varTFBridge — processed FOODIE footprints and variant–TF binding–gene linkage tables

Processed datasets accompanying the manuscript Genome-wide maps of transcription factor footprints identify noncoding variants rewiring gene regulatory networks (Lin, Dong, Zhang, Xie, Jing, Zhao, Ma, Kang, Jiang, Xie & Zhao; Changping Laboratory). varTFBridge is an integrative framework that combines high-resolution FOODIE (single-molecule deaminase footprinting) transcription-factor footprints with UK Biobank whole-genome sequencing to identify causal noncoding variants and elucidate the regulatory mechanisms by which they rewire gene regulatory networks. Applied to 490,640 UK Biobank genomes across 13 erythroid traits, varTFBridge prioritises 113 high-confidence regulatory variants (104 common, 9 rare), encompassing 2,173 variant–TF binding–gene–trait linkages across 64 TFs and 108 genes, and recapitulates the causal variant rs112233623 acting through disruption of a GATA1/TAL1 co-binding motif at a CCND3 enhancer. Analysis code is publicly available at https://github.com/JasonLinjc/varTFBridge under the MIT License. This Zenodo deposit contains all processed datasets generated and used in the study: FOODIE footprints for K562 (188,484 footprints, median 19 bp) and GM12878 cell lines on GRCh38, plus matched ATAC-/DNase-derived footprints used as benchmark annotations. Common-variant pipeline outputs: per-trait variant–footprint overlap tables, VAR2TFBS predictions (Create / Disrupt / Increase / Decrease classifications), ABC-FP-Max variant-to-gene assignments, and AlphaGenome TF-ChIP scores for PIP > 0.7 variants. Rare-variant pipeline outputs: leave-one-variant-out (LOO) driver variant tables for footprint-wide significant burden hits, VAR2TFBS predictions, and ABC-FP-Max gene assignments. Final high-confidence variant–TF binding–gene–trait resource (113 variants, 2,173 linkages) together with interactive Plotly HTML reports for novel-mechanism discovery scoring. S-LDSC heritability enrichment tables for K562 FOODIE, DNase, and ATAC annotations across 28 UK Biobank blood traits. Trait metadata for the 13 erythroid traits and full set of manuscript Supplementary Data files (Supplementary Data 1–6, Supplementary Tables 1–2). See README.md and FILE_MANIFEST.md (rendered on this page and also included inside the archive) for a per-file description, provenance, and reproduction instructions.本数据集配套于论文《全基因组转录因子足迹图谱识别重编程基因调控网络的非编码变异》（作者Lin, Dong, Zhang, Xie, Jing, Zhao, Ma, Kang, Jiang, Xie & Zhao；单位：昌平实验室）。 varTFBridge是一款整合性分析框架，它将高分辨率FOODIE（单分子脱氨酶足迹检测技术）转录因子足迹数据与英国生物银行（UK Biobank）全基因组测序数据相结合，以识别致病性非编码变异，并阐明这些变异重编程基因调控网络的调控机制。本研究针对13种红系性状的490640份英国生物银行全基因组数据应用varTFBridge框架，最终筛选得到113个高置信度调控变异（其中104个为常见变异、9个为稀有变异），涵盖64个转录因子、108个基因的2173条变异-转录因子结合-基因-性状关联通路，并验证了致病变异rs112233623通过破坏CCND3增强子区域的GATA1/TAL1共结合基序发挥调控功能。本研究的分析代码已基于MIT许可证公开部署于https://github.com/JasonLinjc/varTFBridge。 本Zenodo存档包含本研究中生成并使用的全部已处理数据集： 1. K562细胞系（188484个足迹，中位长度19bp）与GM12878细胞系在GRCh38参考基因组上的FOODIE足迹数据，以及作为基准注释的匹配ATAC-/DNase足迹数据； 2. 常见变异分析流程输出结果：逐性状变异-足迹重叠表、VAR2TFBS预测结果（创建/破坏/增强/减弱分类）、ABC-FP-Max变异-基因关联分配结果，以及PIP>0.7变异的AlphaGenome转录因子染色质免疫沉淀（ChIP）评分数据； 3. 稀有变异分析流程输出结果：全足迹范围显著的变异载荷命中的留一变异（LOO）驱动变异表、VAR2TFBS预测结果，以及ABC-FP-Max基因关联分配结果； 4. 最终高置信度变异-转录因子结合-基因-性状关联资源集（含113个变异、2173条关联），以及用于新机制发现评分的交互式Plotly HTML报告； 5. K562细胞系FOODIE、DNase及ATAC注释在28项英国生物银行血液性状中的连锁不平衡得分回归（S-LDSC）遗传力富集表； 6. 13种红系性状的性状元数据，以及论文全部补充数据文件（补充数据1-6、补充表1-2）。 详细的单文件说明、数据溯源与复现步骤，请参阅本页面及存档内附带的README.md与FILE_MANIFEST.md文档。