search

abyzovlab / CNVnator

a tool for CNV discovery and genotyping from depth-of-coverage by mapped reads
Other
206 stars 65 forks source link

calls not being generated #238

Open kghimire09 opened 3 years ago

kghimire09 commented 3 years ago

Hi,

I have bam files from WGS 10x coverage data. I used bwa to generate the bam file with GRCH38 as my reference and sorted it with gatk and removed duplicates with gatk as well. These are my lines of code: 1) ./cnvnator -root file.root -tree file.bam 2) cnvnator -rootfile.root -hist 1000 -fasta 3) cnvnator -root file.root -stat 1000 4) cnvnator -root file.root -partition 1000 5) cnvnator -root file.root -call 1000 > file.txt

I am only interested in the CNV calls and not the SNPs and indels. I generated a vcf file with the command: cnvnator2VCF.pl -prefix study1 -reference GRCh38 file.txt > final.vcf

I have a few questions.

  1. During the histogram generation do I need to provide individual chr1.fa files? if so where can I find one for GRCH38. Or do I just pull it out of my big fasta file.
  2. After I get the vcf file I annotate it using KaiWang's script but I do not find the gene that I know is a deletion. Do you have an annotation script that I can use?

Thank you

kghimire09 commented 3 years ago

@abyzov can you help please?

suvakov commented 3 years ago

Sorry for delay.

1) You can provide directory with separate files for chromosomes:

$ ./cnvnator -root file.root -his 1000 -d dir_with_genome_fa/

or whole genome gziped fasta file:

$ ./cnvnator -root file.root -his 1000 -fasta file_genome.fa.gz

2) I am not sure about annotation. We do have annotation implemented but only in CNVpytor using Ensembl REST API.

kghimire09 commented 3 years ago

Hi, Thank you for your response @suvakov. How would you recommend I prepare the bam files?