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).

• Directing stem cell differentiation by chromatin state approximation.

  PMID 41734818 · PMC12956330 · Nucleic acids research · 2026 · 8 claims · 8 setups

  Full-text index only

  Greedy selection of culture conditions by chromatin (ATAC-seq) distance to target is a viable optimisation strategy for differentiation protocols

  cd34+ cd34+-derived erythroid bone atac-seq flow rt-qpcr

• Multi-omic identification of key transcriptional regulatory programs during endurance exercise training in rats.

  PMID 41862462 · PMC13168590 · Nature communications · 2026 · 8 claims · 7 setups

  Full-text index only

  Gene regulatory responses to endurance exercise training (EET) are predominantly tissue-specific rather than shared across tissues

  tissues same skm-gn rna-seq atac-seq rrbs

• ChromBERT: A foundation model for learning interpretable representations for context-specific transcriptional regulatory networks.

  PMID 41592570 · PMC13069865 · Cell genomics · 2026 · 8 claims · 7 setups

  Full-text index only

  ChromBERT is pre-trained via masked reconstruction on the Cistrome-Human-6K dataset (6,391 cistromes, 991 transcription regulators) to learn genome-wide interaction syntax of transcription regulators

  lines chromosome embryonic transcription chip-seq chromhmm hi-c

• Cross-species prediction reveals chromatin regions with increased accessibility in humans.

  PMID 41984952 · PMC13082337 · Science advances · 2026 · 8 claims · 8 setups

  Full-text index only

  CNNs trained exclusively on human ATAC-seq data achieve cross-species prediction performance in chimpanzees and macaques comparable to species-specific models

  lymphoblastoid prefrontal six chimpanzee bulk single-cell cnn

• Early feature extraction drives model performance in high-resolution chromatin accessibility prediction.

  PMID 41526189 · PMC12951969 · Genome research · 2026 · 8 claims · 6 setups

  Full-text index only

  Early feature extraction (via ConvNeXt V2 blocks), rather than downstream architecture type, is the primary determinant of prediction accuracy in high-resolution chromatin accessibility prediction.

  lines cancer sets gm12878 atac-seq peak input-shift

• 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

  Full-text index only

  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.

  species gm12878 types chromatin cross-protein cross-cell-type

• Hi-Compass: a depth-aware deep learning framework for predicting cell-type-specific 3D genome organization from single-cell to spatial resolution.

  PMID 41980945 · PMC13250166 · Nature communications · 2026 · 8 claims · 8 setups

  Full-text index only

  Hi-Compass predicts cell-type-specific Hi-C contact maps using only ATAC-seq as cell-type-specific input, plus DNA sequence and a generalized CTCF binding profile

  gm12878 curated bulk training hi-c ctcf matched

• 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

• 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

• Systematic annotation of orphan RNAs reveals blood-accessible molecular barcodes of cancer identity and cancer-emergent oncogenic drivers.

  PMID 41579861 · PMC12923976 · Cell reports. Medicine · 2026 · 8 claims · 8 setups

  Full-text index only

  oncRNA binary presence-absence patterns constitute digital molecular barcodes that capture cancer type and subtype identity

  tcga exrna cptac mda-mb-231 small machine atac-seq

• Machine learning-predicted chromatin organization landscape across pediatric tumors.

  PMID 41904260 · PMC13039956 · Scientific reports · 2026 · 8 claims · 5 setups

  Full-text index only

  SuPreMo-Akita (built on the Akita CNN) enables systematic in silico prediction of somatic SV effects on 3D genome folding across large SV cohorts where experimental testing is infeasible

  somatic per-sample tumor paired mutagenesis dimensionality h3k27ac

• Parameter-efficient fine-tuning enables scalable transfer of regulatory sequence models to novel contexts.

  PMID 41618434 · PMC12930932 · Genome biology · 2026 · 8 claims · 7 setups

  Full-text index only

  PEFT enables accurate transfer of Borzoi to new datasets while significantly reducing GPU memory and runtime compared to joint training or full fine-tuning

  gtex encode wi-38 thp-1 bulk eqtl variant-flowfish

• Mapping Genetic Regulation of Transcription to Identify Functional Variants and Genes Associated with Pancreatic Cancer Risk.

  PMID 41824785 · PMC13205582 · Advanced science (Weinheim, Baden-Wurttemberg, Germany) · 2026 · 8 claims · 8 setups

  Full-text index only

  A genome-wide cis-eQTL meta-analysis of 482 pancreatic tissues (177 TCGA tumor + 305 GTEx normal) identified 1,123,483 significant SNP-gene pairs, 709,720 unique eQTLs, and 13,758 eGenes (FDR<0.05)

  tcga chinese significant egenes cis-eqtl gwas variant

• DMN-seq enriches DNA hypomethylated regions for biomarker discovery using 5-methylcytosine glycosylase.

  PMID 41673887 · PMC13097799 · Genome biology · 2026 · 8 claims · 9 setups

  Has reproduction · 80

  DMN-seq (DMN+) uses DME to nick DNA specifically at 5mC sites, enabling 5mC detection at single-base resolution via selective adaptor ligation

  synthetic partially embryonic mesc glycosylase dmn-seq ultrafast