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).

• 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

• Multiomics and deep learning dissect regulatory syntax in human development.

  PMID 41951735 · PMC13216069 · Nature · 2026 · 8 claims · 8 setups

  Full-text index only

  The Human Development Multiomic Atlas (HDMA) is a single-cell atlas of chromatin accessibility and gene expression from 817,740 fetal cells across 12 organs, spanning 203 cell types

  fetal annotated embryo chrombpnet share-seq chromvar activity-by-contact

• Machine and Deep Learning Reveal Sequence Determinants Encoding Bivalent Histone Modifications.

  PMID 41946881 · PMC13057473 · Communications biology · 2026 · 8 claims · 7 setups

  Full-text index only

  Bivalent domains have higher GC content and stronger evolutionary conservation than monovalent regions

  embryonic mesc random bivalent chip-seq rna-seq k-mer-based

• Detection of alternative splicing: deep sequencing or deep learning?

  PMID 41520225 · PMC12790623 · Briefings in bioinformatics · 2026 · 8 claims · 8 setups

  Full-text index only

  Sequence-based deep learning tools (AlphaGenome, SpliceAI, DeepSplice) show potential for initial hypothesis development and as additional filters in standard RNA-seq pipelines, especially when sequencing depth is limited.

  dilated dcm sequence subsampled bulk rna-seq deep

• DeepProg: an ensemble of deep-learning and machine-learning models for prognosis prediction using multi-omics data.

  PMID 34261540 · PMC8281595 · Genome medicine · 2021 · 8 claims · 7 setups

  Has reproduction · 84

  DeepProg is a novel ensemble framework of deep-learning and machine-learning approaches that robustly predicts patient survival subtypes using multi-omics data

  cancer breast hepatocarcinoma tcga multi-omics gene rna-seq

• EpiXFormer: a cross-attention neural network for predicting cell type-specific transcription factor binding sites.

  PMID 41527854 · PMC12796812 · Briefings in bioinformatics · 2026 · 8 claims · 8 setups

  Full-text index only

  EpiXFormer achieves high accuracy (mean AUROC ~0.99) predicting binding sites of both TFs and non-sequence-specific DBPs across 199 DBP-cell type pairs

  dbps tfs common ctcf deep comparative feature-ablation

• Prediction of Alzheimer's disease-specific phospholipase c gamma-1 SNV by deep learning-based approach for high-throughput screening.

  PMID 33397809 · PMC7826347 · Proceedings of the National Academy of Sciences of the United States of America · 2021 · 7 claims · 7 setups

  Has reproduction · 48

  An AD-specific frameshift insertion SNV in exon 27 of mouse PLCγ1 causes abnormal exon skipping (exons 26-30) during mRNA maturation in 5xFAD cortex

  cortex forebrain plcγ1 bulk rna-seq-based western

• Deep learning linking mechanistic models to single-cell transcriptomics data reveals transcriptional bursting in response to DNA damage.

  PMID 41779826 · PMC12959883 · eLife · 2026 · 8 claims · 5 setups

  Full-text index only

  DeepTX is an interpretable, scalable deep learning inference framework that links mechanistic transcription models to scRNA-seq data to infer genome-wide transcriptional burst kinetics

  synthetic embryonic colorectal computational stochastic scrna-seq deep

• Interpretable prediction models for widespread m6A RNA modification across cell lines and tissues.

  PMID 37995291 · PMC10697738 · Bioinformatics (Oxford, England) · 2023 · 8 claims · 8 setups

  Has reproduction · 79

  CLSM6A is a set of CNN-based deep learning models that predict single-nucleotide-resolution m6A RNA modification sites across eight cell lines and three tissues in H. sapiens

  hct116 a549 brain hek293t miclip m6a-clip-seq m6a-seq

• 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

• Annotation-free prediction of immunotherapy response in melanoma using single-cell transcriptomic data.

