Experiments

Searchable full-text extractions: founding hypothesis, core claims, experimental setups, key results and statistics — pulled out of each paper as structure. Search a cell line, an assay or an entity (e.g. HUH7) and find every paper that worked with it. This corpus stands on its own: most entries carry no reproduction assessment (yet).

• Molecular Classification Models for Triple Negative Breast Cancer Subtype Using Machine Learning.

  PMID 34575658 · PMC8472680 · Journal of personalized medicine · 2021 · 6 claims · 4 setups

  Has reproduction · 86

  A training gene set of 719 unique upregulated DEGs (subtype-specific) can be used to build ML models that classify TNBC into BLIA, BLIS, MES, and LAR subtypes.

  known tnbc unique unknown gene machine term

• Atlas - a data warehouse for integrative bioinformatics.

  PMID 15723693 · PMC554782 · BMC bioinformatics · 2005 · 8 claims · 3 setups

  Full-text index only

  Atlas is a biological data warehouse that locally stores and integrates sequences, molecular interactions, homology information, functional annotations, and ontologies

  in-house gene omim genome molecular gene-disease

• Artificial Intelligence Meets Whole Slide Images: Deep Learning Model Shapes an Immune-Hot Tumor and Guides Precision Therapy in Bladder Cancer.

  PMID 36245985 · PMC9553530 · Journal of oncology · 2022 · 6 claims · 8 setups

  Has reproduction

  A deep learning WSI cluster (three-class mini batch K-means on Inception V3 features) is associated with overall survival (P<0.001) and is an independent prognostic predictor (P=0.031) in BLCA.

  tcga-blca blca tcga deep transfer-learning neural-network

• Automatic discovery of 100-miRNA signature for cancer classification using ensemble feature selection.

  PMID 31533612 · PMC6751684 · BMC bioinformatics · 2019 · 7 claims · 8 setups

  Has reproduction · 69

  An ensemble feature selection method based on classifier consensus identifies a robust 100-miRNA signature from TCGA data.

  tcga geo 100-mirna mirna feature bibliographic

• Machine learning and free energy clustering reveal PAH protein binding linked to AD risk.

  PMID 41953002 · PMC13053772 · iScience · 2026 · 7 claims · 8 setups

  Has reproduction · 78

  An integrated framework of bioinformatics, machine learning, and ΔG clustering can prioritize PAHs for AD-associated neurotoxicity.

  proteins microarray ad-associated pahs target differential wgcna

• PA-GOSUB: a searchable database of model organism protein sequences with their predicted Gene Ontology molecular function and subcellular localization.

  PMID 15608166 · PMC540074 · Nucleic acids research · 2005 · 7 claims · 4 setups

  Full-text index only

  PA-GOSUB significantly extends the coverage of GO molecular function and subcellular localization annotations for 10 model organism proteomes compared with existing databases (GOA, Swiss-Prot).

  sequences query swiss-prot naive-bayes blast feature

• iRefIndex: a consolidated protein interaction database with provenance.

  PMID 18823568 · PMC2573892 · BMC bioinformatics · 2008 · 6 claims · 3 setups

  Full-text index only

  A reproducible key (ROG) for each protein interactor and a corresponding key (RIG) for each interaction record can be generated by anyone using only primary sequence, taxonomy identifier, and the SHA-1 algorithm (SEGUID).

  sequence nine interactor seguid parsing mapping

• Integrative transcriptomic and machine learning framework reveals candidate genes and potential mechanisms of aflatoxin B1 exposure in breast cancer.

  PMID 41688730 · PMC12982753 · Scientific reports · 2026 · 7 claims · 8 setups

  Has reproduction · 83

  Twenty-two genes lie at the intersection of AFB1-predicted targets and breast cancer-associated co-expression modules/DEGs

  afb1 breast training candidate computational bulk weighted

• Reconstruction of human protein interolog network using evolutionary conserved network.

  PMID 17493278 · PMC1885812 · BMC bioinformatics · 2007 · 8 claims · 7 setups

  Full-text index only

  A relative conservation score derived from maximal quasi-cliques in protein interaction networks, combined with other interaction features, can score and rank predicted human interologs for confidence.

  ppi sequences interaction predicted interolog orthology quasi-clique

• Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning.

  PMID 19756158 · PMC2742196 · PLoS computational biology · 2009 · 8 claims · 4 setups

  Full-text index only

  MIP active learning is a novel active learning method that preferentially seeks informative Positive (functionally active) examples rather than only maximizing classifier accuracy.

  engineered p53 computational yeast-based regional homology

• Anopheles gambiae genome reannotation through synthesis of ab initio and comparative gene prediction algorithms.

  PMID 16569258 · PMC1557760 · Genome biology · 2006 · 8 claims · 7 setups

  Full-text index only

  An exon-gene-union (EGU) algorithm followed by an open-reading-frame-selection algorithm can synthesize ab initio (GENSCAN, GeneMark, SNAP) and comparative (Ensembl/Genewise) predictions into a single, more complete CDS set

  anopheles gambiae reanocds05 initio full-length rt-pcr