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).

• MiPred: classification of real and pseudo microRNA precursors using random forest prediction model with combined features.

  PMID 17553836 · PMC1933124 · Nucleic acids research · 2007 · 8 claims · 8 setups

  Full-text index only

  A hybrid feature combining local contiguous triplet structure-sequence composition, MFE of the secondary structure, and P-value of a randomization test improves classification of real vs pseudo pre-miRNAs

  real combined pre-mirna training random independent repeated

• Searching for interpretable rules for disease mutations: a simulated annealing bump hunting strategy.

  PMID 16984653 · PMC1618409 · BMC bioinformatics · 2006 · 8 claims · 6 setups

  Full-text index only

  The proposed feature set outperforms existing published feature sets for predicting effects of amino acid substitutions

  coli bacteriophage lac proteins svm simulated pearson

• Multimodal data integration for biologically-relevant artificial intelligence to guide adjuvant chemotherapy in stage II colorectal cancer.

  PMID 40472802 · PMC12171563 · EBioMedicine · 2025 · 6 claims · 7 setups

  Has reproduction · 83

  AI-derived radiological clustering identifies stage II CRC patients with significantly different survival benefit from adjuvant chemotherapy

  stage patients tumour whole-tumour radiomics bulk cellular

• Workflow sharing with automated metadata validation and test execution to improve the reusability of published workflows.

  PMID 36810800 · PMC9944229 · GigaScience · 2022 · 8 claims · 5 setups

  Has reproduction · 50

  Yevis is a system that builds a workflow registry which automatically validates and tests workflows prior to publication, ensuring they are 'reusable with confidence'.

  github submitter's zenodo yevis-based workflow file trs

• NCBI Reference Sequence (RefSeq): a curated non-redundant sequence database of genomes, transcripts and proteins.

  PMID 15608248 · PMC539979 · Nucleic acids research · 2005 · 7 claims · 5 setups

  Full-text index only

  RefSeq provides a curated, non-redundant, explicitly linked collection of genomic, transcript and protein sequences spanning prokaryotes, eukaryotes and viruses.

  refseq honey sequence computational conserved

• Speeding disease gene discovery by sequence based candidate prioritization.

  PMID 15766383 · PMC1274252 · BMC bioinformatics · 2005 · 7 claims · 8 setups

  Full-text index only

  Disease genes (OMIM) differ significantly from non-disease genes in sequence-based features including gene/cDNA/protein size, exon number, homolog conservation, secretion signal, 3' UTR length, CpG islands, and distance to nearest gene.

  ensembl training same hgmd comparative machine tenfold

• Prognostic biomarker discovery in pancreatic cancer through hybrid ensemble feature selection and multi-omics data.

  PMID 41957754 · PMC13188360 · BioData mining · 2026 · 7 claims · 3 setups

  Has reproduction · 92

  The hEFS framework integrates data subsampling with multiple prognostic models (embedded and wrapper-based), aggregates feature rankings via a voting-theory-inspired approach, and selects the optimal feature subset via Pareto front optimization, eliminating user-defined thresholds.

  pancreatic pdac multi-omics gene monte

• Classification of real and pseudo microRNA precursors using local structure-sequence features and support vector machine.

  PMID 16381612 · PMC1360673 · BMC bioinformatics · 2005 · 7 claims · 7 setups

  Full-text index only

  A 32-dimensional triplet structure-sequence feature vector combined with SVM (triplet-SVM) can distinguish real human pre-miRNAs from pseudo pre-miRNA hairpins with ~90% accuracy.

  pre-mirnas test conserved-hairpin cross-species secondary svm cross-species

• A comparison of classification methods for predicting Chronic Fatigue Syndrome based on genetic data.

  PMID 19772600 · PMC2765429 · Journal of translational medicine · 2009 · 7 claims · 3 setups

  Full-text index only

  The naive Bayes model with the wrapper-based feature selection approach performed best among all predictive models tested for distinguishing CFS from controls.

  subjects naive svm c4.5

• 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

• A Decentralized Kidney Transplant Biopsy Classifier for Transplant Rejection Developed Using Genes of the Banff-Human Organ Transplant Panel.

