Development Of Data Analysis Software For ATAC-seq And Its Application In Obesity-induced Chronic Inflammation Research

Chromatin accessibility,also known as the openness of chromatin,reflects the transcriptional activity of chromatin,and is an important aspect to study the transcriptional regulation of gene expression.The research of chromatin accessibility is now playing a more and more significant role in the construction of the epigenetic landscapes for cell differentiation and development,and various diseases.Combined with the development of next generation sequencing technology,a variety of experimental methods for detecting the chromatin accessibility have been developed.Several tools or databases can be used to analyze the chromatin accessibility,however these tools can only solve partial aspects of the chromatin accessibility analysis.As ATAC-seq becomes the mainstream method for studying the chromatin accessibility,it is particularly necessary to develop a software specifically for analyzing ATAC-seq data.In order to meet this demand,we have integrated some existing tools and algorithms,such as the sequence alignment software Bowtie2,the signal peak calling software MACS2 and DESeq algorithm,etc.and built a software for ATAC-seq data analysis,named ATAC-pipe.Based on this software,we attempted to describe the chromatin accessibility of mononuclear-macrophages in mice under inflammatory conditions induced by obesity.Mononuclear-macrophages play an important role in innate immunity and also play a regulatory role in metabolic inflammation associated with metabolic diseases.During the progression of obesity,adipose tissue enlargement is often accompanied with enrichment of inflammatory monocyte and macrophages.In recent years,some researchers have begun to explore the regulatory mechanisms of obesity-induced inflammatory responses at the epigenome level.Nonetheless,our understanding of the regulatory mechanisms at the epigenome level in this process is still limited,and systematically studying of the regulatory networks with critical transcription factors and the comparison of transcriptional regulation between monocytes and macrophages have rarely been reported.Therefore,we performed this research,which can be mainly divided into two major parts.In the first part of the study,we first summarized some basic requirements of ATAC-seq data analysis through literature research.Starting from these basic needs,we successfully developed a comprehensive and easy ATAC-seq data analysis software,ATAC-pipe,by integrating some existing second-generation sequencing data analysis software and algorithms.The tool can perform preliminary analysis such as sequence alignment,quality control,file conversion,data visualization,signal peak search and signal strength statistics,Statistical analysis such as correlation analysis between samples,differential analysis between samples,and cluster analysis of differential open regions,as well as in-depth analysis such as transcription factor search,transcription factor footprint analysis,and nucleosome localization analysis.Using ATAC-seq data of peripheral blood CD4+ T cells in cutaneous T-cell lymphoma patients and normal individuals as sample data,we show the workflow and results of ATAC-pipe analysis.Further,we compared the performance of ATAC-pipe and existing ATAC-seq data analysis software in the analysis of ATAC-seq data of mouse cancer cell from orthotropic lung and liver metastatic site.We found that ATAC-pipe obtained similar quality control results to other analysis software but took only half the time.At the same time,ATAC-pipe can identify more obvious signal peaks,effectively separate samples from different sources and more significantly enrich related transcription factors.The software provides great convenience for studying epigenetic changes during disease development and progression.In the second part of the study,we used ATAC-pipe to analyze ATAC-seq data of peripheral monocytes and adipose tissue macrophages from high-fat food-induced obese mice and normal mice.First,we confirmed the quality of the obtained samples by pretreatment and quality control of sequencing data and simultaneously detecting the accessibility of promoter regions of monocyte and macrophage surface markers gene Cd11b,Ly6c and F4/80.Next,by principal component analysis,we found that the chromatin accessibility of monocytes and adipose macrophages in the inflammatory state was significantly changed compared with the normal state.Inflammatory factor gene Tnf and inflammatory markers gene Ccr2,Nos2 were also in more open state.Through systematic comparative analysis,we screened 2418 and 1126 more open regions in inflammatory monocytes and macrophages,respectively,which were mainly enriched biological function such as immune response,leukocyte activation,differentiation,and migration.To search transcription factors that activate in inflammatory states,we predicted transcription factors that were enriched in the new open regions.We successfully predicted some common transcription factors associated with monocyte or macrophage-mediated inflammatory responses,such as Sp1,AP-1,Nfkb,and others.At the same time,relatively new or less reported transcription factors such as Nfy,Klfl4,Bhlhe40,Ctcf,Yyl,Nrf,etc.were also discovered.Combining with the predicted binding sites of transcription factors,we filtered genes regulated by Bhlhe40 and Ctcf,and mapped their binding at predicted binding sites,respectively.Footprint analysis showed that Bhlhe40 and Ctcf showed more significant binding at their binding sites in the inflammatory state compared to the normal state.Finally,to fully describe the interaction between transcription factors in monocytes and macrophages in the inflammatory state,we constructed a transcription factor regulatory network.Some transcription factors,such as Nfkb,Klf14,Sp1,etc.,have a regulatory or regulated relationship with a variety of other transcription factors.At the same time,monocytes and macrophages exhibit different transcriptional regulatory networks.