Run times with CNVnator

[vivek_]

Hi,

If anyone has run CNVnator before for cnv detection can you give me an idea of the approximate run time for the steps?

I'm trying to run the histogram generation step with a histogram size of 100bp and its been running without any result for close to 6 hours.

I'm trying to run this on a single chromsome and I tried with both wgs and exome samples with no luck.

Thanks for the help!

[gprakhar]

Hello,

What all commands did you use?


For us it took,
22 mins to run the CNVnator tool on a single 3.2 GHz Intel I3 processor with a 4GB DDR3 (1333 Mhz), 2TB Western Digital hard disk and 64 MB Buffer RAM on a Ubuntu 10.10 64 bit operating system.
Input data: 30x coverage, Chromosome 1, Illumina paired end, simulated data

--
pg

[vivek_]

I'm using the following command to generate histograms:

cnvnator -root chr22.root -genome name hg19 -chrom 22 -d human_reference/ -his 100

I'm running CNVnator v0.2.5 on a computing cluster.

[vivek_]

Also when you say that you used simulated data, how big was the bam file and was it an exome or whole genome sequence? I tried increasing the bin size, downloading an older version, checking on different chromosomes and nothing seems to be helping this case.

I have the same issues on the 1000 genomes bam files when I tried to check with those.

It also prints "Making directory bin_100" but I don't see this in the working directory.

[gprakhar]

cnvnator -root chr22.root -genome name hg19 -chrom 22 -d human_reference/ -his 100

I think the "-genome" is used incorectly, the word "name" should not be there. Only the name of the genome should come after the "-genome" option.

Also the "-chrom" and "-genome" options are something you should skip and try.

Also when you say that you used simulated data, how big was the bam file and was it an exome or whole genome sequence?

It was WGS data, for chr1. I dont remember the size of the BAM, it was gnerated using dwgsim.

We have version 0.2.2

I hope this works for you.

--
pg

[vivek_]

Yeah I just wrote the wrong command there for genome option.

Anyways the software worked and the issue was having more than 60 bases per line in fasta input files. The software fails when there are no line breaks within fasta files.

[IsmailM]

Multi-Thread CNVnator?

Hi all,

Is there a way to make CNVnator use multiple cores when it runs? I am on a 64 core Linux SGE running CentOS. I am on the partitioning step and it is taking a long time for one chromosome and I have ~20 more to go so more speed would definitely help this process. Please give me any suggestions about cnvnator or information about multicore running processes.

[sachin]

Yeah I just wrote the wrong command there for genome option.

Anyways the software worked and the issue was having more than 60 bases per line in fasta input files. The software fails when there are no line breaks within fasta files.

Thank you vivek..I also encounter this problem..solved by your "60 bases per line in fasta input files" post