Sign inSign up

benjamindn/evar

By benjamindn

Updated about 8 years ago

Variants calling by de novo assembly

Image
1

244

benjamindn/evar repository overview

evar

Variants calling pipieline by de novo assembly using SGA, BWA, Samtools and R script for pool sequencing


files needed

reference genome in .fa 
genomic annotation in .gff
single-end one file fastq.gz or double-end two files R1.fastq.gz R2.fastq.gz

! your reference can't have any N in place of A, T, G or C --> sript will crash

default parameters

# -r = remove-adapter-fwd (str)
# -c = remove-adapter-rev (str)
SGA_PREPROC_FLAGS:= -r AGATCGGAAGAGCACACGTCTGAACTCCAGTC -c AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTG

# -x = min x kmer coverage in a read (int)
# -k = size of kmer (int)
SGA_FILTER_FLAGS:= -x15 -k61

# -m = minimum overlap required between two reads to merge (int)
SGA_FMMERGE_FLAGS:= -m61

# -k = size of kmer (int)
SGA_KMERCOUNT_FLAGS:= -k61

# -t = number of thread (int)
SGA_THREAD:= -t4

# -m = minimum overlap required between two reads to merge (int)
SGA_OVERLAP_FLAGS:= -m61

Create Makefile.config in your working directory with the following template

REF = ref.fa
REF_GFF = ref.gff

SGA_PREPROC_FLAGS:= -r AGATCGGAAGAGCACACGTCTGAACTCCAGTC -c AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTG
SGA_FILTER_FLAGS:= -x15 -k61
SGA_FMMERGE_FLAGS:= -m61
SGA_KMERCOUNT_FLAGS:= -k61
SGA_THREAD:= -t4
SGA_OVERLAP_FLAGS:= -m61
Some parameters of rules were not in parameters but can be change in the original makefile
For more details on parameters see the help for SGA, BWA and samtools

docker command to run the application

```
docker run --rm -v $(pwd):/export benjamindn/evar-call fastq.all
```

IGV visualisation

#all your reads are mapped
.bwamem.bam

#De novo assembly 
.preproc.filter.pass.kmercount.bwamem.bam

#Region of interest by kmer
.preproc.filter.pass.rmdup.merged.bwamem.bam

The best way to visualize your result is to put together the de novo assembly and the region of interest in IGV

Example

Create a directory data 
Put your fastq.gz, reference.fa and annotation.gff in this directory
Go in the data directory
Run this command 
```
docker run --rm -v $(pwd):/export benjamindn/evar-call fastq.all
```
Wait for all the processus (more your genome to analyse is big more times is needed)
At the end you get the three files for an IGV visualisation, a file in .tsv you can open with a tablesoftware, a repartition plot of mutation and an interactive assembly network to see the de novo assembly.  

Table explanation

work in progress

Tag summary

Content type

Image

Digest

Size

984.9 MB

Last updated

about 8 years ago

docker pull benjamindn/evar