Fix/prepare fastqs self compatibility 20240521 (#9)

* prepare_fastqs should now be able to import its output

* fix: allow to keep alignment files via --keep_alignment_file, v.0.12.7

bin/prepare_fastqs.py

@@ -166,9 +166,14 @@ def process_sample(
 
 		print("PRE", prefixes, file=sys.stderr)
 
+		def rx_filter(s, x=1):
+			return re.search(r"[._R]" + str(x) + r"$", s) and not re.search(r"(orphan|single)s?", s)
+
 		# partition fastqs into R1, R2, and 'other' sets
-		r1 = [(p, f) for p, f in zip(prefixes, fastqs) if re.search(r"[._R]1$", p)]
-		r2 = [(p, f) for p, f in zip(prefixes, fastqs) if re.search(r"[._R]2$", p)]
+		# r1 = [(p, f) for p, f in zip(prefixes, fastqs) if re.search(r"[._R]1$", p)]
+		r1 = [(p, f) for p, f in zip(prefixes, fastqs) if rx_filter(p, x=1)]
+		# r2 = [(p, f) for p, f in zip(prefixes, fastqs) if re.search(r"[._R]2$", p)]
+		r2 = [(p, f) for p, f in zip(prefixes, fastqs) if rx_filter(p, x=2)]
 		others = sorted(list(set(fastqs).difference({f for _, f in r1}).difference({f for _, f in r2})))
 
 		# check if R1/R2 sets have equal sizes or are empty
@@ -180,7 +185,7 @@ def process_sample(
 		if len(r1) == len(r2) and r1:
 			check_pairwise(r1, r2)
 
-		# sort R1/R2 for concatenation, get rid off prefixes
+		# sort R1/R2 for concatenation, get rid of prefixes
 		r1 = sorted(f for _, f in r1)
 		r2 = sorted(f for _, f in r2)
 

nevermore/modules/profilers/gffquant.nf

@@ -12,7 +12,7 @@ process stream_gffquant {
 		tuple val(sample), path("profiles/${sample}/*.{txt.gz,pd.txt}"), emit: results //, optional: (!params.gq_panda) ? true : false
 		tuple val(sample), path("profiles/${sample}/*.{txt.gz,pd.txt}"), emit: profiles //, optional: (params.gq_panda) ? true : false
 		tuple val(sample), path("logs/${sample}.log")
-		tuple val(sample), path("alignments/${sample}*.sam"), emit: alignments, optional: true
+		tuple val(sample), path("alignments/${sample}/${sample}*.sam"), emit: alignments, optional: true
 
 	script:
 			def gq_output = "-o profiles/${sample}/${sample}"
@@ -22,7 +22,7 @@ process stream_gffquant {
 			gq_params += (params.gq_min_seqlen) ? (" --min_seqlen " + params.gq_min_seqlen) : ""
 			gq_params += (params.gq_min_identity) ? (" --min_identity " + params.gq_min_identity) : ""
 			gq_params += (params.gq_restrict_metrics) ? " --restrict_metrics ${params.gq_restrict_metrics}" : ""
-			gq_params += (params.gq_keep_alignments) ? " --keep_alignment_file ${sample}.sam" : ""
+			gq_params += (params.gq_keep_alignments) ? " --keep_alignment_file alignments/${sample}.sam" : ""
 			gq_params += (params.gq_unmarked_orphans) ? " --unmarked_orphans" : ""
 
 			gq_params += " -t ${task.cpus}"
@@ -57,10 +57,12 @@ process stream_gffquant {
 			}
 
 			def gq_cmd = "gffquant ${gq_output} ${gq_params} --db GQ_DATABASE --reference \$(readlink ${reference}) --aligner ${params.gq_aligner} ${input_files}"
+			def mkdir_alignments = (params.keep_alignment_file != null && params.keep_alignment_file != false) ? "mkdir -p alignments/${sample}/" : ""
 
 			"""
 			set -e -o pipefail
 			mkdir -p logs/ tmp/ profiles/
+			${mkdir_alignments}
 			echo 'Copying database...'
 			cp -v ${gq_db} GQ_DATABASE
 			${gq_cmd} &> logs/${sample}.log

nevermore/version.json

@@ -1,3 +1,3 @@
 {
-	"version": "0.12.6"
+	"version": "0.12.7"
 }

nextflow.config

@@ -4,5 +4,5 @@ manifest {
 	description = "Metaomics pipeline toolbox"
 	name = "nevermore"
 	nextflowVersion = ">=22.10.6"
-	version = "0.12.6"
+	version = "0.12.7"
 }