From reads to insight: a hitchhiker's guide to ATAC-seq data analysis.

Yan F, Powell DR, Curtis DJ, Wong NC. · 2020

Read From reads to insight: a hitchhiker's guide to ATAC-seq data analysis. by Yan F, Powell DR, Curtis DJ, Wong NC. (2020) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc6996192.txt.

Introduction

Yan and co-authors write a practical guide to analysing ATAC-seq data, from raw reads to insight. Their review appeared in Genome Biology. ATAC-seq maps which parts of the genome sit open and active. It is now a standard tool in chromatin biology, but no full review of its analysis tools existed.

The guide walks through each major step. Early stages cover quality checks and aligning reads to the genome. Core analysis finds the open peaks. Advanced steps compare peaks between conditions, annotate them, hunt for binding motifs, and read protein footprints and nucleosome positions. The authors also cover rebuilding gene control networks by joining ATAC-seq with other data types. Single-cell ATAC-seq and its special tool needs close the review.

Catalog record

CREATORYan F, Powell DR, Curtis DJ, Wong NC.
YEAR2020
COLLECTIONPAPERS
FORMATTXT
FILE SIZE84 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC6996192

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

From reads to insight: a hitchhiker’s guide to ATAC-seq data analysis By Feng Yan, David R. Powell, David J. Curtis, Nicholas C. Wong Published in: Genome Biology DOI: 10.1186/s13059-020-1929-3 ABSTRACT Assay of Transposase Accessible Chromatin sequencing (ATAC-seq) is widely used in studying chromatin biology, but a comprehensive review of the analysis tools has not been completed yet. Here, we discuss the major steps in ATAC-seq data analysis, including pre-analysis (quality check and alignment), core analysis (peak calling), and advanced analysis (peak differential analysis and annotation, motif enrichment, footprinting, and nucleosome position analysis). We also review the reconstruction of transcriptional regulatory networks with multiomics data and highlight the current challenges of each step. Finally, we describe the potential of single-cell ATAC-seq and highlight the necessity of developing ATAC-seq specific analysis tools to obtain biologically meaningful insights. Electronic supplementary material Supplementary information accompanies this paper at 10.1186/s13059-020-1929-3. INTRODUCTION Mammalian DNA is highly condensed through three major hierarchical scales; the first is the nucleosome which then wraps into chromatin leading to the third hierarchy, the chromosome [1–6]. Chromatin can dynamically switch between transcriptionally active euchromatin and inactive heterochromatin [7, 8]. All three scales of DNA condensation and their interplay contribute to gene regulation. Recent gene regulation studies have focused on epigenetics, and through the advances of high-throughput sequencing technologies, various assays have been developed to decipher the epigenetic landscape. These include Assay of Transposase Accessible Chromatin sequencing (ATAC-seq) [9, 10], DNase I hypersensitive sites sequencing (DNase-seq) [11–13] and Formaldehyde-Assisted Isolation of Regulatory Elements sequencing (FAIRE-seq) [14], all of which interrogate chromatin accessibility; Chromatin Immuno-Precipitation sequencing (ChIP-seq) which measures transcription factor (TF) binding [15–17] and histone modifications [18, 19]; and Micrococcal Nuclease sequencing (MNase-seq) which detects nucleosome positioning and occupancy [20, 21]. Detailed procedures of these assays are out of scope of this review and discussed in detail elsewhere [22]. Among assays designed for detecting chromatin accessibility, ATAC-seq has gained particular popularity since first described in 2013. An exponential increase of curated ATAC-seq datasets and publications indicates its value in a wide spectrum of biological questions (Fig. 1a), such as depicting enhancer landscapes in healthy mammalian tissue and cell types [23–25], studying accessibility changes between normal hematopoiesis and leukemia [26, 27], as well as the chromatin state within schizophrenia patients and the Cancer Genome Atlas (TCGA) pan-cancer cohort [28, 29]. A schematic diagram of this cutting-edge technology in fundamental and translational research is shown in Fig. 3a. Briefly, ATAC-seq incorporates a genetically engineered hyperactive Tn5 transposase that simultaneously cuts open chromatin leaving a 9-bp staggered nick and ligates high-throughput sequencing adapters to these regions. During this process, the nick is repaired, leaving a 9-bp duplication [30, 31]. Paired-end sequencing is then performed to facilitate higher unique alignment rates of these open regions [32]. Fig. 1Overview of ATAC-seq datasets increase and sample output for pre-analysis and advanced analysis. a The number of ATAC-seq datasets, ATAC-seq publications, DNase-seq datasets, FAIRE-seq datasets, and MNase-seq datasets in PubMed from 1 Jan 2013 to 1 Oct 2019. b Typical fragment size