Processing error

[zuta-osa77]

Hello,

I am testing VT and getting through the STAR alignment stages fine but running into what appear to be issues with doPar parallelizing R filtration of the 'viral bams'. As context, I only aligned to one viral reference (ie the index was host + single viral reference). I'm copying below the stdout and stderr for the entire Rscript call to Viral_Track_scanning.R. Any thoughts?

Also, given that I have gone through alignment and creation of the chromosome specific bams, is there a way to comment out some code in the scan to restart at the failure point (instead of redoing alignment again and again every time something downstream fails)?

Thanks! zuta

##stdout#########################################################################
Loading of the libraries.... ... done ! 1 Fastq files are going to be processed ! 
Mapping MIME23_700X3_8_R2_001_extracted.trimmed_2.fastq file 
Jan 25 07:46:26 ..... started STAR run
Jan 25 07:46:26 ..... loading genome
Jan 25 07:46:42 ..... started 1st pass mapping
Jan 25 08:29:38 ..... finished 1st pass mapping
Jan 25 08:29:39 ..... inserting junctions into the genome indices
Jan 25 08:32:55 ..... started mapping
Jan 25 09:24:04 ..... finished mapping
Jan 25 09:24:10 ..... started sorting BAM
Jan 25 09:53:24 ..... finished successfully
Mapping ofMIME23_700X3_8_R2_001_extracted.trimmed_2.fastq done ! 
All fastq files have been mapped successfully 
Starting the BAM file analysis 
Indexing of the bam file for MIME23_700X3_8_R2_001_extracted.trimmed_2 is done 
Computing stat file for the bam file for MIME23_700X3_8_R2_001_extracted.trimmed_2 is done 
Checking the mapping quality of each virus... 
Export of the viral SAM file done for MIME23_700X3_8_R2_001_extracted.trimmed_2 

##stderr########################################################################
Warning message:
Output directory does not exist ! Creating it ! 
Error in unserialize(socklist[[n]]) : error reading from connection
Calls: %dopar% ... recvOneData -> recvOneData.SOCKcluster -> unserialize
Execution halted

[PierreBSC]

Hi Zuta,

Thank you a lot for your interest in Viral-Track ! Indeed it really seems that the parallelisation complitely failed during the QC generation step ... Could you tell me :

• what are the configuration of the cluster/workstation you are using
• what are the files contained in the output directory at the end of your run.

During the development of Viral-Track I usually put a comment character on line :

• 242 ( system(STAR_mapping_command))
• 327 ( system(temp_export_bam_command)) Tell me if the trick works and then we can go into more details !

Best

Pierre

[zuta-osa77]

hi Pierre,

Before answering these I just checked the viral bam, found < 50 reads had aligned, and then just concatenated the R2 10X fastq files and ran kallisto quant on the viral index and found nearly the same result. It looks as though I have a poor signal generally so I'm going to tackle that prob;em first. But thanks for getting back to me -- I really think this tool will be useful when the virus is actually infecting the cell (as opposed to be circulating in eg plasma/serum).

best zo

[PierreBSC]

Hi Zuta,

No problem ! Good luck with your experiment then :) I will close the issue and you can open a new one if needed later !