r27267 - /trunk/lib/sequence_alignment/needleman_wunsch.py -- January 22, 2015 - 16:09

Author: bugman
Date: Thu Jan 22 16:09:34 2015
New Revision: 27267

URL: http://svn.gna.org/viewcvs/relax?rev=27267&view=rev
Log:
Added sanity checks to the Needleman-Wunsch sequence alignment algorithm.

The residues of both sequences are now checked in needleman_wunsch_align() to 
make sure that they
are present in the substitution matrix.


Modified:
    trunk/lib/sequence_alignment/needleman_wunsch.py

Modified: trunk/lib/sequence_alignment/needleman_wunsch.py
URL: 
http://svn.gna.org/viewcvs/relax/trunk/lib/sequence_alignment/needleman_wunsch.py?rev=27267&r1=27266&r2=27267&view=diff
==============================================================================
--- trunk/lib/sequence_alignment/needleman_wunsch.py    (original)
+++ trunk/lib/sequence_alignment/needleman_wunsch.py    Thu Jan 22 16:09:34 
2015
@@ -24,6 +24,10 @@
 
 # Python module imports.
 from numpy import float32, int16, zeros
+
+# relax module imports.
+from lib.errors import RelaxError
+
 
 # Default scores.
 SCORE_MATCH = 1
@@ -62,6 +66,14 @@
     # The sequence lengths.
     M = len(sequence1)
     N = len(sequence2)
+
+    # Sanity check.
+    for i in range(M):
+        if sequence1[i] not in sub_seq:
+            raise RelaxError("The residue '%s' from the first sequence 
cannot be found in the substitution matrix residues '%s'." % (sequence1[i], 
sub_seq))
+    for j in range(N):
+        if sequence2[j] not in sub_seq:
+            raise RelaxError("The residue '%s' from the second sequence 
cannot be found in the substitution matrix residues '%s'." % (sequence2[j], 
sub_seq))
 
     # Calculate the scoring and traceback matrices.
     matrix, traceback_matrix = needleman_wunsch_matrix(sequence1, sequence2, 
sub_matrix=sub_matrix, sub_seq=sub_seq, gap_open_penalty=gap_open_penalty, 
gap_extend_penalty=gap_extend_penalty)