Brian Haas also behind this Splice Alignment tool.
PASA has quite alot of parts and it could almost be seen as a mini-pipeline. Therefore the user needs to read the documentation well to use the software properly. Please see the links section for PASA's main documentation page.
The canonical command for running it on the marvin cluster is:
Launch_PASA_pipeline.pl -c alignAssembly.config -C -R -g genome_sample.fasta -t all_transcripts.fasta.clean -T -u all_transcripts.fasta -f FL_accs.txt --ALIGNERS blat,gmap --CPU 2
- -c is a configuration file which you, the user, needs to set up. The database name you choose for the analysis needs to be specified in this.
- -C is for creating a new database, if there is one there already, the command will fail.
- --CPU is for the number of threads.
- The PASA mysql user requires full privileges so it can create databases, which it apparently does for every analysis.
- these privileges must also exist for '%', so that all the nodes can also do this.
- the issue of a key being too long for mysql was corrected by replacing 1000 with 500 in this line.
gene_id varchar(1000) default NULL,
in the $PASAHOME/schema/cdna_alignment_mysqlschema file on line 174, which was also an issue back in 2013 (see links section).