sunbymoon / ngsclean-1 Goto Github PK

Pipeline for NGS data analysis
==============================
The idea behind this script is have a simple way to go from raw FASTQ reads to 
BAM files. Briefly:

1) Read QC.
% ./filter_reads.sh file_1.fastq file_2.fastq

If file_2.fastq ommited assumes single end reads.


2) Read mapping. Not included here! Use your favorite read mapping program (eg. 
BWA, Bowtie2, SNAP, ...)


3) SNP QC
% ./filter_SNP.sh run_ID $N_IND $CHR bam_list.txt $REF_SEQ $ANC_SEQ

The output from this step is a BED file with the positions that passed the 
quality filters. This file can be used on all downstream analysis


4) Check VCF and SFS to assess overall quality of data.
% Rscript analyze_vcf.R myvariants.vcf output_file




Script Description
==================

filter_reads.sh      Script to perform raw reads QC. Namely, read filtering, 
		     quality trimming, adaptor trimming, PE read merging,...

filter_SNP.sh        SNP QC based on min/max depth, HWE, quality bias, strand 
		     bias, ... It calls the scripts below.

get_mut_bias.pl	     Script to plot the mutation frequency bias along the read. 
		     It was developed to deal with ancient DNA.

get_depth_thresh.R   R script to automatically define the upper and lower depth
		     coverage limits on SNPcleaner. It fits a mixture of Gamma 
		     and Neg-Binomial distribution to the empirical read coverage 
		     distribution.

SNPcleaner.pl        Filters sites from VCF files following a set of rules.

analyze_vcf.R	     Script to visualize aspects of data in VCF files.