soccin / bacterialcopynumber

Copy number pipeline for bacterial WGS

Bacterial copy number pipeline

Preprocessing

This steps need to be done once for each new genome

1: Create GC profile for genome

GENOMENAME=H37RvCO
mkNucFile.sh ${GENOMENAME}.genome ${GENOMENAME}.fasta

Where ${GENOMENAME}.genome is the BEDTOOLS format genome file. The output of this command will be the base profile of the genome. Convert this to the format the R code wants with

Rscript --no-save mkRDA.R \
	${GENOMENAME}_1000by100_TileBin.nuc.gz

which will create the file ${GENOMENAME}gcpct.rda.

2: Make tiling windows bed file

Use bedtools

WINDOW=100
bedtools makewindows \
	-g ${GENOMENAME}.genome \
	-w ${WINDOW} \
	>${GENOMENAME}.genome.Windows_${WINDOW}.bed

Count bins

Make a list of the BAMs to be processed in a file bamList2 and then count the number of reads per bin per sample with:

bedtools multicov \
	-bed ${GENOMENAME}.genome.Windows_${WINDOW}.bed \
	-bams $(cat bamList2) \
	>counts.txt

Normalize and process with DNAcopy

Use the R script seqDNAcopyGCCorrection.R to normalize and then segment the data with DNAcopy and plot the segmentation profile and write the segments file.

Will need to edit the lines at the top of the file depending on your project files.