https://www.biostars.org/p/9523782/ blast biostar nextflow fasta sequence align

biostar9523782.nf

nextflow.enable.dsl=2

/* full path to query directory */
params.qdir="/DIR1"

/* full path to target/database directory */
params.tdir="/DIR2"


workflow {
		dir_ch = Channel.of(["query",params.qdir],["target",params.tdir])
		fasta_ch = findFasta(dir_ch).splitCsv(header: false,sep:'\t',strip:true)
		query_ch = fasta_ch.filter{T->T[0].equals("query")}.map{T->T[1]}
		target_ch = fasta_ch.filter{T->T[0].equals("target")}.map{T->T[1]}
		blast_ch = blastIt(query_ch.combine(target_ch))
		zipIt(blast_ch.collect())

}

process findFasta {
tag "${type} in ${dir}"
input:
	tuple val(type),val(dir)
output:
	path("paths.txt")
script:
"""
test -d "${dir}"
find "${dir}" -type f  -name "*.fa" -o -name "*.fasta" |\
	sed 's/^/${type}\t/' > paths.txt
test -s paths.txt
"""
}

process blastIt {
	tag "${query} / ${target}"
	input:
		tuple val(query),val(target)
	output:
		path("out.blast")
	script:
	"""
	blastp -query "${query}" -db ${target} -out out.blast
	"""
	}

process zipIt {
	input:
		val(L)
	output:
		path("blast.zip")
	script:
	"""
	zip -9 blast.zip ${L.join(" ")}
	"""
	}