Metody komprese bioinformatických dat pro přenos na HPC infrastrukturu

Abstract

Bioinformatical data are currently one of the most valuable data types for clinical and biological research. The fast development of microscopy techniques leads to more data with higher quality than ever before. This thesis aims to explore methods of compression of bioinformatical data. The compression would result in faster transmission times and smaller requirements for data storage. First, we describe the existing methods of lossless data compression and then we take a closer look at the lossy compression techniques. Specifically, the scalar quantization and vector quantization algorithms. These methods will be tested on real-life data. Results of compression, with a detailed analysis of a lossy compression results and the compression error, will be presented in this work. Obtained results, prove that our algorithms can achieve very good compression for the tested data. We can keep most of the image quality while minimizing the compression ratio.