Add MultiQC to snv-indels module

Snakefile

@@ -210,6 +210,7 @@ rule generate_html_report:
 rule multiqc:
     input:
         qc_stats=qc_seq.module_output.multiqc_files,
+        snv_indel_stats=align.module_output.multiqc_files,
         config=srcdir("cfg/multiqc.yml"),
     params:
         filelist="multiqc_filelist.txt",
@@ -224,7 +225,7 @@ rule multiqc:
         """
         rm -f {params.filelist}
 
-        for fname in {input.qc_stats}; do
+        for fname in {input.qc_stats} {input.snv_indel_stats}; do
             echo $fname >> {params.filelist}
         done
 

cfg/multiqc.yml

@@ -2,3 +2,6 @@ show_analysis_paths: False
 sample_names_replace_regex: True
 sample_names_replace:
   "(.+) \\| qc-seq \\| (.+).cutadapt.json": "\\1"
+  "(.+) \\| snv-indels \\| (.+).ReadsPerGene.out.tab": "\\1"
+  "(.+) \\| snv-indels \\| Log.final.out": "\\1"
+  "(.+) \\| snv-indels \\| .*": "\\1"

includes/qc-seq/Snakefile

@@ -100,7 +100,7 @@ rule multiqc:
     output:
         html="multiqc_qc_seq.html",
     log:
-        "log/multiqc.txt",
+        "log/qc_seq.multiqc.txt",
     container:
         containers["multiqc"]
     shell:

includes/snv-indels/Snakefile

@@ -1,11 +1,6 @@
 include: "common.smk"
 
 
-# Put each sample name in a SimpleNamespace to mimic Snakemake wildcard usage
-# (e.g {wildcards.sample}). This is only used in the 'all' rule.
-samples = [SimpleNamespace(sample=sample) for sample in pep.sample_table["sample_name"]]
-
-
 localrules:
     filter_vep_target,
     filter_vep_high,
@@ -20,6 +15,7 @@ rule all:
         vep_target=[module_output.vep_target(sample) for sample in samples],
         json=[module_output.json(sample) for sample in samples],
         hotspot=[module_output.hotspot(sample) for sample in samples],
+        multiqc="multiqc_snv_indels.html",
 
 
 rule tmpdir:
@@ -464,3 +460,36 @@ rule json_output:
             --exon_cov_stats_path {input.exon_cov_stats} \
             --vep_stats_path {input.vep_stats} > {output} 2> {log}
         """
+
+
+rule multiqc:
+    input:
+        stats=module_output.multiqc_files,
+        config=srcdir("../../cfg/multiqc.yml"),
+    params:
+        filelist="multiqc_filelist.txt",
+        depth=2,
+    output:
+        html="multiqc_snv_indels.html",
+    log:
+        "log/snv_indels.multiqc.txt",
+    container:
+        containers["multiqc"]
+    shell:
+        """
+        rm -f {params.filelist}
+
+        for fname in {input.stats}; do
+            echo $fname >> {params.filelist}
+        done
+
+        multiqc \
+        --force \
+        --dirs \
+        --dirs-depth {params.depth} \
+        --fullnames \
+        --fn_as_s_name \
+        --file-list {params.filelist} \
+        --config {input.config} \
+        --filename {output.html} 2> {log}
+        """

includes/snv-indels/common.smk

@@ -12,9 +12,15 @@ containers = {
     "vep": "docker://quay.io/biocontainers/ensembl-vep:108.2--pl5321h4a94de4_0",
     "star": "docker://quay.io/biocontainers/star:2.7.10b--h9ee0642_0",
     "crimson": "docker://quay.io/biocontainers/crimson:1.1.0--pyh5e36f6f_0",
+    "multiqc": "docker://quay.io/biocontainers/multiqc:1.21--pyhdfd78af_0",
 }
 
 
+# Put each sample name in a SimpleNamespace to mimic Snakemake wildcard usage
+# (e.g {wildcards.sample}). This is only used in the 'all' rule.
+samples = [SimpleNamespace(sample=sample) for sample in pep.sample_table["sample_name"]]
+
+
 def get_forward(wildcards):
     return pep.sample_table.loc[wildcards.sample, "R1"]
 
@@ -48,11 +54,35 @@ def get_hotspot(wildcards):
     return f"{wildcards.sample}/snv-indels/{wildcards.sample}.hotspot.vcf"
 
 
+def multiqc_files():
+    star_count = [
+        f"{wildcards.sample}/snv-indels/{wildcards.sample}.ReadsPerGene.out.tab"
+        for wildcards in samples
+    ]
+
+    star_log = [f"{wildcards.sample}/snv-indels/Log.final.out" for wildcards in samples]
+
+    picard_stats = list()
+    for tool in ["rna_stats", "aln_stats", "insert_stats"]:
+        picard_stats += [
+            f"{wildcards.sample}/snv-indels/{wildcards.sample}.{tool}"
+            for wildcards in samples
+        ]
+
+    vep_stats = [
+        f"{wildcards.sample}/snv-indels/{wildcards.sample}.vep_stats.txt"
+        for wildcards in samples
+    ]
+
+    return star_count + star_log + picard_stats + vep_stats
+
+
 module_output = SimpleNamespace(
     bam=get_bam_output,
     bai=get_bai_output,
     vep_high=get_vep_high,
     vep_target=get_vep_target,
     json=get_json,
     hotspot=get_hotspot,
+    multiqc_files=multiqc_files(),
 )

test/test_hamlet.yml

@@ -156,7 +156,10 @@
         - "sample\tboundary_type\tfuzziness"
         - "SRR8615409\tfuzzy-end\t1"
     - path: "multiqc_hamlet.html"
-
+      contains:
+        - "VEP"
+        - "STAR"
+        - "Picard"
 
 - name: lint-hamlet
   tags:

test/test_snv_indels.yml

@@ -21,6 +21,7 @@
       - "TestSample3/snv-indels/TestSample3.vep.high.txt.gz"
 
       - "TestSample3/snv-indels/TestSample3.hotspot.vcf"
+      - "rule multiqc"
     contains_regex:
       - "STAR .* \"ID:TestSample1\" \"SM:TestSample1\" .* --readFilesIn 'test/data/fastq/SRR8615409 chrM_1.fastq.gz' 'test/data/fastq/SRR8615409 chrM_2.fastq.gz'"
       # Test that params.min_intron_size is used
@@ -118,6 +119,20 @@
       contains:
         - "allele_string"
 
+    # Test that file names are set correctly in MultiQC report
+    - path: multiqc_snv_indels.html
+      must_not_contain:
+        - "ReadsPerGene.out.tab"
+        - "Log.final.out"
+        - "aln_stats"
+        - "insert_stats"
+        - "rna_stats"
+        - "vep_stats.txt"
+      contains:
+        - "VEP"
+        - "STAR"
+        - "Picard"
+
 # VEP should switch to offline mode when "vep_cache" is specified
 - name: test-chrM-vep-cache
   tags: