add workflow for strainphlan on paired-end

workflows/strainphlan/README.md

+# StrainPhlan
+
+This describes a workflow to run [strainphlan](https://github.com/biobakery/MetaPhlAn/wiki/StrainPhlAn-3.0)
+
+In brief it contains 4 steps:
+
+* Run Metaphlan on all samples (paired-ends)
+* Extract markers from metaphlan outputs
+* Run StrainPhlan on all selected clades
\ No newline at end of file

workflows/strainphlan/Snakefile

+configfile: "config.yaml"
+
+# ==== Snakefile paths ====
+__metaphlan3_rules = config.get("snakefiles", {}).get("metaphlan3")
+__metaphlan2_merge_rules = config.get("snakefiles", {}).get("metaphlan2_merge")
+__sample2markers_rules = config.get("snakefiles", {}).get("sample2markers")
+__strainphlan_rules = config.get("snakefiles", {}).get("strainphlan")
+
+__input_dir = config['input_dir']
+__main_output_dir = config.get('output_dir', 'output')
+
+# ---- Metaphlan3
+__metaphlan3_output_dir = __main_output_dir + "/metaphlan3"
+__metaphlan3_input_type = config['metaphlan3'].get('input_type', 'fastq')
+__metaphlan3_input_r1 = "{dir}/{sample}{ext}".format(dir=__input_dir,
+                                                      sample="{sample}",
+                                                      ext="_1.fq.gz")
+__metaphlan3_input_r2 = "{dir}/{sample}{ext}".format(dir=__input_dir,
+                                                      sample="{sample}",
+                                                      ext="_2.fq.gz")
+__metaphlan3_output_profile = "{dir}/{sample}.profile.tsv".format(dir=__metaphlan3_output_dir,
+                                                                  sample="{sample}")
+__metaphlan3_output_bowtie2out = "{output_dir}/{sample}.bowtie2.bz2".format(output_dir=__metaphlan3_output_dir, sample="{sample}")
+__metaphlan3_output_sams = "{output_dir}/{sample}.sam.bz2".format(output_dir=__metaphlan3_output_dir, sample="{sample}")
+
+include: __metaphlan3_rules
+
+# ---- Metaphlan2 merge
+__metaphlan2_merge_output_dir = __main_output_dir + "/metaphlan_merge"
+__metaphlan2_merge_output_file_name = config['metaphlan2_merge'].get('output_file_name',"merged_taxonomic_profiles.txt")
+__metaphlan2_merge_input = expand("{dir}/{sample}.profile.tsv".format(dir=__metaphlan3_output_dir,
+                                                                      sample="{sample}"),
+                                  sample=config['samples'])
+__metaphlan2_merge_output = "{dir}/{file_name}".format(dir=__metaphlan2_merge_output_dir,
+                                                       file_name=__metaphlan2_merge_output_file_name)
+include: __metaphlan2_merge_rules
+
+rule metaphlan_merge_all:
+    input:
+        "{dir}/{file_name}".format(dir=__metaphlan2_merge_output_dir,
+                                   file_name=__metaphlan2_merge_output_file_name)
+
+# ---- Sample2markers ----
+__sample2markers_main_output_dir = __main_output_dir + "/sample2markers"
+__sample2markers_output_dir = "{dir}/{sample}".format(dir=__sample2markers_main_output_dir,
+                                                      sample="{sample}")
+__sample2markers_input = __metaphlan3_output_sams
+__sample2markers_output = "{dir}/{sample}.pkl".format(dir=__sample2markers_output_dir,
+                                                      sample="{sample}")
+
+include: __sample2markers_rules
+
+
+# ---- StrainPhlan ----
+
+__strainphlan_main_output_dir = __main_output_dir + "/strainphlan"
+__strainphlan_clade = "{clade}"
+__strainphlan_output_dir = "{dir}/{clade}".format(dir=__strainphlan_main_output_dir,
+                                                  clade="{clade}")
+
+__strainphlan_input = expand("{dir}/{sample}.pkl".format(dir=__sample2markers_output_dir,
+                                                    sample="{sample}"),
+                             sample=config['samples'])
+__strainphlan_output = "{dir}/{prefix}.{clade}.{ext}".format(dir=__strainphlan_output_dir,
+                                                             prefix="RAxML_bestTree",
+                                                             clade="{clade}",
+                                                             ext="StrainPhlAn3.tre")
+
+include: __strainphlan_rules
+
+
+rule all:
+    input:
+        expand("{dir}/{prefix}.{clade}.{ext}".format(dir=__strainphlan_output_dir,
+                                                             prefix="RAxML_bestTree",
+                                                             clade="{clade}",
+                                                             ext="StrainPhlAn3.tre"),
+               clade=config['strainphlan']['clades'])

workflows/strainphlan/config.yaml

+snakefiles:
+  metaphlan3: /pasteur/zeus/projets/p02/metasig/gitlab/snakemake/tools/metaphlan3/metaphlan/paired/Snakefile
+  metaphlan2_merge: /pasteur/zeus/projets/p02/metasig/gitlab/snakemake/tools/metaphlan2/metaphlan2_merge/Snakefile
+  sample2markers: /pasteur/zeus/projets/p02/metasig/gitlab/snakemake/tools/strainphlan/sample2markers/Snakefile
+  strainphlan: /pasteur/zeus/projets/p02/metasig/gitlab/snakemake/tools/strainphlan/strainphlan/Snakefile
+
+samples:
+  - sample_1
+  - sample_2
+  - sample_2
+
+input_dir: /a/path/to/input/data
+output_dir: output
+
+metaphlan3:
+  threads: 4
+  input_type: fastq
+  options: --bowtie2db /pasteur/zeus/projets/p02/metasig/DBs/metaphlan/metaphlan3bowtie2db/ --index mpa_v30_CHOCOPhlAn_201901
+
+metaphlan2_merge:
+  threads: 1
+
+sample2markers:
+  threads: 4
+
+strainphlan:
+  threads: 8
+  options: ""
+  clades:
+    - s__Escherichia_coli
+    - s__Klebsiella_pneumoniae