Hello everyone, I'm having some trouble running a mpiblast script for blastx analysis... It works perfectly with other datasets, but for some reason I can't understand it doesn't like [this specific file as input]. The only difference I've noticed with working files is that the nucletode sequence is on one single line... I've removed dots and anything strange in the > line as well but this didn't make any difference.
- I've formatted my database (Uniti.fa) as required by mpiblast through the script ./mpiformatdb, disassembling it in 16 different files
./mpiformatdb -i /home/genomica/DATA/Databases/prova_unione/Uniti.fa -l /home/genomica/DATA/Databases/prova_unione/LogTrasformazione.txt -p F -o T -a F -V -t NuovoDatabaseFormattato_2 -N 16
The log file shows nothing abnormal.
- I've run mpiblast from the folder where my reference transcriptome is contained, as follows:
mpirun -np 50 --hostfile /home/genomica/DATA/Software/script/hostfile_blast mpiblast -p blastx -d Uniti.fa -i Tuttie4_conN.fasta -m 9 -e 1e-5 -o Tuttie4_conN_codici.blast -v 5 -b 5
This is the detailed error log I immediately get... Any idea what could be wrong and why this specific file doesn't work? Thanks in advance!
I am not answering your question but I am curious as to what is the advantage of using MPIblast these days. A tool like DIAMOND can be as performant or perhaps more without the complexity associated with MPIblast.
I'll take this into consideration, I'm just a newbie to Bioinformatics and that's what my tutor suggested me to use since it's already installed on our server... Thanks for the suggestion!