.png){kind=icon}
.png){kind=icon}
Return
A multi-objective optimization procedure for solving the high-order epistasis detection problem
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2019.113000.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
There are multiple research works that establish a relationship between Single Nucleotide Polymorphisms (SNPs) and complex diseases. In many cases, these diseases are caused by the interaction of two or more SNPs (epistasis). Therefore, it is important to identify which SNPs lead to the emergence of diseases. However, this problem gets harder when the epistasis order and the number of SNPs in the dataset under study are increased. To tackle this problem, this work presents an implementation of a multi-objective optimization algorithm with a problem-aware offspring process. Furthermore, due to the large amount of data in the SNPs datasets, we have also parallelized this algorithm. We experimentally validate the quality of the proposal under epistasis sizes of 2, 5, and 8 loci, although the application is not limited to those values. Moreover, a thorough comparative study of biological performance with six state-of-the-art methods has been conducted, obtaining better results than all the other biological methods. Our results confirm that, using the proposed approach, large datasets with high-order epistasis sizes can be processed in reasonable time, obtaining solutions of good quality. According to the literature review performed until the acceptance of this article, there are no other authors' works that can process epistasis sizes above 6 loci in large datasets. (C) 2019 Elsevier Ltd. All rights reserved.
Keywords:
Multi-Objective optimization
Evolutionary computation
Single nucleotide polymorphism
High-Order epistasis
Parallelism
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W