Fix astrocyte bedtools version. Also consolidated blacklist code.

workflow/main.nf

@@ -547,14 +547,14 @@ process consensusPeaks {
         if (blacklist) {
           """
           module load python/3.6.1-2-anaconda
-          module load bedtools/2.26.0
+          module load bedtools/2.25.0
           python3 ${baseDir}/scripts/overlap_peaks.py -d ${peaksDesign} -f ${preDiffDesign} -b ${blacklistFile}
           """
         }
         else {
           """
           module load python/3.6.1-2-anaconda
-          module load bedtools/2.26.0
+          module load bedtools/2.25.0
           python3 ${baseDir}/scripts/overlap_peaks.py -d ${peaksDesign} -f ${preDiffDesign}
           """
         }

workflow/scripts/overlap_peaks.py

@@ -40,8 +40,7 @@ def get_args():
 
     parser.add_argument('-b', '--blacklist',
                         help="Bed file of blacklisted regions to remove",
-                        required=False,
-                        default="None")
+                        required=False)
 
     args = parser.parse_args()
     return args
@@ -240,17 +239,17 @@ def main():
     design_anno = pd.DataFrame(columns=anno_cols)
 
     # Find consenus overlap peaks for each experiment
-    for experiment, df_experiment in design_peaks_df.groupby('experiment_id'):
-        replicated_peak, chr_peak = overlap(experiment, df_experiment)
-        design_diff.loc[design_diff.experiment_id == experiment, "peak"] = replicated_peak
-        design_anno.loc[experiment] = [experiment, chr_peak]
 
-    # Remove blacklist regions; if blacklist = True
-    if os.path.exists(blacklist):
-        for experiment, df_experiment in design_peaks_df.groupby('experiment_id'):
+    for experiment, df_experiment in design_peaks_df.groupby('experiment_id'):
+        # Remove blacklist regions; if blacklist = True
+        if blacklist and os.path.exists(blacklist):
             bl_peaks, bl_chr_peak = blacklist_peaks(experiment, blacklist)
             design_diff.loc[design_diff.experiment_id == experiment, "peak"] = bl_peaks
             design_anno.loc[design_anno.Condition == experiment, "Peaks"] = bl_chr_peak
+        else:
+            replicated_peak, chr_peak = overlap(experiment, df_experiment)
+            design_diff.loc[design_diff.experiment_id == experiment, "peak"] = replicated_peak
+            design_anno.loc[experiment] = [experiment, chr_peak]
 
     # Write out file
     design_diff.columns = ['SampleID',