Assays for transposase-accessible chromatin sequencing (ATAC-seq) provides an innovative approach to study chromatin status in multiple cell types. Moreover, it is also possible to efficiently extract differentially accessible chromatin (DACs) regions by using state-of-the-art algorithms (e.g. DESeq2) to predict gene activity in specific samples. Furthermore, it has recently been shown that small dips in sequencing peaks can be attributed to the binding of transcription factors. These dips, also known as footprints, can be used to identify trans-regulating interactions leading to gene expression. Current protocols used to identify footprints (e.g. pyDNAse and HINT-ATAC) have shown limitations resulting in the discovery of many false positive footprints. We generated a novel approach to identify genuine footprints within any given ATAC-seq dataset.Herein, we developed a new pipeline embedding DACs together with bona fide footprints resulting in the generation of a Predictive gene regulatory Network (PreNet) simply from ATAC-seq data. We further demonstrated that PreNet can be used to unveil meaningful molecular regulatory pathways in a given cell type.

中文翻译：

---

Prenet：来自 ATAC-SEQ 数据的预测网络

转座酶可及染色质测序 (ATAC-seq) 分析为研究多种细胞类型中的染色质状态提供了一种创新方法。此外，通过使用最先进的算法（例如 DESeq2）来预测特定样本中的基因活性，还可以有效地提取差异可及的染色质 (DAC) 区域。此外，最近表明，测序峰的小幅下降可归因于转录因子的结合。这些下降，也称为足迹，可用于识别导致基因表达的反式调节相互作用。当前用于识别足迹的协议（例如 pyDNAse 和 HINT-ATAC）已显示出局限性，导致发现许多假阳性足迹。我们生成了一种新方法来识别任何给定 ATAC-seq 数据集中的真实足迹。在这里，我们开发了一个新的管道，将 DAC 与真正的足迹一起嵌入，从而仅从 ATAC-seq 数据生成预测基因调控网络 (PreNet)。我们进一步证明了 PreNet 可用于揭示给定细胞类型中有意义的分子调控途径。