Division of Developmental Biology, the Roslin Institute, University of Edinburgh, Easter Bush Campus, Midlothian, EH25 9RG, UK.
BMC Bioinformatics. 2018 Nov 20;19(Suppl 14):409. doi: 10.1186/s12859-018-2377-x.
Transcription regulation is a major controller of gene expression dynamics during development and disease, where transcription factors (TFs) modulate expression of genes through direct or indirect DNA interaction. ChIP sequencing has become the most widely used technique to get a genome wide view of TF occupancy in a cell type of interest, mainly due to established standard protocols and a rapid decrease in the cost of sequencing. The number of available ChIP sequencing data sets in public domain is therefore ever increasing, including data generated by individual labs together with consortia such as the ENCODE project.
A total of 1735 ChIP-sequencing datasets in mouse and human cell types and tissues were used to perform bioinformatic analyses to unravel diverse features of transcription control. 1- We used the Heat*seq webtool to investigate global relations across the ChIP-seq samples. 2- We demonstrated that factors have a specific genomic location preferences that are, for most factors, conserved across species. 3- Promoter proximal binding of factors was more conserved across cell types while the distal binding sites are more cell type specific. 4- We identified combinations of factors preferentially acting together in a cellular context. 5- Finally, by integrating the data with disease-associated gene loci from GWAS studies, we highlight the value of this data to associate novel regulators to disease.
In summary, we demonstrate how ChIP sequencing data integration and analysis is powerful to get new insights into mammalian transcription control and demonstrate the utility of various bioinformatic tools to generate novel testable hypothesis using this public resource.
转录调控是发育和疾病过程中基因表达动态的主要控制器，转录因子（TFs）通过直接或间接的 DNA 相互作用调节基因的表达。ChIP 测序已成为在感兴趣的细胞类型中获得全基因组范围内 TF 占据的最广泛使用的技术，主要是由于建立了标准协议和测序成本的快速下降。因此，公共领域中可用的 ChIP 测序数据集的数量正在不断增加，包括各个实验室以及 ENCODE 项目等联盟生成的数据。
共使用了 1735 个在小鼠和人类细胞类型和组织中进行的 ChIP-seq 数据集来进行生物信息学分析，以揭示转录控制的多样化特征。1. 我们使用 Heat*seq 网络工具来研究 ChIP-seq 样本之间的全局关系。2. 我们证明了因子具有特定的基因组位置偏好，这些偏好在大多数情况下在物种间是保守的。3. 因子在启动子近端的结合在细胞类型间更保守，而远端结合位点在细胞类型间更具特异性。4. 我们确定了在细胞环境中优先共同作用的因子组合。5. 最后，通过将数据与 GWAS 研究中的疾病相关基因座进行整合，我们强调了将新型调节剂与疾病联系起来的价值。
总之，我们展示了如何整合和分析 ChIP 测序数据，以深入了解哺乳动物转录调控，并展示了使用此公共资源生成新的可测试假设的各种生物信息学工具的实用性。
