Tool:cnv_facets: somatic Copy Number Variant calling using the facets package

Question

[dariober]

I'd like to bring to your attention cnv_facets, a command line tool for detecting copy number variant (CNV), based on the facets package (Shen R and Seshan VE, Nucleic Acid Res, 2016).

Recently I've been using the facets package for detecting CNVs in tumour-normal sample pairs. I think facets works great but facets being a collection of R functions (plus an external program to prepare the input), it requires manual interaction from the user and so it doesn't play well with a streamlined analysis pipeline.

cnv_facets wraps all the required steps, from BAM files to VCF output and summary plots, in a simple command line call:

cnv_facets.R -t tumor.bam -n normal.bam -vcf snps.vcf.gz -o tex

Installation can be done under regular user permission (no admin rights required) as:

git clone https://github.com/wwcrc/cnv_facets.git
cd cnv_facets
bash setup.sh /dir/on/your/path

For more information, check out the README file. Feel free to use and report any bugs, comments, questions.

[sdjnqq106]

Could you show me your code of 'cnv_facets.R' script. I found some errors when I run this command:

cnv_facets.R -t tumor.bam -n normal.bam -vcf snps.vcf.gz -o tex

# Load required packages
library(facets)
library(argparse)

# Define the run_facets function
run_facets <- function(tumor_bam, normal_bam, vcf_file, output_prefix) {
    rcmat <- readSnpMatrix(vcf_file, tumor_bam, normal_bam)
    preproc <- preProcSample(rcmat)
    proc <- procSample(preproc, cval = 150)
    fit <- emcncf(proc)

    pdf(paste0(output_prefix, "_facets.pdf"))
    plotSample(proc, fit)
    dev.off()

    write.table(fit$cncf, paste0(output_prefix, "_facets_cncf.txt"), sep = "\t", quote = FALSE, row.names = FALSE)
    write.table(fit$segs, paste0(output_prefix, "_facets_segs.txt"), sep = "\t", quote = FALSE, row.names = FALSE)
}

# Argument parsing
parser <- ArgumentParser(description = 'Run FACETS for CNV calling')
parser$add_argument('-t', '--tumor', required = TRUE, help = 'Tumor BAM file')
parser$add_argument('-n', '--normal', required = TRUE, help = 'Normal BAM file')
parser$add_argument('-vcf', '--vcf', required = TRUE, help = 'SNPs VCF file')
parser$add_argument('-o', '--output', required = TRUE, help = 'Output prefix')
args <- parser$parse_args()

#test function
run_facets(<normal_bam>, <tumour_bam>, <vcf_file>, <output_prefix>)

# Run the function
run_facets(args$tumor, args$normal, args$vcf, args$output)

The error happens on when I run the run_facets function and it is as follows:

Error in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,  : 
  scan() expected 'a real', got 'Description="Allfilterspassed">'
5.
scan(file = file, what = what, sep = sep, quote = quote, dec = dec,
nmax = nrows, skip = 0, na.strings = na.strings, quiet = TRUE,
fill = fill, strip.white = strip.white, blank.lines.skip = blank.lines.skip,
multi.line = FALSE, comment.char = comment.char, allowEscapes = allowEscapes, ...
4.
read.table(file = file, header = header, sep = sep, quote = quote,
dec = dec, fill = fill, comment.char = comment.char, ...)
3.
read.csv(filename, stringsAsFactors = FALSE, colClasses = rep(c("character",
"numeric", "character", "numeric"), c(1, 1, 2, 8)))
2.
readSnpMatrix(vcf_file, tumor_bam, normal_bam)
1.
run_facets("LN88_chr19.sorted.bam", "Tail88_chr19.sorted.bam",
"Tail88_chr19.variants.vcf", "cnv_output")

[dariober]

I'm confused about how you are using cnv_facets. You are showing the command to be executed on the terminal (cnv_facets.R -t ...). This is how cnv_facets.R is supposed to be used and the command looks about right. But then you show some R commands with functionrun_facets redefined in such way that it cannot work (readSnpMatrix is not meant to read vcf files, hence the error you get). So what is it your trying to do...?