distribution plot shows enrichment around 100 and 200 bp, indicating nucleosome-free and mono-nucleosome-bound fragments. c Typical TSS enrichment plot shows that nucleosome-free fragments are enriched at TSS, while mono-nucleosome fragments are depleted at TSS but enriched at flanking regions. d Typical peak annotation pie chart shows that more than half of the peaks fall into enhancer regions (distal intergenic and intronic regions), and only around 25% of the peaks are in promoter regions. TSS: transcription start site The hyperactivity of Tn5 transposase makes the ATAC-seq protocol a simple, time-efficient method that requires 500–50,000 cells [9]. The sensitivity and specificity are comparable to DNase-seq but superior to FAIRE-seq where both methods require millions of cells as input material [9]. Because ATAC-seq does not involve rigorous size selection during library preparation, it can also identify nucleosome positions using fragments representing nucleosome monomer and multi-mers [9]. Recently, single-cell ATAC-seq (scATAC-seq) has been described, using fluorescence-activated cell sorting (FACS), microfluidic, and nano-well-based approaches [33–35]. ScATAC-seq can be applied in multiple situations including clinical specimens and developmental biology to study the heterogenous cell populations at single-cell resolution [23, 29]. Despite the simplicity and robustness of ATAC-seq, a major impediment exists as there are few bioinformatic analysis tools developed specifically for ATAC-seq data [32, 36]. Analysis tools used in ChIP-seq and DNase-seq have been applied to ATAC-seq assuming similar data characteristics [37]. However, this assumption has not been evaluated systematically. The major focus of this review is to discuss current resources for ATAC-seq analysis. We aim to provide an annotated guide for ATAC-seq data analysis instead of an exhaustive collection of tools. Previous reviews regarding ATAC-seq data analysis have focused mainly on peak callers and modeling regulatory networks [37, 38], but a systematic review covering major parts of ATAC-seq data analysis is urgently needed. This review will cover the four most important steps listed in the flowchart (Fig. 2). These include (1) pre-analysis (quality control (QC) and alignment), (2) core analysis (peak calling), (3) advanced analysis at the level of peaks, motifs, nucleosomes, and TF footprints, and (4) integration with multiomics data to reconstruct regulatory networks. These steps will allow researchers to conduct robust analysis on ATAC-seq data and generate more biological meaningful results. Finally, we will present the challenges and opportunities of ATAC-seq analysis and scATAC-seq. Fig. 2Roadmap of a typical ATAC-seq analysis. Four major steps are listed, including pre-analysis, core analysis, advanced analysis, and integration with multiomics data. Pre-analyses include pre-alignment QC, alignment and post-alignment processing, and QC. Core analysis includes peak calling. Advanced analyses include peak, motif, footprint, and nucleosome analysis. Multiomics data integration includes integration with ChIP-seq and RNA-seq data and regulatory network reconstruction. Text in each box emphasizes the important considerations in each analysis step. We suggest researchers start with FastQC, trimmomatic, and BWA-MEM for pre-analysis, MACS2 for peak calling, csaw for peak differential analysis, ChIPseeker for annotation and visualization, MEME suite for motif detection and enrichment, HMMRATAC for nucleosome detection, HINT-ATAC for footprint analysis, and PCEA for regulatory network reconstruction with RNA-seq. QC: quality check; TSS: transcription start site; TF: transcription factor; DEG: differentially expressed gene PRE-ANALYSIS: QUALITY CONTROL AND ALIGNMENT The first step of ATAC-seq analysis involves pre-alignment QC, read alignment to a reference genome, and post-alignment QC and processing (Fig. 2a) [32]. Pre-alignment quality control The pre-alignment QC and read alignment steps are standard for most high-throughput sequencing technologies. For example, FastQC [39] can be used to visualize base quality scores, GC content, sequence length distribution, sequence duplication levels, k-mer overrepresentation and contamination of primers and adapters in the sequencing data. An overall high base quality score with a slight drop towards the 3′ end of sequencing reads is acceptable. No obvious deviation from expected GC content and sequence read length should be observed. Moreover, the metrics should be homogeneous among all samples from the same experimental batch and sequencing run. Currently, due to the ubiquitous use of Illumina’s Nextera library for ATAC-seq, overrepresentation of Nextera sequencing adapters is often observed and should be removed for accurate read alignment. Most adapter removal tools employ different variations of dynamic programming, such as cutadapt [40], AdapterRemoval v2 [41], Skewer [42], and trimmomatic [43] all requiring input of known

Read the complete text