Add mapped read depuplication

Snakefile

@@ -108,10 +108,18 @@ rule all:
                sample = SAMPLES, ref = REFERENCE_NAMES),
         #Paired reads mapped to reference
 
+        expand("data/tmp/Deduplicated-metrics_{sample}_to_{ref}-paired.txt",
+               sample = SAMPLES, ref = REFERENCE_NAMES),
+        #Paired deduplicated mapped reads
+
         expand("data/tmp/Mapped-only_{sample}_to_{ref}-unpaired.bam",
                sample = SAMPLES, ref = REFERENCE_NAMES),
         #Unpaired reads mapped to reference
 
+        expand("data/tmp/Deduplicated-metrics_{sample}_to_{ref}-unpaired.txt",
+               sample = SAMPLES, ref = REFERENCE_NAMES),
+        #Unpaired deduplicated mapped reads
+
         expand("data/tmp/counts/Mapped_read_counts-{sample}_to_{ref}-paired.tsv",
                sample = SAMPLES, ref = REFERENCE_NAMES),
         #Counts of mapped paired reads
@@ -442,6 +450,25 @@ rule filter_unaligned_paired_reads:
 samtools view -b -F 4 -F 256 -F 2048 {input} > {output} 2> {log}
         """
 
+rule deduplicate_paired_reads:
+    input:
+        "data/tmp/Mapped-only_{sample}_to_{ref}-paired.bam"
+    output:
+        bam="data/tmp/Deduplicated-mapped_{sample}_to_{ref}-paired.bam",
+        metrics="data/tmp/Deduplicated-metrics_{sample}_to_{ref}-paired.txt",
+    conda:
+        "envs/picard.yaml"
+    threads: 1
+    log:
+        "log/deduplicate_paired_reads-{sample}_to_{ref}.txt"
+    benchmark:
+        "log/benchmark/deduplicate_paired_reads-{sample}_to_{ref}.txt"
+    shell:
+        """
+picard MarkDuplicates --REMOVE_DUPLICATES -I {input} -O {output.bam} -M {output.metrics}
+        """
+
+
 rule map_unpaired_trimmed_reads:
     input:
         unpaired="data/raw/trimmed_reads/{sample}_unpaired.fq",
@@ -486,6 +513,24 @@ rule filter_unaligned_unpaired_reads:
 samtools view -b -F 4 -F 256 -F 2048 {input} > {output} 2> {log}
         """
 
+rule deduplicate_unpaired_reads:
+    input:
+        "data/tmp/Mapped-only_{sample}_to_{ref}-unpaired.bam"
+    output:
+        bam="data/tmp/Deduplicated-mapped_{sample}_to_{ref}-unpaired.bam",
+        metrics="data/tmp/Deduplicated-metrics_{sample}_to_{ref}-unpaired.txt",
+    conda:
+        "envs/picard.yaml"
+    threads: 1
+    log:
+        "log/deduplicate_unpaired_reads-{sample}_to_{ref}.txt"
+    benchmark:
+        "log/benchmark/deduplicate_unpaired_reads-{sample}_to_{ref}.txt"
+    shell:
+        """
+picard MarkDuplicates --REMOVE_DUPLICATES -I {input} -O {output.bam} -M {output.metrics}
+        """
+
 rule count_mapped_paired_reads:
     input:
         "data/tmp/Mapped-only_{sample}_to_{ref}-paired.bam",

envs/picard.yaml

+name: picard
+
+channels:
+ - default
+ - bioconda
+ - conda-forge
+
+dependencies:
+ - picard=2.23.3