The idea of this pipeline is to parallelise PASA to make it faster to run. Having PASA would be nice because it can be used to predict gene from evidence that can be used in different ways:
PASA makes probably a a better pure evidence-based annotation than MAKER.
the PASA annotation can be used to improve/polish an annotation made with another tool.
the PASA annotation can be used to train Abinitio tools
PASA is already available on Conda. The difficulty is to parallelise it. Marc has already implemented something in esga see here.
Can we use it like it is?
It sounds we could make it more generalised and better commented. For GFF related task (e.g GffToFasta we can use AGAT's script, see other NBIS pipelines to use it) and for fastaSplitSize we can use GAAS script gaas_fasta_splitter.pl.
See #17 for the general picture.
The idea of this pipeline is to parallelise PASA to make it faster to run. Having PASA would be nice because it can be used to predict gene from evidence that can be used in different ways:
PASA is already available on Conda. The difficulty is to parallelise it. Marc has already implemented something in esga see here. Can we use it like it is? It sounds we could make it more generalised and better commented. For GFF related task (e.g GffToFasta we can use AGAT's script, see other NBIS pipelines to use it) and for fastaSplitSize we can use GAAS script gaas_fasta_splitter.pl.