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).
-
Has reproduction · 86
Molecular Classification Models for Triple Negative Breast Cancer Subtype Using Machine Learning.
PMID 34575658 · PMC8472680 · Journal of personalized medicine · 2021 · 6 claims · 4 setups
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.
-
Full-text index only
Prediction of candidate primary immunodeficiency disease genes using a support vector machine learning approach.
PMID 19801557 · PMC2780952 · DNA research : an international journal for rapid publication of reports on genes and genomes · 2009 · 6 claims · 3 setups
An SVM trained on 69 binary features of known PID genes can accurately classify PID vs non-PID genes and predict novel candidate PID genes
-
Full-text index only
Predicting positive p53 cancer rescue regions using Most Informative Positive (MIP) active learning.
PMID 19756158 · PMC2742196 · PLoS computational biology · 2009 · 8 claims · 4 setups
MIP active learning is a novel active learning method that preferentially seeks informative Positive (functionally active) examples rather than only maximizing classifier accuracy.
-
Full-text index only
Swarm intelligence based wavelet coefficient feature selection for mass spectral classification: an application to proteomics data.
PMID 19733729 · PMC2748225 · Analytica chimica acta · 2009 · 8 claims · 4 setups
ACA-based wavelet coefficient feature selection can achieve up to 100% classification accuracy on training, validating, and independent testing sets using only 5 selected features.