and molecular docking used to rationalize binding to COX targets. The work represents a medicinal chemistry effort to develop improved analgesic/anti-inflammatory agents based on the paracetamol scaffold.

drug-discoverymethodologyclinical
RSC advances 2026-11-04 Low

Xanthine oxidase inhibitory evaluation of prenylated flavonoids from peanut (Arachis hypogaea L.) shells: in vitro and in silico molecular docking studies.

This study isolated two new prenylated flavonoids (arachisflavones A and B) along with several known prenylated flavones and chalcones from peanut shells, an agricultural waste product. The compounds were tested for xanthine oxidase inhibition in vitro and evaluated by molecular docking, identifying prenylated flavonoids as potential xanthine oxidase inhibitors with relevance to gout or hyperuricemia management.

metabolomicsdrug-discoveryplants
RSC advances 2026-11-04 Low

Rational design and synthesis of furan-based Schiff base quinazolin-4-amine hybrids: spectral elucidation, molecular docking, DFT studies, and in silico ADMET-antioxidant study.

The authors synthesized a series of eight furan-based quinazolin-4-amine Schiff base hybrids using a multi-step route and confirmed their structures spectroscopically. They then used computational methods—DFT calculations, molecular docking, and in silico ADMET/antioxidant prediction—to assess the compounds' potential as drug candidates. The work is purely computational and synthetic, with no experimental biological validation reported.

drug-discoverymethodologymachine-learning
ACS omega 2026-11-03 Medium

Blood-Brain Barrier Permeability Prediction by Integrating Deep Learning Representations with Chemical Compound Knowledge.

This paper presents a multimodal machine learning framework for predicting blood-brain barrier permeability by combining transformer-derived molecular embeddings with traditional physicochemical descriptors using late-stage ensemble fusion. The model was trained on B3DB and NeuTox datasets and evaluated on the BBBP benchmark, showing strong predictive performance. The approach aims to address limitations of both descriptor-based methods (lack of structural context) and deep learning methods (overfitting).

drug-discoverymachine-learningdeep-learning
RSC advances 2026-11-03 Low

Design, click synthesis, and biological evaluation of novel uracil-1,2,3-triazole hybrids as dual EGFR/CDK1 inhibitors: antiproliferative activity, mechanistic insights, and in silico studies.

The authors designed and synthesized a series of uracil-1,2,3-triazole hybrid compounds using click chemistry and tested them as dual inhibitors of EGFR and CDK1, two kinases important for cancer cell growth. They screened the compounds for antiproliferative activity against four human cancer cell lines and supported the results with mechanistic and computational (in silico) docking studies. The work identifies a new class of potential anticancer agents built on a uracil-triazole scaffold.

drug-discoverycancermethodology
JMIR AI 2026-11-03 Medium

Conversational Multi-Agentic Network-Driven Disease Module Identification and Drug Repurposing Prediction With ChatDRex: Platform Development Study.

ChatDRex is a conversational multi-agent platform that integrates heterogeneous biomedical data and specialized algorithms to identify disease modules in networks and predict drug repurposing candidates. It aims to streamline collaboration across pharmacology, medicine, biology, and bioinformatics by coordinating multiple agents through a chat interface. The platform addresses the fragmentation of existing tools and data sources, offering an end-to-end workflow for in silico repurposing.

drug-discoverynetwork-analysisLLM
Microbiology spectrum 2026-11-03 Medium

Exploiting Neosartorya (Aspergillus) fischeri antifungal protein 2-defined fungal targets for drug repurposing against Candida albicans.

This study uses the fungus-specific antifungal protein NFAP2 as a guide to identify its intracellular protein targets in Candida albicans (including glutamate decarboxylase and mitochondrial ATP synthase) and then applies structure-based drug repurposing to find existing compounds that hit these targets. The approach aims to accelerate antifungal development by leveraging drugs with known safety profiles against a pathogen with limited treatment options and rising resistance.

drug-discoverystructural-biologymicrobiology
Briefings in bioinformatics 2026-11-02 Medium

Graphics processing unit-accelerated quantum simulation for molecular biology and drug discovery: algorithms, platforms, and emerging applications.

