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).

• Exhaustive prediction of disease susceptibility to coding base changes in the human genome.

  PMID 18793467 · PMC2537574 · BMC bioinformatics · 2008 · 8 claims · 7 setups

  Full-text index only

  Inter-species conservation is the strongest single predictor of disease-associated coding mutations among the factors tested.

  hgmd refseq aligned svm svm reciprocal inter-species

• nsSNPAnalyzer: identifying disease-associated nonsynonymous single nucleotide polymorphisms.

  PMID 15980516 · PMC1160133 · Nucleic acids research · 2005 · 6 claims · 4 setups

  Full-text index only

  nsSNPAnalyzer is a web server that predicts whether a query nsSNP is disease-associated or functionally neutral using a Random Forest classifier combining structural and evolutionary information

  curated homologous structure random structural secondary

• Application of machine learning in SNP discovery.

  PMID 16398931 · PMC1955739 · BMC bioinformatics · 2006 · 8 claims · 6 setups

  Full-text index only

  PolyBayes produces high false-positive SNP predictions even with stringent parameters

  diverse soybean candidate same sts polybayes polyphred

• Protein ranking by semi-supervised network propagation.

  PMID 16723003 · PMC1810311 · BMC bioinformatics · 2006 · 8 claims · 5 setups

  Full-text index only

  RankProp, a diffusion-based network propagation algorithm on a PSI-BLAST-derived protein similarity network, significantly outperforms local search methods (BLAST/PSI-BLAST) at detecting remote homologs.

  scop similarity same psi-blast rankprop probabilistic

• Fast, accurate, and racially unbiased pan-cancer tumor-only variant calling with tabular machine learning.

  PMID 36611079 · PMC9825621 · NPJ precision oncology · 2023 · 7 claims · 8 setups

  Has reproduction

  Tabular ML classifiers (TabNet, XGBoost, LightGBM) trained on tumor-only-derived features achieve state-of-the-art somatic vs germline classification, with AUC>94% on TCGA holdout and AUC>85% on metastatic melanoma.

  tcga metastatic tumor-only whole-exome wes copy-number

• Computational disease gene identification: a concert of methods prioritizes type 2 diabetes and obesity candidate genes.

  PMID 16757574 · PMC1475747 · Nucleic acids research · 2006 · 6 claims · 8 setups

  Full-text index only

  Applying seven independent computational disease-gene prioritization methods in concert to 9556 positional candidate genes identifies a prioritized set of likely T2D and obesity candidate genes

  pubmed genes same randomly geneseeker evoc disease

• Identification of immune-related signatures and pathogenesis differences between thoracic aortic aneurysm patients with bicuspid versus tricuspid valves via weighted gene co-expression network analysis.

  PMID 37883426 · PMC10602290 · PloS one · 2023 · 6 claims · 7 setups

  Has reproduction · 97

  TAA/TAV pathogenesis is more associated with immune-related gene expression than TAA/BAV, with two WGCNA gene modules (brown and blue) enriched for immune functions.

  aortic brown genes training wgcna biological gsva

• Predicting failure rate of PCR in large genomes.

  PMID 18492719 · PMC2441781 · Nucleic acids research · 2008 · 7 claims · 8 setups

  Full-text index only

  The number of predicted primer-binding sites in genomic DNA is the most important factor determining PCR failure.

  genomic genome pcr combined large-scale exact-match mismatch-tolerant