diff home/ubuntu/lefse_to_export/lefse2circlader.py @ 1:db64b6287cd6 draft

Modified datatypes
author george-weingart
date Wed, 20 Aug 2014 16:56:51 -0400
parents
children
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/home/ubuntu/lefse_to_export/lefse2circlader.py	Wed Aug 20 16:56:51 2014 -0400
@@ -0,0 +1,43 @@
+#!/usr/bin/env python
+
+from __future__ import with_statement
+
+import sys
+import os
+import argparse
+
+def read_params(args):
+    parser = argparse.ArgumentParser(description='Convert LEfSe output to '
+                        'Circlader input')
+    parser.add_argument(    'inp_f', metavar='INPUT_FILE', nargs='?', 
+                            default=None, type=str, 
+                            help="the input file [stdin if not present]")    
+    parser.add_argument(    'out_f', metavar='OUTPUT_FILE', nargs='?', 
+                            default=None, type=str, 
+                            help="the output file [stdout if not present]")
+    parser.add_argument('-l', metavar='levels with label', default=0, type=int)
+
+    return vars(parser.parse_args()) 
+
+def lefse2circlader(par):
+    finp,fout = bool(par['inp_f']), bool(par['out_f'])
+
+    with open(par['inp_f']) if finp else sys.stdin as inpf:
+        put_bm = (l.strip().split('\t') for l in inpf.readlines()) 
+    biomarkers = [p for p in put_bm if len(p) > 2]
+
+    circ = [    [   b[0],
+                    "" if b[0].count('.') > par['l'] else b[0].split('.')[-1],
+                    b[2],
+                    b[2]+"_col" ] for b in biomarkers]
+
+    with open(par['out_f'],'w') if fout else sys.stdout as out_file:
+        for c in circ:
+            out_file.write( "\t".join( c ) + "\n" )
+
+if __name__ == '__main__':
+    params = read_params(sys.argv)
+    lefse2circlader(params)
+
+
+