Image compression has been a very fertile field, as can be confirmed by the vast literature available. However, the majority of the image compression methods that can be found are not designed for numerical data and rely on properties that are not defined for symbolic data, such as order relations (e.g., being able to say that A is greater than B). A particular symbolic data that gained some interest in the last years is whole genome alignments. These alignments are 2D DNA data sets which are created by aligning the chromosomes of several species according to a certain criteria. Furthermore, it is well know by the scientific community that genomic data are been produced at a pace that exceeds the growth of the media capacity to store them. The pressure and need to find efficient compression algorithms for genomic data is therefore very high and is being felt worldwide.
The goal of my research work consist on developing sophisticated compression methods that can be applied to genomic data, namely whole genome alignments. One important characteristic of these compression tools is the capability of splitting large files into several parts and compress each one as an independent piece. This is important because it will decrease the encoding/decoding time using a parallel approach. Moreover, this splitting approach allows users with limited storage and processing power, to download/decode a portion of the genomic data instead of all of it.
Details of my publications and my research experience can be found in my resume.
©2009-2015 University of Aveiro
Last Update 2015/31/10 - Luís M. O. Matos