starting to modularize

1 files changed, 47 insertions, 0 deletions

diff --git a/quikr_train b/quikr_train
new file mode 100755
index 0000000..9e9caa7
--- /dev/null
+++ b/quikr_train
@@ -0,0 +1,47 @@
+#!/usr/bin/python
+import numpy as np
+import quikr
+import os
+import sys
+import gzip
+from subprocess import *
+import platform
+import argparse
+
+def main():
+  """
+  You can call this script independently, and will save the 
+  trained matrix as a numpy file.
+  example: python quikr-train.py -i input.fasta -k 6 -o trained_matrix.npy
+ 
+  """
+  parser = argparse.ArgumentParser(description=
+  " quikr_train returns a custom trained matrix that can be used with \
+    the quikr function. \n You must supply a kmer. \n ")
+
+  parser.add_argument("-i", "--input", help="training database of sequences (fasta format)", required=True)
+  parser.add_argument("-o", "--output", help="sensing matrix (text file)", required=True)
+  parser.add_argument("-k", "--kmer", help="kmer size (integer)", 
+                      type=int, required=False )
+  parser.add_argument("-z", "--compress", help="compress output (integer)", 
+                      action='store_true', required=False)
+
+  args = parser.parse_args()
+
+  if not os.path.isfile(args.input):
+    parser.error( "Input database not found")
+
+  # call the quikr train function, save the output with np.save
+  matrix = quikr.quikr_train(args.input, args.kmer)
+
+  if args.compress: 
+    output_file = gzip.open(args.output, "wb")
+  else:
+    output_file = open(args.output, "wb")
+
+  np.save(output_file, matrix)
+
+  return 0
+  
+if __name__ == "__main__":
+    sys.exit(main())