The score returned by NWALIGN does not contain intuitive values when I use ambiguous symbols ‘W’ or ‘V’ in the Bioinformatics Toolbox 3.0 (R2007b).
For example when using two short nucleotide sequences that I wish to align: "AGCT" and "AACT". The alignment of these two sequences with NWALIGN gives a score of 7.67. The alignment of "AACT" with its own copy ("AACT") produces a score of 9.33 (maximum score this alignment can have). Now, if I replace the "G" in "AGCT" with the ambiguous symbol "W" that can be either "A" or "T", the alignment should give me a rather high score, since "A" will match with "W". I obtain score 6.67, which is lower that the real mismatch score. Shouldn't a match between "W" and "A" score better than a match between "G" and "A"? Moreover, is the fact that the sequence "AVCT" ("V" stands for any nucleotide except for "A") scores 7.67, which is higher that for alignment between "AWGT"
The following code illustrates the example:
Score = nwalign('AGCT','AACT')Score = nwalign('AACT','AACT')Score = nwalign('AWCT','AACT')Score = nwalign('AVCT','AACT')
Best Answer