Statistics Seminar: Sue Wilson -- Evaluation of Similarity Between Two Sequences

DATE: Friday 14th March at 2:00pm.
LOCATION: University of Sydney, Carslaw 375.
SPEAKER: Susan Wilson (Australia National University) 

Title: Evaluation of Similarity Between Two Sequences 

Abstract: The question ”How should we measure and evaluate the similarity between two
sequences?” is the focus of this presentation.  A commonly encountered situation,
particularly in biology, is to have a query sequence and want to find which,say DNA or
protein, sequences in a large database have ”significant” similarity to this query
sequence.  The widely accepted solution to the question is based on the notion of
alignment, and the BLAST program is the most frequently used method.  There are very
many situations though for which the inherent assumption underlying the use of local
alignment methods is violated.  Hence there has been much interest in development of
alignment-free sequence comparison algorithms.  Arguably the best of these is the number
of k-words shared between two sequences.  The statistic, called D2, is simple and
extremely fast to compute.  Its distribution and asymptotic properties are being
studied, and recent results and unsolved problems will be overviewed.