cfDNApipe: a comprehensive quality control and analysis pipeline for cell-free DNA high-throughput sequencing data

Abstract

Motivation

Cell-free DNA (cfDNA) is gaining substantial attention from both biological and clinical fields as a promising marker for liquid biopsy. Many aspects of disease-related features have been discovered from cfDNA high-throughput sequencing (HTS) data. However, there is still a lack of integrative and systematic tools for cfDNA HTS data analysis and quality control (QC).

Results

Here, we propose cfDNApipe, an easy-to-use and systematic python package for cfDNA whole-genome sequencing (WGS) and whole-genome bisulfite sequencing (WGBS) data analysis. It covers the entire analysis pipeline for the cfDNA data, including raw sequencing data processing, QC and sophisticated statistical analysis such as detecting copy number variations (CNVs), differentially methylated regions and DNA fragment size alterations. cfDNApipe provides one-command-line-execution pipelines and flexible application programming interfaces for customized analysis.

Availability and implementation

https://xwanglabthu.github.io/cfDNApipe/.

Supplementary information

Supplementary data are available at Bioinformatics online.

1 Introduction

Recently, Cell-free DNA (cfDNA), which is present in body fluids like plasma and urine, has attracted significant attentions due to its non-invasive nature for disease diagnosis. The rapid development and application of high-throughput sequencing (HTS) technologies provide deep insight into the information contained by cfDNA. Researchers have found lots of signatures that can be derived from cfDNA HTS data, such as fragment size alterations (Cristiano et al., 2019; Mouliere et al., 2018), copy number variations (CNVs) (Jiang et al., 2015) and DNA methylation changes (Li et al., 2018). These features are promising to be applied in various scenarios such as prenatal diagnosis, early cancer detection and therapy response monitoring (Kilgour et al., 2020; Oliveira et al., 2020).

However, there still lacks an integrative and systematic tool for cfDNA HTS data processing and analysis. Researchers need to build their own pipeline with diverse tools and finely tuning the parameters to tackle cfDNA data, which is tedious and time-consuming. To fill this gap, we propose a highly integrated python package named cfDNApipe to deal with widely used cfDNA HTS data including both whole-genome sequencing (WGS) and whole-genome bisulfite sequencing (WGBS) data. cfDNApipe covers essential steps for cfDNA HTS data processing, including raw data pre-processing, quality control (QC) and state-of-the-art statistical analysis such as CNV detection and fragmentation profile analysis. The package is based on conda/bioconda and all its dependencies can be easily installed by one command line in shell.

2 Design and implementation

2.1 cfDNA data analysis workflow

The flowchart of cfDNApipe is shown in Figure 1.The entire workflow includes three major parts, data pre-processing, QC and statistical analysis.

Fig. 1.

The flowchart of cfDNApipe. PE, paired-end data

In the pre-processing part, cfDNApipe takes raw fastq files of WGS or WGBS data directly as the input. It removes adapters and then employs bowtie2 (Langmead and Salzberg, 2012) or Bismark (Krueger and Andrews, 2011) for read alignment of different data types. Post-alignment operations such as sorting, filtering as well as removing duplicates are executed seamlessly.

Multi-level QC functions are provided in cfDNApipe. Quality reports for raw reads and post-alignment features such as coverage at different genomic regions will be generated. It executes fragment length QC to examine whether samples are contaminated. The bisulfite conversion rate can also be calculated to estimate the technical noise for WGBS data.

The statistical analysis step performs comprehensive analysis to obtain multi-aspect signatures. In terms of genomic features, cfDNApipe provides a method to detect unbiased CNVs accompanied with a GC-bias correction approach (Jiang et al., 2015). Besides, it identifies single nucleotide variations as well as the genomic sequences from virus for WGS data. As for methylomics, cfDNApipe distinguishes differentially methylated regions between case and control samples. Deconvolution can be performed according to the methylome profiles to infer the fraction of each component as well as the tissue-of-origin. For fragmentomics, cfDNApipe integrates the nucleosome positioning inference method (Snyder et al., 2016) and alterations of short and long fragmentation ratio are detected at whole-genome scale (Cristiano et al., 2019). Orientation-aware cfDNA fragmentation (OCF) analysis (Sun et al., 2019) is also performed to detect the tissue-specific open chromatin signals. Detailed functions and usage guidelines of cfDNApipe are shown in the Supplementary Materials.

2.2 Implementation

For the convenience of users, cfDNApipe provides automatic downloading and index building for genome references. The whole pipeline can be executed simply with one Python command. An HTML report will be generated for visualization of analysis results, and outputs of each step are well-arranged in their own folder separately.

The pipeline is organized by a built-in dataflow mechanism with a strictly defined up- and down-stream data interface. Therefore, users can execute particular modules or customize their own pipeline through the application programming interface easily.

cfDNApipe provides parallel computing options for maximizing its efficiency. It also contains breakpoint detection to continue the rest of computational tasks after interruption.

3 Conclusion

Here, we propose cfDNApipe as a highly integrated cfDNA HTS data analysis toolkit. Users without sophisticated programming background can start it easily to discover the information in cfDNA HTS data comprehensively and efficiently. This toolkit will make the analysis of cfDNA HTS data much easier in a wide range of scenarios to promote the development and application of cfDNA-related researches.