This review surveys how GPUs can accelerate classical simulation of quantum algorithms for molecular biology and drug discovery, covering algorithms, platforms, and emerging applications. It argues that GPU-accelerated quantum simulation serves as a practical bridge for developing and benchmarking quantum algorithms despite scarce, noisy quantum hardware.

drug-discoverymethodologyreview
Briefings in bioinformatics 2026-11-02 Medium

Machine learning for RNA-small molecule interactions: methodologies, evaluation, and future directions.

This review surveys machine learning and deep learning methods for predicting RNA-small molecule interactions, focusing on three core tasks: binding site prediction, binding preference prediction, and related interaction modeling. It organizes the computational landscape, discusses evaluation practices, and outlines future directions for this emerging therapeutic target area.

machine-learningdrug-discoveryreview
Journal of cheminformatics 2026-11-02 Low

ADMET property prediction and toxicity profiling of CDK2 inhibitors: a practical cheminformatics tutorial using ADMET-AI 2.0 and RDKit.

This tutorial presents a reproducible Jupyter notebook workflow for large-scale ADMET profiling of 200 CDK2 inhibitors from ChEMBL, using the graph neural network platform ADMET-AI 2.0 combined with RDKit for cheminformatics preprocessing. It demonstrates how to compute physicochemical properties and predict pharmacokinetic and toxicity endpoints in a practical, accessible manner. The work serves as a hands-on guide for researchers wanting to integrate ADMET prediction into early drug discovery pipelines.

drug-discoverymachine-learningtool
Journal of medical Internet research 2026-11-02 Low

Can AI Help Repurpose Drugs to Fight Against Antibiotic Resistance?

This news and perspectives article reports on recent efforts to use AI models for drug repurposing and development against antimicrobial resistance. It highlights how machine learning can help identify candidate drugs to treat resistant infections, potentially reducing the cost and time of traditional drug discovery.

machine-learningdrug-discoverymicrobiology
PLOS digital health 2026-11-01 Low

Utilising digital technologies for mycetoma management: A comprehensive review of advances and opportunities.

This review examines how digital technologies—particularly AI for analyzing clinical images—could improve the diagnosis and management of mycetoma, a neglected tropical disease causing chronic subcutaneous infections in resource-limited regions. The authors argue that such tools may help overcome delayed diagnosis and limited access to specialized care, though the abstract emphasizes opportunities rather than reporting new empirical results.

clinicalmachine-learningreview
Chemical biology & drug design 2026-11-01 Low

Machine Learning-Guided Discovery of Novel Inhibitors of Escherichia coli MurB.

The authors built five machine learning models to predict MurB inhibition and used the best one (RDKit 2D descriptors) to screen an in-house compound library, yielding 238 hits that were then narrowed to 16 candidates via FRED molecular docking. MurB is a validated antibacterial target in E. coli cell wall biosynthesis, so the work provides a computational pipeline and candidate inhibitor set for further experimental testing.

machine-learningdrug-discoverymicrobiology
Expert opinion on drug discovery 2026-11-01 Medium

Crystallographic fragment screening in cancer and antiviral drug discovery: from electron density to translational outcomes.

This review discusses crystallographic fragment screening (CFS) as an experimental method for early-stage drug discovery, where small chemical fragments are soaked into protein crystals and their binding sites and poses are determined by X-ray crystallography. The authors argue that CFS provides a structural starting point that reduces uncertainty and risk before major medicinal chemistry investment, with applications in cancer and antiviral drug discovery.

drug-discoverystructural-biologymethodology
EClinicalMedicine 2026-11-01 Medium

A multimodal digital NeuroMarker platform with composite and domain-level validity for identification of mild cognitive impairment: a cross-sectional study.

