cfMBD-seq

Author: Jinyong Huang

Introduction

The code within the LiangWanglab/cfMBD-seq repository is used for quick quality controls of cfMBD-seq data, including FASTQ to SAM by bowtie2, SAM to BAM by samtools, quality controls using R packages RaMWAS/MEDIPS, and call CpG annotations coverage using bedtools. Bash programming and pre-installation of software are required. Access to a high-performance computing cluster is also recommended.

Requirements

Computer running a Linux system (≥ 8 GB RAM) Cluster computing is HIGHLY recommended when working with FASTQ/BAM files
Modules: Bowtie2 (Version 2.4.2); reference genome-human/hg19; samtools (Version 1.11); bedtools (Version 2.28.0); Integrative Genomics Viewer. R (Version 4.0.3 or greater)
R Packages: BSgenome.hsapiens.UCSC.hg19; RaMWAS (Version 1.12.0); MEDIPS (Version 1.40.0).

Procedure

1. Bash to process raw data FASTQ files to BAM files.
   ./fastq_to_bam.sh
2. Quality control using RaMWAS, which generates summary QC (including duplicate rate%, non-CpG coverage, CpG coverage, and noise) and coverage by CpG density plot.
   Rscript RaMWAS.R
3. Quality control using MEDIPS, which generates wiggle files for visualization of rpkm normalized data. MEDIPS can also do saturation analysis and genome-wide correlation between samples.
   Rscript MEDIPS.R
4. Call CpG annotations coverage using bedtools and normalize the coverage using R.
   ./CpG_annotations.sh