GPU accelerated microarray data analysis using random matrix theory

Joey Ingram, Michelle Zhu

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

1 Citation (Scopus)

Abstract

Recent advances in high-throughput genomic technology, such as micro arrays, usually produce vast amounts of gene expression data under many experimental conditions. Analyzing such data is often difficult due to the colossal data size and the intensive computing involved. In addition, many existing analysis tools often require the inference of experienced analysts and subjective judgments. In this paper, we developed a parallel approach based on Random Matrix Theory (RMT) to generate transcription networks using Graphical Processing Units (GPUs). Recently, GPUs have been redesigned into a more unified architecture, which has allowed them to be used more readily in general purpose computing. This architectural advancement has resulted in GPUs becoming easily programmable parallel processors with performance that is vastly superior to CPUs. Our GPU-based approach makes automated micro array data analysis faster, more accurate and noise resistant without engaging remote high performance computing facilities, such as a cluster or supercomputer. The implementation moves some computationally intensive tasks, such as the calculations of Pearson correlation coefficients, tridiagonal reduction, back transformation of eigenvectors, and orthogonal rotation, to the GPU. Experimental results on real micro array datasets show that our GPU implementation runs faster than a CPU version using highly optimized LAPACK routines. The runtime speedup gets higher as the number of genes and sample points in a micro array dataset increases.

Original languageEnglish
Title of host publicationProc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 - Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011
Pages839-844
Number of pages6
DOIs
StatePublished - 24 Nov 2011
Event13th IEEE International Workshop on FTDCS 2011, the 8th International Conference on ATC 2011, the 8th International Conference on UIC 2011 and the 13th IEEE International Conference on HPCC 2011 - Banff, AB, Canada
Duration: 2 Sep 20114 Sep 2011

Publication series

NameProc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 -Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011

Other

Other13th IEEE International Workshop on FTDCS 2011, the 8th International Conference on ATC 2011, the 8th International Conference on UIC 2011 and the 13th IEEE International Conference on HPCC 2011
CountryCanada
CityBanff, AB
Period2/09/114/09/11

Fingerprint

Microarrays
Processing
Program processors
Supercomputers
Transcription
Gene expression
Eigenvalues and eigenfunctions
Genes
Throughput

Cite this

Ingram, J., & Zhu, M. (2011). GPU accelerated microarray data analysis using random matrix theory. In Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 - Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011 (pp. 839-844). [6063085] (Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 -Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011). https://doi.org/10.1109/HPCC.2011.119
Ingram, Joey ; Zhu, Michelle. / GPU accelerated microarray data analysis using random matrix theory. Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 - Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011. 2011. pp. 839-844 (Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 -Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011).
@inproceedings{1fc2414028cf4d1a8439c9f204e4b44d,
title = "GPU accelerated microarray data analysis using random matrix theory",
abstract = "Recent advances in high-throughput genomic technology, such as micro arrays, usually produce vast amounts of gene expression data under many experimental conditions. Analyzing such data is often difficult due to the colossal data size and the intensive computing involved. In addition, many existing analysis tools often require the inference of experienced analysts and subjective judgments. In this paper, we developed a parallel approach based on Random Matrix Theory (RMT) to generate transcription networks using Graphical Processing Units (GPUs). Recently, GPUs have been redesigned into a more unified architecture, which has allowed them to be used more readily in general purpose computing. This architectural advancement has resulted in GPUs becoming easily programmable parallel processors with performance that is vastly superior to CPUs. Our GPU-based approach makes automated micro array data analysis faster, more accurate and noise resistant without engaging remote high performance computing facilities, such as a cluster or supercomputer. The implementation moves some computationally intensive tasks, such as the calculations of Pearson correlation coefficients, tridiagonal reduction, back transformation of eigenvectors, and orthogonal rotation, to the GPU. Experimental results on real micro array datasets show that our GPU implementation runs faster than a CPU version using highly optimized LAPACK routines. The runtime speedup gets higher as the number of genes and sample points in a micro array dataset increases.",
author = "Joey Ingram and Michelle Zhu",
year = "2011",
month = "11",
day = "24",
doi = "10.1109/HPCC.2011.119",
language = "English",
isbn = "9780769545387",
series = "Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 -Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011",
pages = "839--844",
booktitle = "Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 - Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011",