This cross-sectional study presents a multimodal digital NeuroMarker platform that combines multiple digital assessments to identify mild cognitive impairment (MCI). The platform demonstrates both composite and domain-level validity, suggesting it can capture the heterogeneity of MCI while remaining scalable. This approach could help address the underdiagnosis of MCI by providing an accessible screening tool.

clinicalmachine-learningneuroscience
Chemistry & biodiversity 2026-11-01 Medium

Biosynthetic Pathways and Regulation of Antibacterial Compounds From Endophytic Fungi: From Chemistry to Genomics and Biotechnology.

This review integrates the chemistry, biosynthetic pathways, and genomic regulation of antibacterial natural products from endophytic fungi. It highlights that most biosynthetic gene clusters (BGCs) in these fungi are transcriptionally silent under standard laboratory conditions, leaving much of their chemical diversity unexplored, and discusses strategies for activating these cryptic BGCs.

genomicsmicrobiologydrug-discovery
Molecular informatics 2026-11-01 Medium

Interpretable Drug-Target Prediction via Geometric Alignment: A Linear Alternative to Deep Architectural Complexity.

The authors propose Multi-View Linear Manifold (MVLM), a linear method that aligns latent representations from ChemBERTa (for small molecules) and ESM-2 (for proteins) to predict kinase inhibitor–target interactions. By replacing complex deep architectures with a geometric alignment approach, they aim to improve generalization to new chemical scaffolds despite limited interaction data. The work positions linear alignment as an interpretable alternative to deep learning for drug-target prediction.

drug-discoverymachine-learningprotein-structure
bioRxiv : the preprint server for biology 2026-10-31 High

Small molecule structure elucidation from NMR spectra with generative and reasoning models.

The authors present an AI system that combines a molecular generative model (CHEFNMR2, an atomic flow-matching model) with a scientific reasoning agent to automatically determine the structure of unknown small molecules from NMR spectra. This addresses the labor-intensive, expert-dependent bottleneck of NMR-based structure elucidation, which is central to chemistry, pharmacology, and drug discovery.

metabolomicsmachine-learningdrug-discovery
Nucleic acids research 2026-10-31 Medium

MultiTargetDB: a comprehensive database of multitarget ligands for polypharmacology research and drug discovery.

MultiTargetDB is a new curated database of small molecules that bind multiple biological targets, designed to support polypharmacology research and multitarget drug discovery. It integrates chemical structures, molecular scaffolds, drug development stages, and detailed biological annotations with measurement-level provenance. The resource addresses the scarcity of well-curated multitarget ligand data for complex diseases.

drug-discoverydatabasechemical-biology
Journal of the Egyptian National Cancer Institute 2026-10-29 Low

Computational network pharmacology and molecular simulations to explore the potential mechanisms of Pyrrosia piloselloides compounds against prostate cancer.

This study uses computational network pharmacology and molecular dynamics simulations to predict which compounds in the dragon scale fern (Pyrrosia piloselloides) might act against prostate cancer and through which molecular targets. The authors identify candidate bioactive compounds and propose mechanisms based on target networks and binding simulations, but the findings are purely computational and lack experimental validation.

cancernetwork-analysisdrug-discovery
Molecules (Basel, Switzerland) 2026-10-21 Low

In Silico Investigation of Candidate Anticancer Targets for Aryltetralin Lignans Previously Reported from Cleistanthus Species.

This study used computational docking and target-prediction approaches to investigate which protein targets might explain the anticancer activity of six aryltetralin lignans previously isolated from Cleistanthus species. No new experiments were performed; the work integrates prior chemical and cytotoxicity data with in silico target identification. The findings propose candidate anticancer targets for these natural products, offering hypotheses for future experimental validation.

drug-discoverymachine-learningcancer
RSC advances 2026-10-21 Low

GNPS-assisted identification of bioactive constituents from Scoparia dulcis and their anti-inflammatory mechanisms via network pharmacology and experimental validation.

