Error in cnvDeletions(CONTIG_coverage, peakDistanceThreshold) : object 'DEL_DF' not found

Question

Error in cnvDeletions(CONTIG_coverage, peakDistanceThreshold) : object 'DEL_DF' not found

miniluphy opened this issue 2 years ago · 17 comments

Hello，Thank you for your tools.
I am running the following code in RStudio for CNproScan:

devtools::install_github("robinjugas/CNproScan")
library("CNproScan")
# Working directory with files
setwd("D:\\R\\CNproScan_project")
# File paths
fasta_file <- "reference.fasta"
bam_file <- "file.bam"
coverage_file <- "file.coverage"
bedgraph_file <- "mapp_genmap.bedgraph"

# For only GC normalization
DF <- CNproScanCNV(coverage_file, bam_file, fasta_file, 
                   GCnorm=TRUE, MAPnorm=FALSE, cores=4)

Then, the error occurred as follow:

Error in cnvDeletions(CONTIG_coverage, peakDistanceThreshold) : 
  object 'DEL_DF' not found

The link contains my files. Hope this helps.
https://drive.google.com/file/d/1RBgW4xe2SAS8kLyMdWV1lGkXn9qr0C14/view?usp=sharing
Thank you very much.

Answer 1 · 2022-11-08T09:27:42.000Z

Hi, thank you for using the package. ľll look into it in the evening and reply you.
Cheers!

Answer 2 · 2022-11-09T13:47:07.000Z

Hi, try to install the latest version commited to GitHub as
devtools::install_github("robinjugas/CNproScan",force = TRUE,ref = "main")

and then try to rerun it again. I made some exceptions handling so it could work. I got 23 events detected so you should have the same amount. Let me know if it works.

Cheers.

Answer 3 · 2022-11-09T17:51:51.000Z

Thank you for your reply.
I have updated my script according to your advice:

devtools::install_github("robinjugas/CNproScan",force = TRUE,ref = "main")
library("CNproScan")
# Working directory with files
setwd("D:\\R\\CNproScan_project\\CNproScan_project")
# File paths
fasta_file <- "reference.fasta"
bam_file <- "file.bam"
coverage_file <- "file.coverage"
bedgraph_file <- "mapp_genmap.bedgraph"

# For only GC normalization
DF1 <- CNproScanCNV(coverage_file, bam_file, fasta_file, 
                   GCnorm=TRUE, MAPnorm=FALSE, cores=4)

However, there is still an error about GC:
Error in if (GC_act == 0 | GC_tab[GC_act, "GC"] == 0 | is.na(GC_tab[GC_act, : argument is of length zero
It seems that the "GC_act" argument matters. Sorry to bother you again.
Thank you.

Answer 4 · 2022-11-09T20:22:56.000Z

That's okay, I am happy someone discovered new bug which should be repaired.
So, I debugged it and you can try reinstall repaired version again. It seems it works, I run it without GC normalization before thats why.

Answer 5 · 2022-11-10T15:37:04.000Z

Thank you most sincerely.
I have completed it and got 57 events with the GC normalization condition.
Thank you!

Answer 6 · 2024-03-04T08:48:49.000Z

Hello，Thank you for your tools.
I am running the following code in RStudio for CNproScan:
`devtools::install_github("robinjugas/CNproScan",force = TRUE,ref = "main")
library(CNproScan)
setwd("D:/educational/CNV")
fasta_file <- "./K+N.fasta"
bam_file <- "./N_VS_K+N.sort.bam"
coverage_file <- "./N_VS_K+N.coverage"
bedgraph_file <- "./mapp_genmap.bedgraph"

For only GC normalization

DF <- CNproScanCNV(coverage_file, bam_file, fasta_file, GCnorm=TRUE, MAPnorm=FALSE, ORICnorm=FALSE, cores=4)`

Then, the error occurred as follow:

> DF <- CNproScanCNV(coverage_file, bam_file, fasta_file, GCnorm=TRUE, MAPnorm=FALSE, ORICnorm=FALSE, cores=4) Error in CNproScanCNV(coverage_file, bam_file, fasta_file, GCnorm = TRUE, : object 'CNV_DF' not found
I followed you suggestion to install the latest version commited to GitHub as
devtools::install_github("robinjugas/CNproScan",force = TRUE,ref = "main")
but the error still occurred.
I have tried many ways to solve this error, I really need your help. Looking forward for your reply.

Answer 7 · 2024-03-04T08:51:44.000Z

Hello,
I will look into it.
Would you please share your files with me, over cloud or something?
Best regards Robin

Answer 8 · 2024-03-04T08:59:22.000Z

Uploading CNV.zip…

Answer 9 · 2024-03-04T09:01:43.000Z

Thank you very much for your reply despite your busy schedule.

Answer 10 · 2024-03-04T09:12:52.000Z

Hey, unfortunately the link doesnt work...

Answer 11 · 2024-03-04T09:28:36.000Z

Can I send it to you by email?

Answer 12 · 2024-03-04T09:35:19.000Z

Yeah, robinjugas at gmail.com

Answer 13 · 2024-03-04T12:07:49.000Z

This seems to be a special usecase the tool can't handle.
In FASTA you have 412 sequences.
In coverage file you have 162 sequences / contigs.
In BAM there is zero. I use BAM RNAME tag to scan the BAM file alignments and this one is empty.

So, you could process your files as recommended (should work I guess) or you can try ProcaryaSV. There you start with sequencing reads and everything is taken care of.

Did you use the recommended way? Or maybe, the condition I assume that
numberOfFastaContigs == numberOfCoverageContigs == numberOfBAMContigs
is wrong, and some contigs can have zero coverage.
But still, the BAM is an issue.

Best regards, Robin

Answer 14 · 2024-04-01T17:13:51.000Z

Oh, I’m back again, encountering the same issue as Sylviasmu. Following the advice, I’ve generated a series of files, which I’ve already sent to your email.

However, when I input the command: DF <- CNproScanCNV(coverage_file, bam_file, fasta_file, GCnorm=TRUE, MAPnorm=FALSE, ORICnorm=FALSE, cores=4), an error occurs: Error in CNproScanCNV(coverage_file, bam_file, fasta_file, GCnorm = TRUE, : object 'CNV_DF' not found.

Looking forward to your response. Thank you!

Answer 15 · 2024-04-07T08:05:05.000Z

I will look into it.

Answer 16 · 2024-04-07T18:56:17.000Z

Hello, thanks for using the package.
I uploaded the modified code, so reinstall the package:
devtools::install_github("robinjugas/CNproScan",force = TRUE,ref = "main")

Also, sort and index your BAM file...
And try to run it again. It's testing commit, so we will see.
However, your coverage shows there are only zeroes.

Answer 17 · 2024-04-15T12:37:18.000Z

Thank you for your response, the issue has been successfully resolved.