Copy Number Variation Detection in Bacterial Circular Genome
git clone https://github.com/LinjieWu/CNV-BAC.git
R version >= 2.6
R package 'mgcv' and 'graph'
Python2
cd /PATH/TO/CNV-BAC
python setup.py
Usage: python CNV-BAC.py [options]
-h display help info.
-i bam file generate by BWA. (required)
-s Location of origin of replications seperated by comma (required), one can get this information from DoriC database http://tubic.org/doric/public/index.php/index.
-r fasta file of reference file. (required)
-m Mappability file (required)
-l penalty for BIC-seq2 (Default 1.5)
-o prefix of output file (Default CNV-BAC_tmp).
-b length of bin. (Defaut 1000)
-p subsample percentage for normalization. (Default 0.5)
-d the directory for output files. (Default /home/ruibinxi_pkuhpc/lustre1/ljwu/software/CNV-BAC/tmp)
--norm method for correction of normalization either Gaussian or Poisson (Default Gaussian)
--SV SV regions file (Default FALSE)
--PT Pvalue cutoff for CNVs detection (Default 0.001)
--RT log 2 copy number ratio cutoff for CNVs detection (Default 0.2)
Step 1: Generate mappability file
See https://github.com/LinjieWu/GenerateMappability
Step 2: Extract SV regions from Breakdancer or Delly results (optional)
/Path/to/CNV-BAC/ExtractSVRegions.py -i BreakDancer.out -o SV.out -s breakdancer
Step 3: Filter CNVs
python CNV-BAC.py -i input.bam -r reference.fa -s start_of_Oric_region,end_of_Oric_region -m example_mappability -l 2 -SV SV.out
*_mappability: Regions of reference with mappability 1 (calculated by gemtools).
*_GR.txt: spearman correlation, pearson correlation between log2 copy number ratio and distance to origin of replication.
*_seg.out: Segmentation result for CNV calling.
*_norm.out： Record the parameters for normalization.
*.norm.bin: Normalized, GC corrected and oriC corrected counts.
*.CNVs: Detected CNVs.