Vertical lossless genomic data compression tools for assembled genomes: A systematic literature review

Abstract

The recent decrease in cost and time to sequence and assemble of complete genomes created an increased demand for data storage. As a consequence, several strategies for assembled biological data compression were created. Vertical compression tools implement strategies that take advantage of the high level of similarity between multiple assembled genomic sequences for better compression results. However, current reviews on vertical compression do not compare the execution flow of each tool, which is constituted by phases of preprocessing, transformation, and data encoding. We performed a systematic literature review to identify and compare existing tools for vertical compression of assembled genomic sequences. The review was centered on PubMed and Scopus, in which 45726 distinct papers were considered. Next, 32 papers were selected according to the following criteria: to present a lossless vertical compression tool; to use the information contained in other sequences for the compression; to be able to manipulate genomic sequences in FASTA format; and no need prior knowledge. Although we extracted performance compression results, they were not compared as the tools did not use a standardized evaluation protocol. Thus, we conclude that there's a lack of definition of an evaluation protocol that must be applied by each tool.

Design of Unique Auto generated DNA Digital Dictionary using permutations of finite structures for Genome Sequence Compression

“…Kredens KV et.al. [24] presented a detailed systematic literature survey and Tang T and Li J [25] presented a comparative analysis of the existing compression tools for Genome sequences.…”

“…Input data: 11 24 11 43 24 11 Frequency of Input data is given as (u,v) where u represents the data and v represents the frequency of the data. Huffman Tree for the Input data 11 24 11 43 24 11 having the frequency (11,3), (24,2), (43,1) is shown in Fig. 4.…”

“…The Dictionary table created using the Huffman tree is shown in Table 1 Thus, from Table 1, it is clear that the Huffman Encoding replaces the Input data 11 24 11 43 24 11 by 010011100…”

PQSDC: a parallel lossless compressor for quality scores data via sequences partition and run-length prediction mapping

“…The development of short reads and long reads sequencing technologies has greatly reduced the cost of obtaining genomics sequencing data, the price dropping from $5292.390/MB in 2002 to $0.006/MB in 2022 ( Wetterstrand 2023 ). This has propelled rapid advancements in virus tracing, precision diagnosis treatment, and new drug development ( Hernaez et al 2019 , Kredens et al 2020 , Liu and Li 2021 , Sun et al 2023b ). As a result, the growth rate of sequencing data surpasses the Moore’s law ( Schaller 1997 , Hernaez et al 2019 ).…”