Hyphens seem to cause issues when running alignment (STAR) step

[SergeWielhouwer]

While running the COMPSRA example workflow with the following command java -jar COMPSRA.jar -ref hg38 -qc -ra TGGAATTCTCGGGTGCCAAGG -rb 4 -rh 20 -rt 20 -rr 20 -rlh 8,17 -aln -mt star -ann -ac 1,2,3,4,5,6 -inf ./example/sample.list -out ./example_out/ from a working directory such as /mnt/example/COMPSRA-1.0.3 the --outFileNamePrefix from STAR unexpectedly changes to /mnt/example/COMPSRA-1STAR instead of /mnt/example/COMPSRA-1.0.3/example_out/sample01/sample01_17to50_FitReadSTAR which causes the the files to be written to the wrong directory.

[cougarlj]

Hi, SergeWiehouwer,

There is no need to set the parameter "--outFileNamePrefix" in STAR. COMPSRA will set it through "-out" automatically.

Best Wishes, Jiang

[SergeWielhouwer]

Hi Jiang,

Thank you for your quick response.

I understand that the "--outFileNamePrefix" will be automatically set through COMPSRA. The issue that I am experiencing is that if a path contains a hyphen (-) that the parameter value for --outFileNamePrefix created by COMPSRA is incorrect.

Btw, may I ask how you determined the most optimal STAR parameters for use with Small RNA data? Did you make use of suggestions made by the developer of STAR (Alexander Dobin) https://groups.google.com/g/rna-star/c/RBWvAGFooMU ?

Thanks,

Serge

[cougarlj]

Dear Serge,

OK, I know what your mean. This is a bug that I will fix it in the next version. I think we didn't define the best parameters for small RNA-Seq. We just tried this using our small RNA-Seq data and compared with our tools. Finally, we though this set might be a good one.

Best Wishes, Jiang

[SergeWielhouwer]

Dear Jiang,

Thanks for the update. My compliments for designing this tool, easy to use workflows for small RNA (not just miRNAs) data analysis are currently quite rare and therefore more than welcome!

Good to know on how you determined the STAR parameters and that small changes could possibly yield better results 👍.

Regards,

Serge