Entering edit mode
7.1 years ago
XBria
▴
90
Hi,
I am trying to run STAR to map. The code is :
STAR --runThreadN 8 --genomeDir /home/XBria/bin/chrX.fa --sjdbGTFfile /home/
XBria /my_rnaseq_exp/chrX_data/genes/chrX.gtf --readFilesIn /home/
XBria /my_rnaseq_exp/chrX_data/samples/ERR188044_chrX_1.fastq.gz /home/
XBria /my_rnaseq_exp/chrX_data/samples/ERR188044_chrX_2.fastq.gz --outFileNamePrefix STAR
result:
Nov 28 19:35:54 ..... started STAR run
Nov 28 19:35:54 ..... loading genome
EXITING because of FATAL ERROR: could not open genome file /home/XBria/bin/chrX.fa/genomeParameters.txt
SOLUTION: check that the path to genome files, specified in --genomeDir is correct and the files are present, and have user read permsissions
Nov 28 19:35:54 ...... FATAL ERROR, exiting
I drag the chrX.fa and forward and reverse reads files into the terminal. Any advice is appreciated.
Thanks.
Looks like you have spaces in several file paths? (or is that just a side effect of copy-paste?) Remove those.
I just edited and tested. Unfortunately the same error occurs !
Have you created STAR genome index (in this case for chrX) already?
Oops. no! I 'm trying to run with the Hisat indices I've already mapped !
Many thanks for the hint !
Dear Genomax,
I run making indices and got the following :
Nov 28 20:12:02 ..... started STAR run Nov 28 20:12:02 ... starting to generate Genome files Nov 28 20:12:04 ... starting to sort Suffix Array. This may take a long time... Nov 28 20:12:05 ... sorting Suffix Array chunks and saving them to disk... Killed
Is it done right !!
Well, it's "done" in that it's not running, but the "Killed" means that your system killed it. I'm guessing that you ran out of memory.
That doesn't look right. Do you have sufficient memory available?
I am not indexing a large file. it is only chrX ! RAM of the system is 8G. fa file is 158 and GTF is 7 MB ! Are they still too much ?
I think there are some tricks to make the indexing work if you have less RAM if I recall right. Search for STAR genome index threads here. You may need to use one of those workarounds.
After sorting out the memory issue, you should also add the parameter
--readFilesCommand zcat
, since your fastq files are zipped.You can check the memory usage of the process using htop or top.