Merge branch 'master' of git.wur.nl:unlock/scripts

metagenomics/assembly_bins_readstats.py

@@ -35,9 +35,10 @@ binContigs = set()
 bins = 0
 prev_bin = ""
 for line in open(binContigsFile).readlines():
-    if line.split()[0] != prev_bin: bins+=1
-    binContigs.add(line.strip().split()[1])
-    prev_bin = line.split()[0]
+    if not "unbinned" in line:
+        if line.split()[0] != prev_bin: bins+=1
+        binContigs.add(line.strip().split()[1])    
+        prev_bin = line.split()[0]
 
 total_assembly_size = 0
 mapped_reads = 0

metagenomics/bins_summary.py

+#!/usr/bin/python3
+import sys
+import argparse
+import os
+import pandas as pd
+
+def parse_options():
+  usage = "\nbins_summary.py -b bins_folder -i idxstats_file -f flagstat_file -b busco_summaries_folder -c checkm_report -o output_summary.tsv"
+  description = "Creates a summary of bins from different tools"
+  parser = argparse.ArgumentParser(usage=usage, description=description)
+
+  input_group = parser.add_argument_group('Required arguments')
+  input_group.add_argument("-b","--bin_folder", dest="bin_folder",    help="Folder with bins in fasta format", required=True, metavar="")
+  input_group.add_argument("-d","--bindepths",  dest="bin_depths",     help="MetaBAT2 aggregateDepths file", required=True, metavar="")
+  input_group.add_argument("-B","--busco",      dest="busco_folder",  help="Folder with BUSCO reports", required=False,  metavar="")
+  input_group.add_argument("-c","--checkm",     dest="checkm_report", help="Checkm report file", required=False,  metavar="")
+  input_group.add_argument("-o","--output_file",dest="output_file",   help="Output name", required=True,  metavar="")
+
+  inputs = parser.parse_args()
+  return inputs
+
+def bin_stats(bin_folder):
+  bin_file_list = os.listdir(bin_folder)
+
+  for bin in bin_file_list:
+    bin_file_path = bin_folder+"/"+bin
+
+    # TOTAL SIZE, CONTIG SIZES, GC%
+    total_size = 0
+    contig_sizes = []
+    contig_len = 0
+    GC = 0
+    for line in open(bin_file_path,"r").readlines():
+      if not line.startswith(">"):
+        GC += line.count("G")+line.count("C")
+        total_size += len(line.strip())
+        contig_len += len(line.strip())
+      else:
+        contig_sizes.append(contig_len)
+        contig_len = 0
+
+    GC = round(GC/total_size*100,1)
+    contigs = len(contig_sizes)
+    largest_contig = max(contig_sizes)
+
+    # N50
+    size = []
+    for contig in sorted(contig_sizes, reverse=True):
+      size.append(contig)
+      if sum(size) >= total_size * 0.5:
+          n50 = contig
+          break
+
+    df.loc[bin, ['Contigs','Size','Largest_contig','N50','GC']] = [contigs,total_size,largest_contig,n50,GC]
+
+def bin_depths(bin_depths):
+  for i,line in enumerate(open(bin_depths,"r").readlines()):
+    if i > 0:
+      line = line.strip().split()
+      bin = line[0].split("/")[-1]
+      df.at[bin,'avgDepth'] = line[2]
+
+def read_checkm_report(checkm_report):
+  checkm_df = pd.read_fwf(checkm_report, colspecs="infer",skiprows=3,header=None,infer_nrows=16)
+  # Drop last row because it's dashes 
+  checkm_df.drop(checkm_df.tail(1).index,inplace=True)
+
+  for idx, row in checkm_df.iterrows():
+    df.loc[row[0]+".fa", ['CheckM_Completeness','CheckM_Contamination','CheckM_Strain-heterogeneity']] = [row[11]+"%",row[12]+"%",row[13]+"%"] 
+
+def read_busco(busco_folder):
+  busco_file_list = os.listdir(busco_folder)
+  for busco_file in busco_file_list:
+    if "specific" in busco_file:
+      bin = ".".join(busco_file.split(".")[3:-1])
+      taxonomy = busco_file.split(".")[2]
+      busco_file_path = busco_folder+"/"+busco_file
+      for line in open(busco_file_path):
+        if line.strip().startswith("C"):
+          busco_score = line.strip()
+          df.loc[bin, ['BUSCO_Taxonomy','BUSCO_score']] = [taxonomy,busco_score]
+
+def main(argv):
+  inputs = parse_options()
+
+  out_columns = ['Contigs','Size','Largest_contig','N50','GC','avgDepth','BUSCO_Taxonomy','BUSCO_score','CheckM_Completeness','CheckM_Contamination','CheckM_Strain-heterogeneity']
+  bin_file_list = os.listdir(inputs.bin_folder)
+
+  global df
+  df = pd.DataFrame(columns=out_columns, index=bin_file_list)
+
+  bin_stats(inputs.bin_folder)
+  bin_depths(inputs.bin_depths)
+  read_checkm_report(inputs.checkm_report)
+
+  read_busco(inputs.busco_folder)
+  df.index.name = "Bin"
+  df.to_csv(inputs.output_file, sep="\t")
+
+
+if __name__ == "__main__":
+	main(sys.argv[1:])