GeCo2: An Optimized Tool for Lossless Compression and Analysis of DNA Sequences

摘要

The development of efficient DNA data compression tools is fundamental for reducing the storage, given the increasing availability of DNA sequences. The importance is also reflected for analysis purposes, given the search for optimized and new tools for anthropological and biomedical applications. In this paper, we describe the characteristics and impact of the GeCo2 tool, an improved version of the GeCo tool. In the proposed tool, we enhanced the mixture of models, where each context model or tolerant context model has now a specific decay factor. Additionally, specific cache-hash sizes and the ability to run only a context model with inverted repeats was developed. A new command line interface, twelve new pre-computed levels, and several optimizations in the code were also included. The results show a compression improvement using less computational resources (RAM and processing time). This new version permits more flexibility for compression and analysis purposes, namely a higher ability of addressing different characteristics of the DNA sequences. The decompression is performed using symmetric computational resources (RAM and time). The GeCo2 is freely available, under GPLv3 license, at https://github.com/pratas/geco2.