This study used LC-HRMS/MS molecular networking (GNPS) to profile the metabolites of Scoparia dulcis aerial parts, then combined network pharmacology, molecular docking, and molecular dynamics simulations with experimental validation to identify bioactive constituents and their anti-inflammatory mechanisms. The work links specific phytochemicals to putative inflammatory targets, offering a mechanistic rationale for the traditional use of this edible herb.

metabolomicsnetwork-analysisdrug-discovery
RSC advances 2026-10-21 Medium

Lipid tail length dictates membrane disruption by carmofur analogs and reveals noncanonical modes of action.

This study shows that the length of the lipid tail on carmofur analogs determines how they disrupt lipid membranes, revealing membrane interactions beyond carmofur's known enzyme-inhibition mechanisms. The findings suggest that these noncanonical membrane effects may contribute to the drug's activity and could inform the design of related compounds.

drug-discoverymethodologybiophysics
bioRxiv : the preprint server for biology 2026-10-20 Medium

Integrating Multisource Target Evidence with Machine-Learning Potency Prediction for Alzheimer's Disease Drug Discovery.

The authors built a traceable data-to-model platform that integrates Alzheimer's disease target evidence from biological, chemical, structural, regulatory, and trial-registry sources, covering 967 targets, 452,249 compounds, and 1,430,260 measurements. They show that target-specific fine-tuning of the Boltz-2 affinity prediction head substantially improves potency prediction accuracy on a curated PIK3CA dataset (Pearson r from 0.248 to 0.560, with reduced MAE). The work aims to make AD target evaluation and modeling more reproducible and integrated.

drug-discoverymachine-learningdatabase
bioRxiv : the preprint server for biology 2026-10-20 Medium

Advancing CD38 from a marker of disease activity to an active regulator of interferon responses in juvenile dermatomyositis.

This study investigates whether CD38, an NAD+-consuming enzyme elevated in juvenile dermatomyositis (JDM), is merely a marker of disease activity or an active driver of the type I interferon pathway that underlies the disease. The work aims to clarify CD38's mechanistic role, potentially informing more targeted therapies than current broad immunosuppression.

immunologytranscriptomicsclinical
SLAS technology 2026-10-20 Medium

NU-1025 modulates immunogenic cell death in bladder cancer: A multi-omics insight.

This study used integrative bioinformatics on TCGA and GEO bladder cancer transcriptomes to characterize immunogenic cell death (ICD)-associated subtypes, build a prognostic risk model, and profile immune infiltration. The authors then used Connectivity Map analysis to identify NU-1025 as a compound that modulates ICD, proposing it as a potential strategy to boost tumor immunogenicity and improve immune checkpoint blockade response.

cancertranscriptomicsimmunology
Frontiers in artificial intelligence 2026-01-01 Low

Navigating the Pharma 4.0 revolution: how AI-enabled applications and predictive modelling shaped pharmaceutical development in the post-pandemic era: a systematic review.

This systematic review examines how AI-enabled Pharma 4.0 technologies and predictive modelling have influenced pharmaceutical development since the COVID-19 pandemic, with particular attention to their use in discovering potential drug compounds for COVID-19. The authors synthesize evidence on how these implementations affect operational efficiency, resilience, and productivity across pharmaceutical systems, and discuss their potential to improve clinical outcomes.

drug-discoverymachine-learningreview
Research (Washington, D.C.) 2026-01-01 Medium

Artificial Intelligence as a Multifaceted Toolkit for Antibacterial Drug Discovery against Resistance.

This review argues that AI can serve as an integrated toolkit across the antibacterial drug discovery pipeline to combat antimicrobial resistance, rather than as an isolated technique. It emphasizes AI's strengths in multi-omics integration, complex pattern recognition, and addressing diverse resistance mechanisms at different translational stages. The paper frames AI as a rational, multifaceted strategy to counter slow antibiotic discovery and limited target diversity.

drug-discoverymachine-learningreview
Frontiers in pharmacology 2026-01-01 Medium

Computational drug screening for drug-resistant tuberculosis: advances, challenges, and future perspectives.

