|
Abstract:
|
A gene sequence comparison algorithm provides researchers a means for judging the similarity of DNA sequences. Evolutionary changes are tracked by considering the relative similarity of different DNA sequences, whose differences may be the result of possible insertions, deletions, and/or substitutions of nucleotides that can occur over time. The renowned Smith-Waterman algorithm uses a dynamic programming approach along with a tunable scoring function to quantify the goodness-of-fit for alignments of sequences. Scoring-function values using randomly-generated sequences are known to follow an extreme value distribution; further, empirical evidence suggests that scoring-function values using actual DNA sequences that are not "well-matched" also tend to be extreme value. Here, we will investigate the distributions of scores using both artificially-generated and actual DNA sequence data.
|
