metaspades megahit

assembly.smk

@@ -13,7 +13,7 @@ rule khmer:
 	threads:
 		10
 	params:
-		mem_tot = 90 * 1e9,
+		mem_tot = int(90 * 1e9),
 		K = 32,
 		C = 100
 	shell:
@@ -26,3 +26,70 @@ rule khmer:
 		"--output {output.R1R2} "
 		"--gzip "
 		"{input.R1R2}"
+
+def assembly_input(wildcards):
+	if wildcards.reads == "coassembly":
+		if config["NORMALIZATION"]:
+			return [expand("work/khmer/{reads}_R1R2.fastq.gz", reads=config["SAMPLES"])]
+		elif config["SORTMERNA"]:
+			return [expand("work/sortmerna/{reads}_mRNA.fastq.gz", reads=config["SAMPLES"])]
+		else:
+			return [expand("DATA/trim/{reads}_R1R2.fastq.gz", reads=config["SAMPLES"])]
+	else:
+		if config["NORMALIZATION"]:
+			return {"R1R2": "work/khmer/{wildcards.reads}_R1R2.fastq.gz".format(wildcards=wildcards)}
+		elif config["SORTMERNA"]:
+			return {"R1R2": "work/sortmerna/{wildcards.reads}_mRNA.fastq.gz".format(wildcards=wildcards)}
+		else:
+			return {"R1R2": "DATA/trim/{wildcards.reads}_R1R2.fastq.gz".format(wildcards=wildcards)}
+
+rule megahit:
+	input:
+		unpack(assembly_input)
+	output:
+		contigs = "work/megahit/{reads}/{reads}.contigs.fa"
+	threads:
+		20
+	params:
+		min_len = config["CONTIGS_LEN"],
+		mem_tot = int(250 * 1e9),
+		input = lambda wildcards, input: ",".join(input),
+		output = lambda wildcards, output: os.path.dirname(str(output.contigs))
+	shell:
+		"megahit "
+		"--12 {params.input} "
+		"--continue "
+		"--preset meta-large "
+		"--num-cpu-threads {threads} "
+		"--memory {params.mem_tot} "
+		"--tmp-dir /projet/tmp "
+		"--out-dir {params.output} "
+		"--out-prefix {wildcards.reads} "
+		"--min-contig-len {params.min_len} "
+		"--verbose"
+
+rule metaspades:
+	input:
+		unpack(assembly_input)
+	output:
+		contigs = "work/metaSPADES/{reads}/{reads}.contigs.fasta"
+	threads:
+		20
+	params:
+		min_len = config["CONTIGS_LEN"],
+		mem_tot = 250,
+		input = lambda wildcards, input: " --12 ".join(input),
+		output = lambda wildcards, output: os.path.dirname(str(output))
+	shell:
+		"spades.py "
+		"--threads {threads} "
+		"--memory {params.mem_tot} "
+		"--tmp-dir /projet/tmp "
+		"--meta "
+		"--12 {params.input} "
+		"-o {params.output} "
+		" ; "
+		"filterSeq.py "
+		"-i work/metaSPADES/{wildcards.reads}/contigs.fasta "
+		"-o {output.contigs} "
+		"--min-length {params.min_len} "

cluster.json

@@ -19,12 +19,12 @@
 	"megahit" :
 	{
 		"queue" : "highmem.q,maiage.q",
-		"cluster" : "-l h_vmem={MEM}G"
+		"cluster" : "-l h_vmem=12G"
 	},
 	"metaspades" :
 	{
 		"queue" : "highmem.q,maiage.q",
-		"cluster" : "-l h_vmem={MEM}G"
+		"cluster" : "-l h_vmem=12G"
 	},
 	"diamond" :
 	{