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).
-
Full-text index only
Prediction of myeloid malignant cells in Fanconi anemia using machine learning.
PMID 41557613 · PMC12818649 · PloS one · 2026 · 6 claims · 7 setups
A DNN classifier trained on AML scRNA-seq data accurately predicts AML-like transcriptional profiles at single-cell resolution
-
Full-text index only
AI-guided multi-omics analysis identifies NPC1-modulated susceptibility to SARS-CoV-2 infection under PM(2.5) exposure.
PMID 41912520 · PMC13194984 · Nature communications · 2026 · 8 claims · 6 setups
A fine-tuned Geneformer (single-cell transcriptomics transformer) model classifies PM2.5 exposure status and generalizes better than BERT models trained from scratch, especially with limited data.
-
Full-text index only
scMILD: Single-cell multiple instance learning for sample classification and associated subpopulation discovery.
PMID 41907409 · PMC13019583 · iScience · 2026 · 8 claims · 8 setups
scMILD identifies condition-associated cells using only sample-level labels via a dual-branch MIL architecture with a shared encoder
-
Full-text index only
A generic reference defined by consensus peaks for single-cell ATAC-seq data analysis.
PMID 41663439 · PMC12996591 · Nature communications · 2026 · 7 claims · 7 setups
Aggregating peaks from 624 high-quality bulk ATAC-seq datasets defines ~1.4 million observed consensus peaks (cPeaks) covering ~30% of the genome.
-
Full-text index only
CLAMP: predicting specific protein-mediated chromatin loops in diverse species with a chromatin accessibility language model.
PMID 41555433 · PMC12903630 · Genome biology · 2026 · 8 claims · 8 setups
CLAMP, a chromatin-accessibility language model, predicts protein-mediated chromatin loops across 10 species, 18 proteins, and 24 cell types with superior performance versus existing methods.