  PMID 35619722 · PMC9128066 · Frontiers in immunology · 2022 · 6 claims · 6 setups

  Has reproduction · 90

  A random forest model trained solely on B-HOT panel genes (B-HOT Model) accurately classifies kidney transplant biopsies as NR, ABMR, or TCMR.

  kidney gene combined microarray machine sequential

• Constructing support vector machine ensembles for cancer classification based on proteomic profiling.

  PMID 16689692 · PMC5173238 · Genomics, proteomics & bioinformatics · 2005 · 7 claims · 4 setups

  Full-text index only

  CSVME, built by selecting a subset of base SVMs via SVM-RFE ranking and fusing them with a trained upper-layer SVM, achieves better classification performance than an ensemble of all base SVMs.

  serum computational seldi-tof svm-rfe-based 10-fold

• Identification of deleterious non-synonymous single nucleotide polymorphisms using sequence-derived information.

  PMID 18588693 · PMC2446391 · BMC bioinformatics · 2008 · 8 claims · 5 setups

  Full-text index only

  A decision tree built on 10 selected sequence-derived features classifies SAPs as Disease or Polymorphism with 82.6% accuracy and 0.607 MCC in cross-validation.

  sap sequence-derived humvar same decision independent greedy

• Gene-expression patterns in peripheral blood classify familial breast cancer susceptibility.

  PMID 26538066 · PMC4634735 · BMC medical genomics · 2015 · 8 claims · 5 setups

  Has reproduction · 83

  A multigene peripheral-blood gene-expression biomarker accurately classifies which women from high-risk families develop familial breast cancer.

  peripheral pbmcs blood normal gene-expression lymphocyte gene

• STAT1 and IL-7 as potential diagnostic biomarkers for distinguishing high-grade from low-grade serous ovarian cancer: a multi-cohort analysis.

  PMID 42058211 · PMC13120972 · Frontiers in immunology · 2026 · 7 claims · 6 setups

  Has reproduction · 50

  STAT1 and IL-7 are differentially expressed immune-related genes that can distinguish HGSOC from LGSOC and may serve as ancillary diagnostic biomarkers.

  hgsoc training independent paraffin-embedded transcriptome machine-learning diagnostic

• Boosting accuracy of automated classification of fluorescence microscope images for location proteomics.

  PMID 15207009 · PMC449699 · BMC bioinformatics · 2004 · 8 claims · 8 setups

  Full-text index only

  New classifiers (SVMs, ensembles) and new wavelet-derived (Gabor, Daubechies) features improve recognition of protein subcellular location patterns over the previous neural network approach

  hela automated stepwise training-set-size

• Interaction profile-based protein classification of death domain.

  PMID 15189571 · PMC459208 · BMC bioinformatics · 2004 · 7 claims · 6 setups

  Full-text index only

  An SVM-based classifier using Residue Pair Interaction Profiles (RPIPs) can classify death domain superfamily members into subfamilies with 89% average cross-validation accuracy

  homology-modeled death homology putative rigid-body homology model

• Prioritization of candidate cancer genes--an aid to oncogenomic studies.

  PMID 18710882 · PMC2566894 · Nucleic acids research · 2008 · 8 claims · 8 setups

  Full-text index only

  Computational classifiers using combinations of protein conservation, gene structure, protein domains, protein interactions, and regulatory data can distinguish known cancer genes (CD/CR) from unlabelled human genes

  genes cancer interpro interactome averaged cross-validation domain

• Comparative analysis of molecular signatures reveals a hybrid approach in breast cancer: Combining the Nottingham Prognostic Index with gene expressions into a hybrid signature.

  PMID 35143511 · PMC8830616 · PloS one · 2022 · 8 claims · 6 setups

  Has reproduction · 48

  A hybrid signature combining the Nottingham Prognostic Index with SIS-selected gene expressions can be built in a data-driven fashion (NPI treated as a gene expression during feature selection).

  metabric scan-b gene rna-seq feature

• Smart spatial omics (S2-omics) optimizes region of interest selection to capture molecular heterogeneity in diverse tissues.

  PMID 41298871 · PMC12662399 · Nature cell biology · 2025 · 7 claims · 6 setups

  Has reproduction · 100

  S2-omics is an end-to-end workflow that automatically selects ROIs from H&E histology images to maximize molecular information content for spatial omics profiling.

  gastric colorectal additional breast xenium visium histology