Whale song analyses using bioinformatics sequence analysis approaches

• Published in: The Journal of the Acoustical Society of America 2005-04, Vol.117 (4_Supplement), p.2470-2470
• Main Authors: Chen, Yian A., Almeida, Jonas S., Chou, Lien-siang
• Format: Article
• Language: English
• Subjects: Marine
• ISSN: 0001-4966; 1520-8524
• DOI: 10.1121/1.4787459

Animal songs are frequently analyzed using discrete hierarchical units, such as units, themes and songs. Because animal songs and bio-sequences may be understood as analogous, bioinformatics analysis tools DNA/protein sequence alignment and alignment-free methods are proposed to quantify the theme similarities of the songs of false killer whales recorded off northeast Taiwan. The eighteen themes with discrete units that were identified in an earlier study [Y. A. Chen, masters thesis, University of Charleston, 2001] were compared quantitatively using several distance metrics. These metrics included the scores calculated using the Smith-Waterman algorithm with the repeated procedure; the standardized Euclidian distance and the angle metrics based on word frequencies. The theme classifications based on different metrics were summarized and compared in dendrograms using cluster analyses. The results agree with earlier classifications derived by human observation qualitatively. These methods further quantify the similarities among themes. These methods could be applied to the analyses of other animal songs on a larger scale. For instance, these techniques could be used to investigate song evolution and cultural transmission quantifying the dissimilarities of humpback whale songs across different seasons, years, populations, and geographic regions. [Work supported by SC Sea Grant, and Ilan County Government, Taiwan.]