  PMID 41758825 · PMC12948085 · PloS one · 2026 · 8 claims · 6 setups

  Full-text index only

  AI-based predictive models built on unannotated scRNA-seq data (cell-by-gene expression matrices) can classify melanoma patients as ICI responders vs. non-responders

  cd45+ scrna-seq independent scrna-seq single-gene-level pathway-level

• Attention-based deep learning for analysis of pathology images and gene expression data in lung squamous premalignant lesions.

  PMID 41952176 · PMC13188385 · Genome medicine · 2026 · 8 claims · 8 setups

  Full-text index only

  A multimodal transformer combining WSIs and gene expression outperforms models trained on either modality alone for classifying bronchial dysplasia or worse vs non-dysplasia.

  endobronchial bronchial colorado jules slide bulk microarray

• Adaptive learning embedding features to improve the predictive performance of SARS-CoV-2 phosphorylation sites.

  PMID 37847658 · PMC10628388 · Bioinformatics (Oxford, England) · 2023 · 8 claims · 6 setups

  Has reproduction · 67

  PSPred-ALE outperforms state-of-the-art SARS-CoV-2 phosphorylation site predictors (e.g. DeepIPs) and handcrafted feature-based methods in benchmarking comparisons

  a549 sequence same bulk pathway gene

• Immunoregulatory Roles of Tumor-Originated Pericytes Identified by Single-Cell Analysis in Glioblastoma.

  PMID 41001759 · PMC12713092 · Advanced science (Weinheim, Baden-Wurttemberg, Germany) · 2025 · 8 claims · 8 setups

  Has reproduction

  CD146 (MCAM) is a superior surface marker for sorting GBM pericytes, labeling nearly all PDGFRβ+ pericytes with strong specificity, unlike NG2 which labels fewer vessel-associated pericytes.

  facs-sorted frozen dissociated scrna-seq scrna-seq immunofluorescence fluorescence-activated

• Constructing eRNA-mediated gene regulatory networks to explore the genetic basis of muscle and fat-relevant traits in pigs.

  PMID 38594607 · PMC11003151 · Genetics, selection, evolution : GSE · 2024 · 8 claims · 8 setups

  Has reproduction · 100

  H3K27ac ChIP-seq and RNA-seq were used to construct eRNA expression profiles across multiple tissues in Enshi Black (ES) and Duroc pigs, revealing tissue-level eRNA regulatory landscapes

  longissimus muscle snps pig h3k27ac strand-specific starr-seq

• Single-cell multiome and enhancer connectome of human retinal pigment epithelium and choroid nominate causal variants in macular degeneration.

  PMID 41528844 · PMC12971065 · Cell reports · 2026 · 8 claims · 8 setups

  Full-text index only

  Generated a single-cell gene expression and chromatin accessibility (multiome) atlas of human RPE and choroid from control and AMD eyes

  rpe-choroid rpe arpe-19 scatac-seq joint h3k27ac activity-by-contact

• The evolution of gene regulation in mammalian cerebellum development.

  PMID 41610256 · PMC7618896 · Science (New York, N.Y.) · 2026 · 8 claims · 8 setups

  Full-text index only

  Combined single-nucleus RNA-seq and ATAC-seq atlases of cerebellum development were generated/integrated across six mammals (human, bonobo, macaque, marmoset, mouse, opossum)

  developing opossum marmoset cultures single-nucleus bulk gene

• 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

• Genetic alternative splicing regulation mapping of cartilage and synovium reveals tissue-specific mechanisms of joint-related traits.

  PMID 41820402 · PMC13121699 · Nature communications · 2026 · 8 claims · 8 setups

  Full-text index only

  Generated a splicing quantitative trait loci (sQTL) resource for human cartilage and synovium from over 200 donors

  degenerated synovium matched paired bulk genotyping differential

• Developing a thyroid cancer differentiation state classification system using deep residual networks and metabolic signature profiling.

  PMID 40993300 · PMC12460824 · NPJ digital medicine · 2025 · 8 claims · 8 setups

  Has reproduction

  Metabolic status is closely tied to tumor differentiation state, with distinct metabolic reprogramming occurring during thyroid cancer dedifferentiation

  thyroid fuscc gse193581 aggregated untargeted whole-exome transcriptomics