}

Ingram, J & Zhu, M 2011, GPU accelerated microarray data analysis using random matrix theory. in Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 - Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011., 6063085, Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 -Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011, pp. 839-844, 13th IEEE International Workshop on FTDCS 2011, the 8th International Conference on ATC 2011, the 8th International Conference on UIC 2011 and the 13th IEEE International Conference on HPCC 2011, Banff, AB, Canada, 2/09/11. https://doi.org/10.1109/HPCC.2011.119

GPU accelerated microarray data analysis using random matrix theory. / Ingram, Joey; Zhu, Michelle.

Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 - Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011. 2011. p. 839-844 6063085 (Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 -Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011).

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - GPU accelerated microarray data analysis using random matrix theory

AU - Ingram, Joey

AU - Zhu, Michelle

PY - 2011/11/24

Y1 - 2011/11/24

N2 - Recent advances in high-throughput genomic technology, such as micro arrays, usually produce vast amounts of gene expression data under many experimental conditions. Analyzing such data is often difficult due to the colossal data size and the intensive computing involved. In addition, many existing analysis tools often require the inference of experienced analysts and subjective judgments. In this paper, we developed a parallel approach based on Random Matrix Theory (RMT) to generate transcription networks using Graphical Processing Units (GPUs). Recently, GPUs have been redesigned into a more unified architecture, which has allowed them to be used more readily in general purpose computing. This architectural advancement has resulted in GPUs becoming easily programmable parallel processors with performance that is vastly superior to CPUs. Our GPU-based approach makes automated micro array data analysis faster, more accurate and noise resistant without engaging remote high performance computing facilities, such as a cluster or supercomputer. The implementation moves some computationally intensive tasks, such as the calculations of Pearson correlation coefficients, tridiagonal reduction, back transformation of eigenvectors, and orthogonal rotation, to the GPU. Experimental results on real micro array datasets show that our GPU implementation runs faster than a CPU version using highly optimized LAPACK routines. The runtime speedup gets higher as the number of genes and sample points in a micro array dataset increases.

AB - Recent advances in high-throughput genomic technology, such as micro arrays, usually produce vast amounts of gene expression data under many experimental conditions. Analyzing such data is often difficult due to the colossal data size and the intensive computing involved. In addition, many existing analysis tools often require the inference of experienced analysts and subjective judgments. In this paper, we developed a parallel approach based on Random Matrix Theory (RMT) to generate transcription networks using Graphical Processing Units (GPUs). Recently, GPUs have been redesigned into a more unified architecture, which has allowed them to be used more readily in general purpose computing. This architectural advancement has resulted in GPUs becoming easily programmable parallel processors with performance that is vastly superior to CPUs. Our GPU-based approach makes automated micro array data analysis faster, more accurate and noise resistant without engaging remote high performance computing facilities, such as a cluster or supercomputer. The implementation moves some computationally intensive tasks, such as the calculations of Pearson correlation coefficients, tridiagonal reduction, back transformation of eigenvectors, and orthogonal rotation, to the GPU. Experimental results on real micro array datasets show that our GPU implementation runs faster than a CPU version using highly optimized LAPACK routines. The runtime speedup gets higher as the number of genes and sample points in a micro array dataset increases.

UR - http://www.scopus.com/inward/record.url?scp=81555213263&partnerID=8YFLogxK

U2 - 10.1109/HPCC.2011.119

DO - 10.1109/HPCC.2011.119

M3 - Conference contribution

SN - 9780769545387

T3 - Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 -Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011

SP - 839

EP - 844

BT - Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 - Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011

ER -

Ingram J, Zhu M. GPU accelerated microarray data analysis using random matrix theory. In Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 - Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011. 2011. p. 839-844. 6063085. (Proc.- 2011 IEEE International Conference on HPCC 2011 - 2011 IEEE International Workshop on FTDCS 2011 -Workshops of the 2011 Int. Conf. on UIC 2011- Workshops of the 2011 Int. Conf. ATC 2011). https://doi.org/10.1109/HPCC.2011.119