fix bug sample merging

Snakemake/1000GB/Snakefile

@@ -2,7 +2,7 @@
 
 """
 __author__ = "Kenza BAZI KABBAJ - Sigenae"
-__version__ = "1.0.0"
+__version__ = "1.0.1"
 Aim: A Snakemake workflow to process paired-end or/and single end WGS data.
 Project: 1000GBovins
 Run locally: snakemake -p
@@ -42,8 +42,7 @@ rule_resources = {
     "bwa_mem_pe" : ["cpu"],
     "bwa_mem_se_R1" : ["cpu"],
     "bwa_mem_se_R2" : ["cpu"],
-    "sample_merge_SE" : ["mem"],
-    "sample_merge_PE" : ["mem"],
+    "sample_merge" : ["mem"],
     "callableLoci" : ["mem"],
     "depthOfCoverage" : ["mem"],
     "markduplicates" : ["mem"],
@@ -126,8 +125,6 @@ include:
     include_prefix + "rules/trim_pe.rules"
 include:
     include_prefix + "rules/bwa_mem_pe.rules"
-include:
-    include_prefix + "rules/sample_merge_PE.rules"
 
 #process single-end reads if available
 include:
@@ -135,7 +132,7 @@ include:
 include:
     include_prefix + "rules/bwa_mem_se.rules"
 include:
-    include_prefix + "rules/sample_merge_SE.rules"
+    include_prefix + "rules/sample_merge.rules"
 
 # bam processing
 include:

Snakemake/1000GB/resources_SLURM.yaml

@@ -47,11 +47,9 @@ bwa_se_sort:
 bwa_se_index:
     cpu: 8
 
-sample_merge_PE:
+sample_merge:
     mem : "40G"
     
-sample_merge_SE:
-    mem : "40G"
     
 markduplicates:
     mem : "20G"

Snakemake/1000GB/rules/sample_merge_PE.rules → Snakemake/1000GB/rules/sample_merge.rules

@@ -7,11 +7,21 @@ __version__ = "1.0.0"
 Merge PE bam files for multiple units into one for the given sample.
 """
 
-rule sample_merge_PE:
+def get_input(wildcards):
+    input=list()
+    if wildcards.sample in PE_samples:
+        for unit in PE_samples[wildcards.sample]:
+            input.append("results/mapping/PE/"+unit+"-pe.sorted.bam")
+            input.append("results/mapping/PE/"+unit+"_1U-se.sorted.bam")
+            input.append("results/mapping/PE/"+unit+"_2U-se.sorted.bam")
+    if wildcards.sample in SE_samples:
+        for unit in SE_samples[wildcards.sample]:
+            input.append("results/mapping/SE/"+unit+"-se.sorted.bam")
+    return input
+
+rule sample_merge:
     input:
-        lambda wildcards: expand("results/mapping/PE/{unit}-pe.sorted.bam", unit=PE_samples[wildcards.sample]),
-        lambda wildcards: expand("results/mapping/PE/{unit}_1U-se.sorted.bam", unit=PE_samples[wildcards.sample]),
-        lambda wildcards: expand("results/mapping/PE/{unit}_2U-se.sorted.bam", unit=PE_samples[wildcards.sample])
+        get_input
     
     output:
         temp("results/sample_merge/{sample}.sorted.bam")
@@ -21,7 +31,7 @@ rule sample_merge_PE:
         stderr="results/sample_merge/logs/{sample}_merge_PE.stderr"
     
     params:
-        mem=config["sample_merge_PE"]["mem"]
+        mem=config["sample_merge"]["mem"]
 
     
     run:

Snakemake/1000GB/rules/sample_merge_SE.rules

-# coding: utf-8
-
-__author__ = "Kenza BAZI KABBAJ - Sigenae"
-__version__ = "1.0.0"
-
-"""
-Merge SE bam files for multiple units into one for the given sample.
-"""
-
-rule sample_merge_SE:
-    input:
-        lambda wildcards: expand("results/mapping/SE/{unit}-se.sorted.bam", unit=SE_samples[wildcards.sample])
-        
-    output:
-        temp("results/sample_merge/{sample}.sorted.bam")
-        
-    log: 
-        stdout="results/sample_merge/logs/{sample}_merge_SE.stdout",
-        stderr="results/sample_merge/logs/{sample}_merge_SE.stderr"
-
-    params:
-        mem=config["sample_merge_SE"]["mem"]
-    
-    run:
-        reads=" I=".join(input)
-        shell("java -Xmx{params.mem} -jar {config[picard]} MergeSamFiles I={reads} VALIDATION_STRINGENCY=LENIENT ASSUME_SORTED=true MERGE_SEQUENCE_DICTIONARIES=true O={output} > {log.stdout} 2> {log.stderr}")