I need to find the most similar regions of two DNA sequences. I already tried parallel alignment algorithms, but I don't need to find the maximum alignments over the entire length of the sequences. I only need one longest possible sequence that matches the two DNA sequences as closely as possible. Are there any tools for this? I did not find.
What you are trying to do is "local alignment", as opposed to "global" which finds the best alignment over the entire length of the two sequences or "glocal" which match an entire sequence to a fraction of the other (longer) sequence.
The are some parameters that will determine how long the alignment will be, mainly penalties for gaps. When a low gap penalty is implemented the alignment will tend to be longer and with a high penalty you will get a short, highly matching alignment.
There are several tools to find local alignments. The classic one is water in the EMBOSS package, named after the original algorithm - Smith-Waterman. You can set the gap open and gap extend penalties.
Another implementation is with BLAST2seq. You will have a set of parameters you can choose for.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6