Abstract
Epistasis detection represents a fundamental problem in bio-medicine to understand the reasons for occurrence of complex phenotypic traits (diseases) across a population of individuals. Exhaustively examining all possible interactions of multiple Single-Nucleotide Polymorphisms provides the most reliable way to identify accurate solutions, but it is both computationally and memory intensive task. To tackle this challenge, this work proposes a modular and self-adaptive framework for high-performance and energy-efficient epistasis analysis on modern tightly-coupled heterogeneous platforms composed of multi-core CPUs and integrated GPUs. To fully exploit the capabilities of these systems, the proposed framework incorporates both task- and data-parallel approaches specifically tailored to enhance single and multi-objective epistasis detection on each device architecture, along with allowing efficient collaborative execution across all devices. The experimental results show the ability of the proposed framework to handle the heterogeneity of an Intel CPU+iGPU system, achieving performance and energy-efficiency gains of up to 5\(\times \) and 6\(\times \) in different parallel execution scenarios.
This work was supported by Intel Corporation, the FCT (Fundação para a Ciência e a Tecnologia, Portugal) and the ERDF (European Regional Development Fund, EU) through the projects UIDB/50021/2020 and LISBOA-01-0145-FEDER-031901 (PTDC/CCI-COM/31901/2017, HiPErBio). Sergio Santander-Jiménez and Diogo Marques are supported by FCT fellowships under Grants SFRH/BPD/119220/2016 and SFRH/BD/136053/2018.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Given two solutions \(s_1\) and \(s_2\) and n objective functions \(\mathbf {f}(s)=[f_1(s),...,f_n(s)]\), \(s_1\) dominates \(s_2\) iff 1) \(\forall \) i \(\in \) [1, 2, ..., n], \(f_i(s_1)\) is not worse than \(f_i(s_2)\) and 2) \(\exists \) i \(\in \) [1, 2, ..., n] such that \(f_i(s_1)\) is better than \(f_i(s_2)\). Those solutions that are not dominated by any other candidate compose the Pareto-optimal set. The representation of this set in the objective space is commonly designated as Pareto front.
References
Che, K., et al.: Epistasis detection using a permutation-based gradient boosting machine. In: 2016 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pp. 1247–1252. IEEE (2016)
Dinu, I., et al.: SNP-SNP interactions discovered by logic regression explain Crohn’s disease genetics. PloS One 7(10), e43035 (2012)
Fan, R., et al.: Entropy-based information gain approaches to detect and to characterize gene-gene and gene-environment interactions/correlations of complex diseases. Genet. Epidemiol. 35(7), 706–721 (2011)
Gallego-Sánchez, D., et al.: Parallel multi-objective optimization for high-order epistasis detection. In: Ibrahim, S., Choo, K.K.R., Yan, Z., Pedrycz, W. (eds.) ICA3PP 2017. LNCS, vol. 10393, pp. 523–532. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-65482-9_38
Gonçalves, F., et al.: Parallel evolutionary computation for multiobjective gene interaction analysis. J. Comput. Sci. 40(101068), 1–15 (2020)
González-Domínguez, J., Schmidt, B.: GPU-accelerated exhaustive search for third-order epistatic interactions in case-control studies. J. Comput. Sci. 8, 93–100 (2015)
Intel Corporation: The Compute Architecture of Intel® Processor Graphics Gen9 (2015)
Jing, P.J., Shen, H.B.: MACOED: a multi-objective ant colony optimization algorithm for SNP epistasis detection in genome-wide association studies. Bioinformatics 31(5), 634–641 (2015)
Joubert, W., et al.: Attacking the opioid epidemic: determining the epistatic and pleiotropic genetic architectures for chronic pain and opioid addiction. In: SC18: International Conference for High Performance Computing, Networking, Storage and Analysis, pp. 1–57. IEEE (2018)
Kässens, J.C., et al.: High-speed exhaustive 3-locus interaction epistasis analysis on FPGAs. J. Comput. Sci. 9, 131–136 (2015)
Li, X., et al.: Nature-inspired multiobjective epistasis elucidation from genome-wide association studies. IEEE/ACM T. Comput. Biol. (2018). https://doi.org/10.1109/TCBB.2018.2849759
Ritchie, M.D.: Finding the epistasis needles in the genome-wide haystack. In: Moore, Jason H., Williams, Scott M. (eds.) Epistasis. MMB, vol. 1253, pp. 19–33. Springer, New York (2015). https://doi.org/10.1007/978-1-4939-2155-3_2
Schüpbach, T., et al.: FastEpistasis: a high performance computing solution for quantitative trait epistasis. Bioinformatics 26(11), 1468–1469 (2010)
Sun, J., et al.: Hidden risk genes with high-order intragenic epistasis in alzheimer’s disease. J. Alzheimer’s Dis. 41(4), 1039–1056 (2014)
Visscher, P.M., et al.: 10 years of GWAS discovery: biology, function, and translation. Am. J. Hum. Genet. 101(1), 5–22 (2017)
Wan, X., et al.: Boost: a fast approach to detecting gene-gene interactions in genome-wide case-control studies. Am. J. Hum. Genet. 87(3), 325–340 (2010)
Weeks, N.T., et al.: High-performance epistasis detection in quantitative trait gwas. Int. J. High Perform. Comput. Appl. 32(3), 321–336 (2018)
Yang, C., et al.: SNPHarvester: a filtering-based approach for detecting epistatic interactions in genome-wide association studies. Bioinformatics 25(4), 504–511 (2009)
Yang, J.K., et al.: Interactions among related genes of renin-angiotensin system associated with type 2 diabetes. Diab. Care 33(10), 2271–2273 (2010)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Campos, R., Marques, D., Santander-Jiménez, S., Sousa, L., Ilic, A. (2020). Heterogeneous CPU+iGPU Processing for Efficient Epistasis Detection. In: Malawski, M., Rzadca, K. (eds) Euro-Par 2020: Parallel Processing. Euro-Par 2020. Lecture Notes in Computer Science(), vol 12247. Springer, Cham. https://doi.org/10.1007/978-3-030-57675-2_38
Download citation
DOI: https://doi.org/10.1007/978-3-030-57675-2_38
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-57674-5
Online ISBN: 978-3-030-57675-2
eBook Packages: Computer ScienceComputer Science (R0)