Queue BaseRecalibrator - RScript file name error

flescai

Hi there, I was trying to debug an error in the RScript generated after base recalibration, while running the DataProcessingPipeline.scala (run as it is). I get the following debug output

 [...]
 Error in file(filename, "r", blocking = TRUE) : 
   cannot open the connection
 Calls: source ... eval.with.vis -> eval.with.vis -> gsa.read.gatkreport -> file
 In addition: Warning messages:
 1: In file(filename, "r", blocking = TRUE) :
   cannot open file '/SAN/scratch3/sample378_TTAGGC_L004_R1_001.fastq.pre_recal.table.recal': No such file or directory
  Execution halted

no file ending with "recal.table.recal" exists, but the file "recal.table" does exist. I couldn't find any step in the scala script where a ".recal" is added to "recal.table", nor a specific trait or class referring to the RScript itself, as I understand it's part of the walker BaseRecalibrator.

is this a small bug in the name handling, or am I doing something wrong somewhere?

thanks, Francesco

Geraldine_VdAuwera

Try upgrading to the latest version. I've checked the code and the name handling looks clean in both the script and the walker.

Also, make sure to have the same versions for GATK and Queue.

Geraldine_VdAuwera

Glad to hear it worked out for you, and thanks for reporting your solution details for other users' benefit!

Geraldine_VdAuwera

Hi Francesco, what version are you running?

flescai

I didn't write it. it's Queue-2.1-8-g5efb575

flescai

Hi Geraldine! I didn't launch again the entire DataProcessingPipeline on my batch. The Calling pipeline however generates a beautiful PDF now.

I tried to relaunch the DataProcessing script with -keepIntermediates, to have it generating the report, but unfortunately this way deleted the jobreport.txt (no data in) and therefore gave an error in the R script.

Is there somewhere the R script to generate the PDF graphs that I could apply to previously generated jobreport.txt files without launching the scala script?

thanks!

flescai

Oh, I found it :-) for the benefit of other users it's located in the github repository, under /public/R/scripts/org/broadinstitute/sting/queue/util

then it's sufficient to launch

 Rscript queueJobReport.R DataProcessingPipeline.jobreport.txt DataProcessingPipeline.jobreport.pdf

and you have it formatted, it works nicely! cheers, F