alexdobin / STAR

RNA-seq aligner
MIT License
1.87k stars 506 forks source link

Segmentation Fault When STARSolo Starts Counting #1166

Open skvanburen opened 3 years ago

skvanburen commented 3 years ago

Hi,

First off, thanks for working on such a great tool that has proven useful to so many over the years!

However, I am having an issue when using STARSolo where I get a “Segmentation fault (core dumped)” error right after the Solo counting starts. I am wondering if it is a bug with the exact combination of counting parameters I have specified, specifically keep all cell barcodes and only collapse exact UMIs. The issue may be somewhat similar to issue 727 (https://github.com/alexdobin/STAR/issues/727) and issue 936 (https://github.com/alexdobin/STAR/issues/936).

I give the command I’m using below and have attached the Log.out and Log.out.progress files here: Log.out.txt Log.progress.out.txt

Happy to provide more details but it seems like this is often enough information for you to diagnose the issue (amazingly to me!).

Note that this is using the newest version, 2.7.8a, though I also tried 2.7.6a and still have the same issue. Happy to provide any assistance I can. Thanks!

/n/home04/skvanburen/bin/STAR-2.7.8a/bin/Linux_x86_64/STAR --runThreadN 6 --soloType CB_UMI_Simple --soloCBwhitelist None --soloCBstart 1 --soloCBlen 19 --soloUMIstart 20 --soloUMIlen 8 --soloBarcodeReadLength 1 --genomeDir /n/holylfs03/LABS/dulac_lab/Users/skvanburen/sciPlex/STARIndex --readFilesCommand zcat --readFilesIn /n/holylfs03/LABS/dulac_lab/Users/skvanburen/sciPlex/210106_NB502063_0429_AHJLYGAFX2_noadapters/SVBPipelineOutput/NewFASTQs/Klenow.R2.Length27.fastq.gz /n/holylfs03/LABS/dulac_lab/Users/skvanburen/sciPlex/210106_NB502063_0429_AHJLYGAFX2_noadapters/SVBPipelineOutput/NewFASTQs/Klenow.R1.Length27.fastq.gz --outFileNamePrefix /n/holylfs03/LABS/dulac_lab/Users/skvanburen/sciPlex/210106_NB502063_0429_AHJLYGAFX2_noadapters/SVBPipelineOutput/STAROutput/R1ReadLength27/Klenow/ --soloFeatures GeneFull --soloUMIdedup Exact --soloCellFilter None --genomeLoad LoadAndRemove --outSAMtype None

alexdobin commented 3 years ago

Hi Scott,

this looks like a bug. Could you please run it with --readMapNumber 1000, and - if it still seg-faults, send me the first 1000 reads from Read1 and Read2?

Cheers Alex

skvanburen commented 3 years ago

Hi Alex,

Thank you for the quick reply, I have just tried running the code with --readMapNumber 1000 and I still receive the same error at the same spot. I have attached the first 1000 reads here (adding .txt to the end just so Github will allow me to upload the files) R1First1000Reads.fastq.gz.txt R2First1000Reads.fastq.gz.txt

Let me know if I can provide any other assistance. Thanks! Scott

alexdobin commented 3 years ago

Hi Scott,

thanks a lot for reporting this bug. Please try the patch I just pushed to GitHub-master, it should have fixed the issue. I will make a 2.7.8b release in a couple of days.

Cheers Alex

skvanburen commented 3 years ago

Hi Alex,

I have tried the patch and it did fix the bug. Many thanks for the quick fix!

Scorr