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).

• DeeReCT-APA: Prediction of Alternative Polyadenylation Site Usage Through Deep Learning.

  PMID 33662629 · PMC9801043 · Genomics, proteomics & bioinformatics · 2022 · 8 claims · 8 setups

  Has reproduction · 50

  DeeReCT-APA quantitatively predicts the usage of all competing PASs of a gene simultaneously, rather than casting the problem as pairwise comparison like prior methods.

  fibroblast genomic same 3'-region 3'-mrna deep

• Uncertainty-aware genomic deep learning with knowledge distillation.

  PMID 41523993 · PMC12779563 · NPJ artificial intelligence · 2026 · 7 claims · 6 setups

  Full-text index only

  DEGU distills an ensemble of teacher DNNs into a single student model by jointly predicting the ensemble mean and the variability (epistemic uncertainty) across ensemble predictions.

  drosophila k562 deepstarr starr-seq starr-seq-derived lentimpra base-resolution

• How negative sampling shapes the performance of transcription factor binding site prediction models.

  PMID 41601205 · PMC12910371 · Bioinformatics (Oxford, England) · 2026 · 7 claims · 5 setups

  Full-text index only

  Negative sampling technique significantly impacts TFBS prediction model performance and interpretation of results

  encode same genome chip-seq atac-seq deep

• AMR-GNN: a multi-representation graph neural network framework to enable genomic antimicrobial resistance prediction.

  PMID 41792137 · PMC13087051 · Nature communications · 2026 · 7 claims · 8 setups

  Full-text index only

  AMR-GNN, a graph neural network integrating multiple genomic representations (unitigs, SNPs, FCGR) via low-rank multimodal fusion, improves AMR phenotype prediction in P. aeruginosa compared to single-representation baseline models.

  pseudomonas aeruginosa escherichia coli whole-genome elastic convolutional

• BiRNA-BERT allows efficient RNA language modeling with adaptive tokenization.

  PMID 41266599 · PMC12635123 · Communications biology · 2025 · 8 claims · 8 setups

  Has reproduction · 50

  BiRNA-BERT uses adaptive dual-tokenization that dynamically selects nucleotide-level (NUC) or byte-pair encoding (BPE) tokens based on input sequence length

  rfam plant lncrna-mirna sequences unsupervised mirna-lncrna length-binned

• A generic reference defined by consensus peaks for single-cell ATAC-seq data analysis.

  PMID 41663439 · PMC12996591 · Nature communications · 2026 · 7 claims · 7 setups

  Full-text index only

  Aggregating peaks from 624 high-quality bulk ATAC-seq datasets defines ~1.4 million observed consensus peaks (cPeaks) covering ~30% of the genome.

  bulk scatac-seq cnn gm12878 bulk scatac-seq deep