Questions about the output of CNVs and the use of assess_significance

[stbio-hbh]

Hello, I'm using the tool Control-FREEC to evaluate the CNVs in my data, but I have some doubts about the _CNVs file given in output. According to the manual information, the CNVs file should have 9 columns of results, but my result only output seven columns.

This is my configuration file and command line:

freec -conf freec_LPS4.txt -sample LPS4.bam

[general] samtools = /data_center_01/home/hubihao/anaconda3/envs/NB/bin/samtools bedtools = /data_center_01/home/hubihao/anaconda3/envs/NB/bin/bedtools sambamba = /data_center_01/home/hubihao/anaconda3/envs/NB/bin/sambamba chrLenFile = /data_center_01/home/hubihao/script/hg19_index/hg19_EBV_target.fasta.fai ploidy = 2 maxThreads = 4 breakPointThreshold = .8 window = 50000 chrFiles = /data_center_01/home/hubihao/script/hg19_chr/ outputDir = /data_center_01/home/hubihao/project/Neuroblastoma/FREEC_analysis/ sex = XX

[sample]

matefile = [samfile]

inputFormat = BAM mateOrientation = 0

[BAF] makePileup = /data_center_01/home/hubihao/project/Neuroblastoma/FREEC_test/hg19_snp142.SingleDiNucl.1based.bed fastaFile = /data_center_01/home/hubihao/script/hg19_index/hg19_EBV_target.fasta SNPfile = /data_center_01/home/hubihao/project/Neuroblastoma/FREEC_test/hg19_snp142.SingleDiNucl.1based.txt.gz

I looked up the previous issue, but I didn't find anything that needed to be adjusted in my config file, because I didn't have a control?

Also, I had some problems with the CNVs using assess_significance.R. cat assess_significance.R | R --slave --args LPS4.bam_CNVs LPS4.bam_ratio.txt

And then this is the script's error message:

Loading required package: GenomicRanges Loading required package: stats4 Loading required package: BiocGenerics

Attaching package: ‘BiocGenerics’

The following objects are masked from ‘package:stats’:

IQR, mad, sd, var, xtabs

The following objects are masked from ‘package:base’:

anyDuplicated, aperm, append, as.data.frame, basename, cbind,
colnames, dirname, do.call, duplicated, eval, evalq, Filter, Find,
get, grep, grepl, intersect, is.unsorted, lapply, Map, mapply,
match, mget, order, paste, pmax, pmax.int, pmin, pmin.int,
Position, rank, rbind, Reduce, rownames, sapply, setdiff, sort,
table, tapply, union, unique, unsplit, which.max, which.min

Loading required package: S4Vectors

Attaching package: ‘S4Vectors’

The following objects are masked from ‘package:base’:

expand.grid, I, unname

Loading required package: IRanges Loading required package: GenomeInfoDb Error in if (class(resultks) == "try-error") return(list(statistic = NA, : the condition has length > 1 Calls: KS In addition: Warning message: In ks.test.default(values, score(normals)) : p-value will be approximate in the presence of ties Execution halted

Pvalue couldn't be added in CNVs. How can I eliminate this error.

Thanks for your reply. : )

[hushaoqiang]

I also have this problem. How can I solve it? Thank you

[stbio-hbh]

I tried to modify assess_significance.R, and the output of the modified script seems to be correct. I posted the modified part in the script below:

Start with line 36 of the original script

numberOfCol=length(cnvs) wscore=c() kscore=c() for (i in c(1:length(cnvs[,1]))) { values <- score(subsetByOverlaps(ratio.bed,cnvs.bed[i])) resultw <- class(try(wilcox.test(values,score(normals)), silent = TRUE)) ifelse(resultw == "try-error", wscore <- c(wscore, "NA"), wscore <- c(wscore, wilcox.test(values,score(normals))$p.value)) resultks <- class(try(ks.test(values,score(normals)), silent = TRUE)) ifelse(resultks == "try-error",kscore <- c(kscore, "NA"),kscore <- c(kscore, ks.test(values,score(normals))$p.value)) } cnvs = cbind(cnvs, wscore, kscore)

End with line 50 of the original script

I hope you can try out the modified script, and I also hope to get some guidance from the author.

[hushaoqiang]

According to what you said, I have made modifications and have successfully run. Thank you