1
$\begingroup$

I want to find my copy number variants. I got DNA testing from the Nebula Genomics and received the raw datas in .cram, .fastq, .vcf formats. Are there any simple and easy method to call my copy number variants from one of these files? I tried GATK, but I was unable to collect Read Counts from 1000 genomes datas.

$\endgroup$
0

1 Answer 1

0
$\begingroup$

Yes and no. Yes, there are tools that can do this, but no, they do not work very well. Short read NGS sequencing just isn't the right tool to find large structural variants like CNVs. As a rule of thumb, you can expect around 30% accuracy when compared to other experimental approaches such as MLPA or array CHG. That said, yes, there are indeed various tools that can do this. My personal favorite is delly because it is accurate, actively maintained and since it is written in C it is quite fast.

Once you have downloaded and installed it, you'll likely need to convert your cram to bam as I don't think it can deal with cram:

samtools view -b foo.cram > foo.bam

Then you can run delly on it with (assuming your cram was aligned against hg38):

delly call -x path/to/delly/excludeTemplates/human.hg38.excl.tsv \
   -g path/to/hg38.fa -o out.bcf foo.bam

You will also need to download a copy of the human genome fasta file that Nebula used to generate your cram. You can ask them where to get one, and change path/to/hg38.fa with the path to that file.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.