STAR error EXITING because of FATAL ERROR: failed reading from temporary file
1
1
Entering edit mode
15 months ago

Hello all,

I'm attempting to use STAR to map some RNA-seq data, but keep getting some sort of error. Here is the command I used:

for i in `ls *_clean.fastq | sed 's/_clean.fastq//'`; do STAR --runThreadN 20 --quantMode GeneCounts --outFileNamePrefix aligned_haploid/$i --outSAMtype BAM SortedByCoordinate --outSAMunmapped Within --genomeDir /media/smasher/4TB_data/pairwise_brb_8-2023/haploid_genome --readFilesIn $i\_clean.fastq; done

Here's the error I get:

EXITING because of FATAL ERROR: failed reading from temporary file: aligned_haploid/A01_STARtmp//BAMsort/2/44

It doesn't happen to every fastq file, I mapped 95 files and 5 of them generated this error. I thought maybe it was something to do with the fastq's themselves because I mapped them individually rather than in a loop and still got this same error. However, when I map them to a different, but very similar, genome, I get the error but for different files.

Anyone know what's going on here and have an idea of how to fix it? I'm running STAR 2.7.10a via Mamba, could that be the problem?

Thanks!

STAR RNA-seq • 1.6k views
ADD COMMENT
2
Entering edit mode
15 months ago
GenoMax 148k

Using STAR to sort the BAM files during alignment requires a lot of additional resources. Looks like you may be running out of temp space. You could sort the files after the alignments using samtools and avoid this error.

Looking at the paths in your command lines you appear to be doing this analysis using some kind of mounted external storage drive, is that correct? That may also be causing issues because of not having enough bandwidth to support 20 cores.

ADD COMMENT
0
Entering edit mode

Looks like this fixed it, thank you! I started it running without sorting after reading your response and every single one has mapped successfully so far, including a couple that were giving the error before. This will be great to know going forward, thanks again!!!

Looking at the paths in your command lines you appear to be doing this analysis using some kind of mounted external storage drive, is that correct?

It's an internal drive, it's just mounted to /media.

ADD REPLY
0
Entering edit mode

thanks, for me it was the number of cores indeed, had always used 8 before, this time kept giving error with 24 cores

ADD REPLY

Login before adding your answer.

Traffic: 2082 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6