This review surveys computational drug discovery approaches—virtual screening, AI-driven modeling, and hybrid frameworks—for identifying new treatments against drug-resistant tuberculosis. It highlights how these methods have expanded the search space beyond conventional experimental pipelines, while discussing remaining challenges and future directions. The paper is a synthesis of existing work rather than a new experimental finding.

drug-discoverymachine-learningreview
Frontiers in cell and developmental biology 2026-01-01 Medium

SLC5A12 defines an invasive epithelial state associated with immune remodeling and therapeutic vulnerability in HNSCC.

This study identifies SLC5A12 as a marker of an invasive epithelial state in head and neck squamous cell carcinoma (HNSCC). This state is linked to immune remodeling within the tumor microenvironment and points to a potential therapeutic vulnerability. The work helps define malignant epithelial heterogeneity in HNSCC and its relationship to immune escape and treatment targets.

cancertranscriptomicsimmunology
Frontiers in bioengineering and biotechnology 2026-01-01 Medium

Design and validation of nucleosome binding peptides using BindCraft.

The authors used the BindCraft protein-design pipeline to computationally design short peptides that bind the nucleosome acidic patch, then experimentally validated their binding. This demonstrates that de novo peptide design can target specific chromatin interfaces, offering a route toward engineered epigenetic tools and probes.

protein-structuremethodologyepigenomics
International journal of chronic obstructive pulmonary disease 2026-01-01 Low

Integrative Single-Cell and Bulk RNA Sequencing Analysis Identifies Ion Channel-Related Biomarkers in Chronic Obstructive Pulmonary Disease.

This study integrates single-cell and bulk RNA sequencing data to identify ion channel-related genes as potential biomarkers in chronic obstructive pulmonary disease (COPD). The authors highlight that human ion channel genes, which mediate transmembrane signaling, may play underappreciated roles in COPD onset and progression. The work nominates candidate ion channel biomarkers for further validation.

transcriptomicssingle-cellclinical
Frontiers in molecular biosciences 2026-01-01 Low

Bone morphogenetic protein 4 meets artificial intelligence: a comprehensive review of AI-driven advances in drug discovery, structural modeling, and clinical translation.

This review surveys how AI and machine learning are being applied to study and target BMP4, a TGF-β family signaling protein implicated in development, cancer, fibrosis, and fibrodysplasia ossificans progressiva. It covers AI-driven advances across drug discovery, structural modeling, and clinical translation, spanning atomic-resolution structure prediction to population-level prognostication. The paper is a synthesis of recent progress rather than a primary research contribution.

reviewmachine-learningdrug-discovery
Frontiers in pharmacology 2026-01-01 Low

Hupehenine suppresses tumor cell phenotypes and modulates macrophage marker profiles in a diabeticmimetic colorectal cancer model.

This study identifies shared molecular features between diabetes and colorectal cancer and tests whether the natural compound hupehenine can modulate these pathways. Using a diabetic-mimetic CRC model, the authors show that hupehenine suppresses tumor cell phenotypes and alters macrophage marker profiles, suggesting a potential therapeutic axis at the diabetes–cancer interface.

cancerdrug-discoverytranscriptomics
Frontiers in pharmacology 2026-01-01 Medium

AI-enabled drug repurposing in Alzheimer's disease: from computational nomination to pharmacological proof.

This review surveys how AI-driven approaches—transcriptomic signature matching, network pharmacology, real-world pharmacoepidemiology, and multi-omics machine learning—have generated hundreds of candidate repurposed drugs for Alzheimer's disease, with repurposed agents now making up 35% of the 2026 AD pipeline and 42% of Phase 3 agents. The authors emphasize that computational nomination has not yet translated into clinical proof, citing the semaglutide trial as an example of the gap between AI prioritization and therapeutic validation.

drug-discoverymachine-learningreview
Scientifica 2026-01-01 Low

Structure-Based Discovery of Coumarin-Triazole Derivatives Targeting the Conserved QGDNQ Motif of Chandipura RdRp.

