Authors:
Anca Bucur
1
;
Jasper van Leeuwen
1
and
Nevenka Dimitrova
2
Affiliations:
1
Philips Research Europe, Netherlands
;
2
Philips Research USA, United States
Keyword(s):
Genomic signal processing, DNA spectral analysis, Spectral alignment, Frequency sorting, Distributed computing.
Related
Ontology
Subjects/Areas/Topics:
Algorithms and Software Tools
;
Bioinformatics
;
Biomedical Engineering
;
Data Mining and Machine Learning
;
Genomics and Proteomics
;
Sequence Analysis
Abstract:
DNA spectral analysis, i.e. the analysis of DNA spectrograms, has been proposed as a method to systematically investigate DNA patterns, which may correspond to relevant biological features. The Frequency Sorting method sorts the sequences in spectral domain based on their frequency content, and detects and groups those sequences exhibiting one or more strong patterns in the same frequencies. In this paper we propose a novel distributed algorithm for Frequency Sorting and report on the performance results of our implementation for the alignment in spectral domain of the human chromosome 21. Distributed Frequency Sorting enables efficient spectral alignment and allows for the easy detection of strong patterns in both single and multiple frequencies.