STAR 2.7.10a_alpha_220601 terminates with signal 11, Segmentation fault

[wlkath]

Hi,

I'm running STAR version=2.7.10a_alpha_220601 on some 10x 3' data.

STAR_2.7.10a --runThreadN 20 --soloType CB_UMI_Simple --soloUMIlen 12 --soloCellFilter EmptyDrops_CR --soloFeatures GeneFull --soloCBwhitelist ./3M-february-2018.txt --genomeDir /home/kath/storage2/dmel104_10a --outFilterType BySJout --alignIntronMax 100000 --quantMode GeneCounts --outSAMtype BAM SortedByCoordinate --limitBAMsortRAM 4715576210 --soloOutFileNames ZT00_10a_solo.out/ ZT00_features.tsv ZT00_barcodes.tsv ZT00_matrix.mtx --outFileNamePrefix ZT00_10a_ --readFilesCommand "zcat" --readFilesPrefix /home/kath/b1042/fatbody_sc_ZT/01.RawData/ZT0/ --readFilesIn ZT0_CKDL220015693-1A_H37TGDSX5_S1_L004_R2_001.fastq.gz ZT0_CKDL220015693-1A_H37TGDSX5_S1_L004_R1_001_trimmed.fastq.gz

The run makes it to

Jul 24 09:34:39 ..... started Solo counting
Jul 24 09:34:39 ... Starting Solo post-map for GeneFull
Jul 24 09:34:39 ... Finished allocating arrays for Solo 1.91906 GiB

and then segfaults. STAR version 2.7.9a is able to complete without error with the same run parameters.

Here's the output of gdb ~/bin/STAR_2.7.10a core.23748

Program terminated with signal 11, Segmentation fault.
#0 0x000000000046a1dc in SoloReadFeature::inputRecords(unsigned int**, unsigned int, std::vector<unsigned int, std::allocator >&, std::vector<readInfoStruct, std::allocator >&, SoloReadFlagClass&, std::vector<unsigned int, std::allocator >&, std::vector<unsigned int, std::allocator >&) ()
Missing separate debuginfos, use: debuginfo-install glibc-2.17-326.el7_9.x86_64 libgcc-4.8.5-44.el7.x86_64 libgomp-4.8.5-44.el7.x86_64 libstdc++-4.8.5-44.el7.x86_64 zlib-1.2.7-20.el7_9.x86_64

Going further with gdb than this is beyond my normal experience but I'm hoping it at least suggests possibilities.

Thanks,

• Bill

[alexdobin]

Hi Bill,

Please try to run it on a small subset of reads, say 100k, with --readMapNumber 100000.
If it still sega-faults, please send me this subset of reads.

Thanks!
Alex

[wlkath]

Alex,

Yes, the same error occurs on the subset of reads.

Here's the command I used:

STAR --runThreadN 20 --soloType CB_UMI_Simple --soloUMIlen 12 --soloCellFilter EmptyDrops_CR --soloFeatures GeneFull --soloCBwhitelist ./3M-february-2018.txt --genomeDir /home/kath/storage2/dmel104_10a --outFilterType BySJout --alignIntronMax 100000 --quantMode GeneCounts --outSAMtype BAM SortedByCoordinate --limitBAMsortRAM 4715576210 --soloOutFileNames ZT00_10a_solo.out/ ZT00_features.tsv ZT00_barcodes.tsv ZT00_matrix.mtx --outFileNamePrefix ZT00_10a_ --readFilesCommand "zcat" --readFilesPrefix /home/kath/b1042/fatbody_sc_ZT/01.RawData/ZT0/ --readMapNumber 100000 --readFilesIn Sampled_100k_R2.fq.gz Sampled_100k_R1.fq.gz

Thanks,

• Bill

P.S. I forgot to say that these are Drosophila melanogaster. I used the Ensemble reference and annotation, v104. I did my best to filter the annotation as recommended by 10x.

Drosophila_melanogaster.BDGP6.32.dna.toplevel.fa
Drosophila_melanogaster.BDGP6.32.104.gtf

Sampled_100k_R1.fq.gz
Sampled_100k_R2.fq.gz

[alexdobin]

Hi @wlkath

I could not reproduce the seg-fault with your FASTQ and parameters, but I used the complete GTF from ENSEMBL.
Could you please send me the filtered GTF as well as Log.out file for the failed run.
Also, I fixed another potential seg-fault, if you could test this patch it would be great:
https://github.com/alexdobin/STAR/releases/tag/2.7.10a_alpha_220818

Thanks!
Alex

[wlkath]

Thanks, Alex. I'll check out the latest alpha version.

In the meantime, here are the GTF and log files that were generated.

gtf_and_logs.zip

[wlkath]

Alex,

2.7.10a_alpha_220818 completes without the seg-fault, so your patch has resolved the issue.

Thanks very much for the fix!

Best regards,

• Bill

[alexdobin]

Great, many thanks for testing it, Bill!