This study uses computational structure-based methods to identify coumarin-triazole derivatives as potential inhibitors of the Chandipura virus RNA-dependent RNA polymerase, targeting its conserved QGDNQ catalytic motif. The work is preliminary and in silico, proposing candidate compounds for a virus that currently lacks any approved antiviral treatment.

drug-discoverystructural-biologymicrobiology
Frontiers in pharmacology 2026-01-01 Low

Targeting IL-1R1 and androgen receptor using natural compounds from Aconitum heterophyllum and Podophyllum hexandrum: an in silico and in vitro study.

This study uses computational docking and in vitro experiments to identify natural compounds from two medicinal plants (Aconitum heterophyllum and Podophyllum hexandrum) that can simultaneously inhibit IL-1R1 (an inflammation receptor) and the androgen receptor (a hormone receptor driving prostate cancer). The dual-targeting approach aims to address both inflammation-driven tumor progression and hormone-dependent prostate cancer growth, potentially overcoming therapeutic resistance.

drug-discoverycancermethodology
Frontiers in bioinformatics 2026-01-01 Low

Comparative multi-target computational profiling of ebelin lactone across cognition-associated proteins.

This study computationally screens ebelin lactone, a triterpenoid derived from bacosides, against multiple proteins linked to cognition, including those involved in synaptic plasticity, neuronal proteostasis, mitochondrial stress, and neurite remodeling. The authors use docking and related in silico methods to profile binding compatibility across these targets, extending prior findings of affinity for M1 muscarinic and 5-HT2A receptors. The work is purely computational and does not include experimental validation.

drug-discoveryneurosciencemethodology
Frontiers in pharmacology 2026-01-01 Low

Mapping the subcellular pharmacokinetic-pharmacodynamic landscape: a systems-pharmacology framework for predicting combination therapies against intracellular bacterial infections.

This review proposes a systems pharmacology framework that maps how drugs move and act across different cellular compartments to predict combination therapies against intracellular bacterial infections. It argues that conventional drug discovery pipelines ignore the compartmentalized, dynamic nature of host cells where these pathogens hide, and that integrating subcellular pharmacokinetics and pharmacodynamics could improve treatment design.

systems-biologydrug-discoverymethodology
The Canadian journal of infectious diseases & medical microbiology = Journal canadien des maladies infectieuses et de la microbiologie medicale 2026-01-01 Low

Plasmid-Mediated AmpC Detection in Escherichia coli and Klebsiella pneumoniae: In Vitro Prevalence and In Silico Inhibitor Design.

This study screened clinical isolates of E. coli and K. pneumoniae from Lahore, Pakistan, for plasmid-mediated AmpC beta-lactamase production, establishing local prevalence rates of these antibiotic-resistance enzymes. The authors then used computational (in silico) docking and design approaches to identify candidate small-molecule inhibitors targeting AmpC beta-lactamases. The work combines regional surveillance data with early-stage structure-based drug design against a clinically important resistance mechanism.

microbiologydrug-discoveryclinical
PloS one 2026-01-01 Low

Investigating potential molecular mechanisms linking 6PPD-quinone exposure to Intervertebral Disc Degeneration: A network toxicology, molecular docking, and molecular dynamics simulation study.

This study uses computational approaches—network toxicology, molecular docking, and molecular dynamics simulation—to explore how exposure to 6PPD-quinone, an environmental pollutant from tire rubber, might contribute to intervertebral disc degeneration. The authors predict that 6PPD-Q activates stress-related pathways (SR-MMP and SR-ARE) and identify potential molecular targets linking the pollutant to disc degeneration. The work is entirely in silico and provides hypotheses rather than experimental validation.

network-analysismethodologydrug-discovery

Ask this channel

Answers come only from bioTriage’s indexed literature, scoped to this channel. Always verify before citing.