Add test for input multiple readgroups

Add test to make sure the markdup and baserecal groups receive the
correct inputs when a sample has multiple readgroups.

Snakefile

@@ -111,7 +111,7 @@ rule align:
 rule markdup:
     """Mark duplicates in BAM file"""
     input:
-        bam = markdup_input_files,
+        bam = sample_bamfiles,
         tmp = ancient("tmp")
     output:
         bam = "{sample}/bams/{sample}.bam",
@@ -119,7 +119,7 @@ rule markdup:
         metrics = "{sample}/bams/{sample}.metrics"
     log: "log/{sample}/markdup.log"
     params:
-        bams=markdup_input_string
+        bams = lambda wc: expand('INPUT={bam}', bam=sample_bamfiles(wc))
     container: containers["picard"]
     shell: "picard -Xmx4G -Djava.io.tmpdir={input.tmp} MarkDuplicates "
            "CREATE_INDEX=TRUE TMP_DIR={input.tmp} "
@@ -136,21 +136,16 @@ def bqsr_bam_input(wildcards):
 rule baserecal:
     """Base recalibrated BAM files"""
     input:
-        bam = lambda wildcards:
-        ("{sample}/bams/{sample}-{read_group}.sorted.bam".format(
-            sample=wildcards.sample, read_group=rg)
-            for rg in get_readgroup(wildcards)),
+        bam = sample_bamfiles,
         ref = config["reference"],
         vcfs = config["known_sites"]
     output: "{sample}/bams/{sample}.baserecal.grp"
     log: "log/{sample}/baserecal.log"
     params:
-        known_sites = " ".join(
-                expand("-knownSites {vcf}", vcf=config["known_sites"])
-        ),
+        known_sites = expand("-knownSites {vcf}", vcf=config["known_sites"]),
         region = f"-L {config['restrict_BQSR']}" if "restrict_BQSR" in config else "",
         gatk_jar = config["gatk_jar"],
-        bams = bqsr_bam_input
+        bams = lambda wc: expand("-I {bam}", bam=sample_bamfiles(wc))
     container: containers["gatk"]
     shell: "java -XX:ParallelGCThreads=1 -jar {params.gatk_jar} -T "
            "BaseRecalibrator {params.bams} -o {output} -nct 8 "

common.smk

@@ -81,10 +81,8 @@ def coverage_files(wildcards):
         files.append(f'{sample}/vcf/{sample}_{threshold}.bed')
     return files
 
-def markdup_input_files(wildcards):
-    """ Determine the input files for markduplicaates
-
-    This is the step where we merge the different per-readgroup bam files
+def sample_bamfiles(wildcards):
+    """ Determine the bam files for a sample (one for each readgroup)
     """
     files = list()
     sample = config['samples'][wildcards.sample]
@@ -92,7 +90,3 @@ def markdup_input_files(wildcards):
     for readgroup in sample['read_groups']:
         files.append(f'{sample_name}/bams/{sample_name}-{readgroup}.sorted.bam')
     return files
-
-def markdup_input_string(wildcards):
-    """Generate the INPUT for each bam file """
-    return [f'INPUT={file} ' for file in markdup_input_files(wildcards)]

tests/test_integration_run.yml

@@ -177,6 +177,9 @@
       - Job counts
       - localrule all
       - (100%) done
+      - BaseRecalibrator -I micro/bams/micro-lib_01.sorted.bam -I micro/bams/micro-lib_02.sorted.bam
+    contains_regex:
+      - MarkDuplicates.*INPUT=micro/bams/micro-lib_01.sorted.bam INPUT=micro/bams/micro-lib_02.sorted.bam
     must_not_contain:
       - rror
   files: