search

broadinstitute / ichorCNA

Estimating tumor fraction in cell-free DNA from ultra-low-pass whole genome sequencing.
GNU General Public License v3.0
160 stars 87 forks source link

Error while running with GenomeInfoDb_1.20.0 installed from tar #58

Closed patidarr closed 5 years ago

patidarr commented 5 years ago

Hi,

I just installed ichorCNA and other dependencies but GenomeInfoDb is not available for R >3.6, I installed it from tar file, but now I am getting this error.

Could you please suggest some fix? 

Sorting by decreasing chromosome size
Correcting Tumour
Error in (function (classes, fdef, mtable)  :
  unable to find an inherited method for function 'seqinfo' for signature '"RangedData"'
Calls: loadReadCountsFromWig ... seqlevels -> seqlevels -> seqlevels -> seqinfo -> <Anonymous>
In addition: There were 13 warnings (use warnings() to see them)
Execution halted

Thanks, Rajesh

gavinha commented 5 years ago

Hi @patidarr

Are you using the latest runIchorCNA.R. Can you check in your version of the script that this line is there?

https://github.com/broadinstitute/ichorCNA/blob/bc89d52dc5a02e77094e5ca770c691494aca0911/scripts/runIchorCNA.R#L164

patidarr commented 5 years ago

Thanks @gavinha for the quick reply.

I was using script from 0.2.0 release. I just tried from the master branch and getting following error: Removed 97 bins near centromeres. Applying filter on data... Correcting for GC bias... Correcting for mappability bias... Filtering low uniqueness regions with mappability score < 0.9 Removed 4 bins near centromeres. Determining gender...Gender female Error: RangedData objects are deprecated and the seqinfo() setter for RangedData objects is now defunct. Please migrate your code to use GRanges or GRangesList objects instead. See IMPORTANT NOTE in ?RangedData In addition: Warning messages: 1: In dir.create(paste0(outDir, "/", id, "/"), recursive = TRUE) : 'ichorCNA_OUT/134332-052-B' already exists 2: In regularize.values(x, y, ties, missing(ties)) : collapsing to unique 'x' values 3: RangedData objects are deprecated. Please migrate your code to use GRanges or GRangesList objects instead. See IMPORTANT NOTE in ?RangedData 4: RangedData objects are deprecated. Please migrate your code to use GRanges or GRangesList objects instead. See IMPORTANT NOTE in ?RangedData 5: RangedData objects are deprecated. Please migrate your code to use GRanges or GRangesList objects instead. See IMPORTANT NOTE in ?RangedData 6: RangedData objects are deprecated. Please migrate your code to use GRanges or GRangesList objects instead. See IMPORTANT NOTE in ?RangedData Execution halted

gavinha commented 5 years ago

Can you install ichorCNA directly from the master branch as well?

Also, can you show me the command you are using?

Thanks, Gavin

patidarr commented 5 years ago

Hi @gavinha,

Just figured out, that I should all the files from master as well. for 4.5X wgs, what settings would you recommend to call tumor fraction below 0.25%?

Thanks, Rajesh

patidarr commented 5 years ago

and I just got this error at the end: runEM: Total elapsed time: 0.12min. runViterbi: Segmenting and classifying Total ULP-WGS HMM Runtime: 0.91 min. Writing segments to ichorCNA_OUT/134332-052-B.seg Outputting to bin-level results to ichorCNA_OUT/134332-052-B.cna.seg Error in which(chrs == chrsNum[i]) : argument to 'which' is not logical Calls: plotSolutions ... plotCorrectionGenomeWide -> getGenomeWidePositions -> which Execution halted

gavinha commented 5 years ago

and I just got this error at the end: runEM: Total elapsed time: 0.12min. runViterbi: Segmenting and classifying Total ULP-WGS HMM Runtime: 0.91 min. Writing segments to ichorCNA_OUT/134332-052-B.seg Outputting to bin-level results to ichorCNA_OUT/134332-052-B.cna.seg Error in which(chrs == chrsNum[i]) : argument to 'which' is not logical Calls: plotSolutions ... plotCorrectionGenomeWide -> getGenomeWidePositions -> which Execution halted

I've seen this error come up in other issues as well. I'll work on fixing this soon.

Thanks for bringing up these issues. Best, Gavin

gavinha commented 5 years ago

Hi @gavinha,

Just figured out, that I should all the files from master as well. for 4.5X wgs, what settings would you recommend to call tumor fraction below 0.25%?

Thanks, Rajesh

I'm not sure if ichorCNA is able to provide the sensitivity for 0.25% at 4-5x WGS. In my own analysis, I was able to get down to that order of magnitude with about 40x WGS.

That being said, you can try to set --normal 0.995 and --ploidy 2. Also, keep with 1Mb bins.

Please refer to Parameter-tuning-and-settings for more details.

patidarr commented 5 years ago

Thanks @gavinha for the suggestions, with my data I can get a prediction of 1% tumor fraction, only if there is a way to validate this :)

Thanks for developing the tool, will wait for the fix for the error I am getting.

Rajesh

patidarr commented 5 years ago

hi @gavinha, could you please let me know how may normal samples you used for creating the PON?

patidarr commented 5 years ago

Hi @gavinha, I have the latest code, (as of y'day) with the fix for plotting but I am still getting this error: Total ULP-WGS HMM Runtime: 3.3 min. Writing segments to Results_ichoCNA_MoChaBaseline/19-32079-004-M_cfNA_Qubit.seg Outputting to bin-level results to Results_ichoCNA_MoChaBaseline/19-32079-004-M_cfNA_Qubit.cna.seg Error in which(chrs == chrsNum[i]) : argument to 'which' is not logical Calls: plotSolutions ... plotCorrectionGenomeWide -> getGenomeWidePositions -> which Execution halted

Thanks, Rajesh

gavinha commented 5 years ago

Hi @patidarr

Can you please share with me (gha@fredhutch.org) the .RData file so that I can debug? I am not able to reproduce this error with my runs. I'm starting to suspect it is related to R package versioning.

Thanks, Gavin