search

Illumina / REViewer

A tool for visualizing alignments of reads in regions containing tandem repeats
GNU General Public License v3.0
73 stars 14 forks source link

Segmentation fault (core dumped) #28

Closed hchetia closed 2 years ago

hchetia commented 2 years ago

Hi @egor-dolzhenko, This is more of a combined REViewer/EH question.

I am trying to generate a Reviewer graph for a locus from RNASeq data. It runs until the plot blueprint stage and then generates the segmentation fault error. [2021-11-08 17:42:57.730] [info] Loading specification of locus NIPA1 [2021-11-08 17:42:57.823] [info] Extracted 84 frags [2021-11-08 17:42:57.823] [info] Calculating fragment length [2021-11-08 17:42:57.824] [info] Fragment length is estimated to be 191 [2021-11-08 17:42:57.824] [info] Extracting genotype paths [2021-11-08 17:42:57.826] [info] Phasing [2021-11-08 17:42:57.830] [info] Found 2 paths defining genotype [2021-11-08 17:42:57.830] [info] Projecting reads onto haplotype paths [2021-11-08 17:42:57.831] [info] Projected 84 read pairs [2021-11-08 17:42:57.831] [info] Generating fragment alignments [2021-11-08 17:42:57.831] [info] Generated 84 fragment alignments [2021-11-08 17:42:57.831] [info] Assigning fragment origins [2021-11-08 17:42:57.833] [info] Found assignments for 84 frags [2021-11-08 17:42:57.833] [info] Generating plot blueprint Segmentation fault (core dumped)

Locus information from the ExpHunter run is as follows- "NIPA1": { "AlleleCount": 2, "Coverage": 14.291508923742562, "FragmentLength": 191, "LocusId": "NIPA1", "ReadLength": 151, "Variants": { "NIPA1": { "CountsOfFlankingReads": "()", "CountsOfInrepeatReads": "()", "CountsOfSpanningReads": "()", "Genotype": "0/0", "GenotypeConfidenceInterval": "0-714/0-714", "ReferenceRegion": "chr15:22786677-22786701", "RepeatUnit": "GCG", "VariantId": "NIPA1", "VariantSubtype": "Repeat", "VariantType": "Repeat"

If the lack of FR, IRR and SR is causing the error, then I wonder how the coverage for the loci was calculated.

Regards, Hasna

egor-dolzhenko commented 2 years ago

Hi Hasna,

Apologies about the error. Would it be permissible to share EH output for this repeat? If not, we can add additional debug information to REViewer output to diagnose the error without seeing your data.

Also, EH calculates coverage using read pairs that align within 1Kb of the repeat, so it is possible for coverage to be non-zero even if there are no spanning, flanking, or in-repeat reads.

Best wishes, Egor

hchetia commented 2 years ago

Hi Egor, we can add additional debug information to REViewer output to diagnose the error without seeing your data. Can we do this? Thanks for the explanation about the coverage. Best, Hasna

egor-dolzhenko commented 2 years ago

Sure! We are already working on this.