Kallisto quant output file abundance.tsv and run_info.json missing

Question

[Abiterkuile]

After running a gird engine job for kallisto, in my output only abundance.h5 file has been produced. Does anyone have any idea why the abundances.tsv and run_info.json output files are missing (copy of script below)? Could this indicate that the job did not run properly and could effect my downstream analysis? Thanks!

kallisto quant -i /users/k1456980/brc_scratch/SNCID/PFC/mapping_PFC/references/kallisto_index_RefSeq.idx -o /users/k1456980/brc_scratch/SNCID/PFC/mapping_PFC/refseq/C"$SGE_TASK_ID" -b 100 --pseudobam C"$SGE_TASK_ID"_FP.fastq.gz C"$SGE_TASK_ID"_RP.fastq.gz

[Kevin Blighe]

I presume that you are running this as a loop, but in which shell? Does it work if you just type the full command for a single FASTQ pair?

Can you also check the messages that are output to terminal, as they may prove informative?

In some situations, you have to wrap variables like this: "${SGE_TASK_ID}". also, it can be a good idea in certain situations to execute your command like this:

echo ´kallisto quant -i /users/k1456980/brc_scratch/SNCID/PFC/mapping_PFC/references/kallisto_index_RefSeq.idx -o /users/k1456980/brc_scratch/SNCID/PFC/mapping_PFC/refseq/C"$SGE_TASK_ID" -b 100 --pseudobam C"$SGE_TASK_ID"_FP.fastq.gz C"$SGE_TASK_ID"_RP.fastq.gz´

Those are back-ticks surrounding the echo command. I don't know the specifics of your shell / OS environment, so, am only speculating that these may be the issues. Your command looks fine, otherwise.

Final idea: execute kallisto using it's full path, and/or ensure that its directory is in the PATH variable.

[Abiterkuile]

Thank you for your reply and help. In the end it was because I had exceeded by disk quota limit without realizing. After freeing up some space the outputs were fine.