STAR: seg fault when using GeneFull_ExonOverIntron or GeneFull_Ex50pAS

Hi Alex,

Thanks so much for developing this awesome method. We have been trying to experiment with the different versions of GeneFull, but haven’t been able to get the GeneFull_ExonOverIntron or GeneFull_Ex50pAS versions to run. We use the same exact command that works perfectly for GeneFull, except changing the --soloFeatures (example below). The genome generation works fine, but once the 1st pass mapping begins, there is a seg fault (core dumped). Is there something that we’re perhaps missing?

Also - we were wondering how GeneFull and GeneFull_ExonOverIntron deal with multimapping. From my understanding, the main limitation of GeneFull (and reason why we’d want GeneFull_ExonOverIntron) is when there are cases like an intron of Gene A falling within an exon of Gene B. In that case, we’d want a read mapping to the overlapping part to be assigned to Gene B in the GeneFull_ExonOverIntron case, but it would be ambiguous in the simple GeneFull case. If multimapping were turned on, is the expected behavior to split the count between Gene A and B in the GeneFull case, but then the read is treated as uniquely mapped to Gene B in the GeneFull_ExonOverIntron case? (Apologies in advance if this is already explained somewhere, and thanks so much for your help!)

Joyce

STAR version: 2.7.10a compiled: 2022-03-11T01:08:49-0500 cn003.research.partners.org:/PHShome/jbk37/tools/STAR-2.7.10a/source STAR --genomeDir /data/srlab2/jkang/scHLA/personalized_final/Randolph2021_NewPanel/testing/HMN171215_NI_GeneFull_ExonOverIntron_gtf2_index/ --readFilesIn /data/srlab2/jkang/scHLA/personalized_final/Randolph2021_NewPanel/testing/data/HMN171215_NI_shuffled.bam --runThreadN 4 --runDirPerm All_RWX --soloUMIlen 10 --outFileNamePrefix /data/srlab2/jkang/scHLA/personalized_final/Randolph2021_NewPanel/testing/pers_results/HMN171215_NI_GeneFull_ExonOverIntron_gtf2/HMN171215_NI_GeneFull_ExonOverIntron_gtf2_ --soloType CB_UMI_Simple --soloCBwhitelist /apps/lib-osver/cellranger/1.3.1/cellranger-cs/1.3.1/lib/python/cellranger/barcodes/737K-august-2016.txt --soloCBmatchWLtype 1MM_multi_Nbase_pseudocounts --soloUMIdedup 1MM_CR --soloFeatures GeneFull_ExonOverIntron --soloBarcodeReadLength 0 --soloMultiMappers EM --twopassMode Basic --outSAMtype BAM SortedByCoordinate --outSAMunmapped Within --readFilesType SAM SE --readFilesCommand samtools view -F 0x100 --soloInputSAMattrBarcodeSeq CR UR --soloInputSAMattrBarcodeQual CY UY --readFilesSAMattrKeep None --outSAMattributes CR CY UR UY GX GN CB UB STAR version: 2.7.10a compiled: 2022-03-11T01:08:49-0500 cn003.research.partners.org:/PHShome/jbk37/tools/STAR-2.7.10a/source Apr 24 19:47:15 … started STAR run Apr 24 19:47:15 … loading genome Apr 24 19:47:43 … started 1st pass mapping

line 58: 265533 Segmentation fault (core dumped)

About this issue

  • Original URL
  • State: closed
  • Created 2 years ago
  • Comments: 17 (9 by maintainers)

Most upvoted comments

@alexdobin they all ran successfully with the new patch 🎉