A fast SCCA algorithm for big data analysis in brain imaging genetics

Alzheimer’s Disease Neuroimaging Initiative

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Mining big data in brain imaging genetics is an emerging topic in brain science. It can uncover meaningful associations between genetic variations and brain structures and functions. Sparse canonical correlation analysis (SCCA) is introduced to discover bi-multivariate correlations with feature selection. However, these SCCA methods cannot be directly applied to big brain imaging genetics data due to two limitations. First, they have cubic complexity in the size of the matrix involved and are computational and memory intensive when the matrix becomes large. Second, the parameters in an SCCA method need to be fine-tuned in advance. This further dramatically increases the computational time, and gets severe in high-dimensional scenarios. In this paper, we propose two fast and efficient algorithms to speed up the structure-aware SCCA (S2CCA) implementations without modification to the original SCCA models. The fast algorithms employ a divide-and-conquer strategy and are easy to implement. The experimental results, compared with conventional algorithms, show that our algorithms reduce the time usage significantly. Specifically, the fast algorithms improve the computational efficiency by tens to hundreds of times compared to conventional algorithms. Besides, our algorithms yield similar correlation coefficients and canonical loading profiles to the conventional implementations. Our fast algorithms can be easily parallelized to further reduce the computational time. This indicates that the proposed fast scalable SCCA algorithms can be a powerful tool for big data analysis in brain imaging genetics.

Original languageEnglish (US)
Title of host publicationGraphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics - 1st International Workshop, GRAIL 2017 6th International Workshop, MFCA 2017 and 3rd International Workshop, MICGen 2017 Held in Conjunction with MICCAI 2017, Proceedings
PublisherSpringer Verlag
Pages210-219
Number of pages10
Volume10551 LNCS
ISBN (Print)9783319676746
DOIs
StatePublished - 2017
Event1st International Workshop on Graphs in Biomedical Image Analysis, GRAIL 2017, 6th International Workshop on Mathematical Foundations of Computational Anatomy, MFCA 2017 and 3rd International Workshop on Imaging Genetics, MICGen 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 - Quebec City, Canada
Duration: Sep 10 2017Sep 14 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10551 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other1st International Workshop on Graphs in Biomedical Image Analysis, GRAIL 2017, 6th International Workshop on Mathematical Foundations of Computational Anatomy, MFCA 2017 and 3rd International Workshop on Imaging Genetics, MICGen 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017
CountryCanada
CityQuebec City
Period9/10/179/14/17

Fingerprint

Canonical Correlation Analysis
Brain
Data analysis
Fast Algorithm
Imaging
Imaging techniques
Genetic Variation
Divide and conquer
Correlation coefficient
Computational Efficiency
Feature Selection
Mining
Speedup
High-dimensional
Efficient Algorithms
Genetics
Big data
Scenarios
Computational efficiency
Experimental Results

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Alzheimer’s Disease Neuroimaging Initiative (2017). A fast SCCA algorithm for big data analysis in brain imaging genetics. In Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics - 1st International Workshop, GRAIL 2017 6th International Workshop, MFCA 2017 and 3rd International Workshop, MICGen 2017 Held in Conjunction with MICCAI 2017, Proceedings (Vol. 10551 LNCS, pp. 210-219). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10551 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-67675-3_19

A fast SCCA algorithm for big data analysis in brain imaging genetics. / Alzheimer’s Disease Neuroimaging Initiative.

Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics - 1st International Workshop, GRAIL 2017 6th International Workshop, MFCA 2017 and 3rd International Workshop, MICGen 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10551 LNCS Springer Verlag, 2017. p. 210-219 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10551 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Alzheimer’s Disease Neuroimaging Initiative 2017, A fast SCCA algorithm for big data analysis in brain imaging genetics. in Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics - 1st International Workshop, GRAIL 2017 6th International Workshop, MFCA 2017 and 3rd International Workshop, MICGen 2017 Held in Conjunction with MICCAI 2017, Proceedings. vol. 10551 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10551 LNCS, Springer Verlag, pp. 210-219, 1st International Workshop on Graphs in Biomedical Image Analysis, GRAIL 2017, 6th International Workshop on Mathematical Foundations of Computational Anatomy, MFCA 2017 and 3rd International Workshop on Imaging Genetics, MICGen 2017 held in Conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017, Quebec City, Canada, 9/10/17. https://doi.org/10.1007/978-3-319-67675-3_19
Alzheimer’s Disease Neuroimaging Initiative. A fast SCCA algorithm for big data analysis in brain imaging genetics. In Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics - 1st International Workshop, GRAIL 2017 6th International Workshop, MFCA 2017 and 3rd International Workshop, MICGen 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10551 LNCS. Springer Verlag. 2017. p. 210-219. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-67675-3_19
Alzheimer’s Disease Neuroimaging Initiative. / A fast SCCA algorithm for big data analysis in brain imaging genetics. Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics - 1st International Workshop, GRAIL 2017 6th International Workshop, MFCA 2017 and 3rd International Workshop, MICGen 2017 Held in Conjunction with MICCAI 2017, Proceedings. Vol. 10551 LNCS Springer Verlag, 2017. pp. 210-219 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{aa92eae1fd224ec08aabbc9348da0319,
title = "A fast SCCA algorithm for big data analysis in brain imaging genetics",
abstract = "Mining big data in brain imaging genetics is an emerging topic in brain science. It can uncover meaningful associations between genetic variations and brain structures and functions. Sparse canonical correlation analysis (SCCA) is introduced to discover bi-multivariate correlations with feature selection. However, these SCCA methods cannot be directly applied to big brain imaging genetics data due to two limitations. First, they have cubic complexity in the size of the matrix involved and are computational and memory intensive when the matrix becomes large. Second, the parameters in an SCCA method need to be fine-tuned in advance. This further dramatically increases the computational time, and gets severe in high-dimensional scenarios. In this paper, we propose two fast and efficient algorithms to speed up the structure-aware SCCA (S2CCA) implementations without modification to the original SCCA models. The fast algorithms employ a divide-and-conquer strategy and are easy to implement. The experimental results, compared with conventional algorithms, show that our algorithms reduce the time usage significantly. Specifically, the fast algorithms improve the computational efficiency by tens to hundreds of times compared to conventional algorithms. Besides, our algorithms yield similar correlation coefficients and canonical loading profiles to the conventional implementations. Our fast algorithms can be easily parallelized to further reduce the computational time. This indicates that the proposed fast scalable SCCA algorithms can be a powerful tool for big data analysis in brain imaging genetics.",
author = "{Alzheimer’s Disease Neuroimaging Initiative} and Yuming Huang and Lei Du and Kefei Liu and Xiaohui Yao and Risacher, {Shannon L.} and Lei Guo and Andrew Saykin and Li Shen",
year = "2017",
doi = "10.1007/978-3-319-67675-3_19",
language = "English (US)",
isbn = "9783319676746",
volume = "10551 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "210--219",
booktitle = "Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics - 1st International Workshop, GRAIL 2017 6th International Workshop, MFCA 2017 and 3rd International Workshop, MICGen 2017 Held in Conjunction with MICCAI 2017, Proceedings",

}

TY - GEN

T1 - A fast SCCA algorithm for big data analysis in brain imaging genetics

AU - Alzheimer’s Disease Neuroimaging Initiative

AU - Huang, Yuming

AU - Du, Lei

AU - Liu, Kefei

AU - Yao, Xiaohui

AU - Risacher, Shannon L.

AU - Guo, Lei

AU - Saykin, Andrew

AU - Shen, Li

PY - 2017

Y1 - 2017

N2 - Mining big data in brain imaging genetics is an emerging topic in brain science. It can uncover meaningful associations between genetic variations and brain structures and functions. Sparse canonical correlation analysis (SCCA) is introduced to discover bi-multivariate correlations with feature selection. However, these SCCA methods cannot be directly applied to big brain imaging genetics data due to two limitations. First, they have cubic complexity in the size of the matrix involved and are computational and memory intensive when the matrix becomes large. Second, the parameters in an SCCA method need to be fine-tuned in advance. This further dramatically increases the computational time, and gets severe in high-dimensional scenarios. In this paper, we propose two fast and efficient algorithms to speed up the structure-aware SCCA (S2CCA) implementations without modification to the original SCCA models. The fast algorithms employ a divide-and-conquer strategy and are easy to implement. The experimental results, compared with conventional algorithms, show that our algorithms reduce the time usage significantly. Specifically, the fast algorithms improve the computational efficiency by tens to hundreds of times compared to conventional algorithms. Besides, our algorithms yield similar correlation coefficients and canonical loading profiles to the conventional implementations. Our fast algorithms can be easily parallelized to further reduce the computational time. This indicates that the proposed fast scalable SCCA algorithms can be a powerful tool for big data analysis in brain imaging genetics.

AB - Mining big data in brain imaging genetics is an emerging topic in brain science. It can uncover meaningful associations between genetic variations and brain structures and functions. Sparse canonical correlation analysis (SCCA) is introduced to discover bi-multivariate correlations with feature selection. However, these SCCA methods cannot be directly applied to big brain imaging genetics data due to two limitations. First, they have cubic complexity in the size of the matrix involved and are computational and memory intensive when the matrix becomes large. Second, the parameters in an SCCA method need to be fine-tuned in advance. This further dramatically increases the computational time, and gets severe in high-dimensional scenarios. In this paper, we propose two fast and efficient algorithms to speed up the structure-aware SCCA (S2CCA) implementations without modification to the original SCCA models. The fast algorithms employ a divide-and-conquer strategy and are easy to implement. The experimental results, compared with conventional algorithms, show that our algorithms reduce the time usage significantly. Specifically, the fast algorithms improve the computational efficiency by tens to hundreds of times compared to conventional algorithms. Besides, our algorithms yield similar correlation coefficients and canonical loading profiles to the conventional implementations. Our fast algorithms can be easily parallelized to further reduce the computational time. This indicates that the proposed fast scalable SCCA algorithms can be a powerful tool for big data analysis in brain imaging genetics.

UR - http://www.scopus.com/inward/record.url?scp=85029748372&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029748372&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-67675-3_19

DO - 10.1007/978-3-319-67675-3_19

M3 - Conference contribution

AN - SCOPUS:85029748372

SN - 9783319676746

VL - 10551 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 210

EP - 219

BT - Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics - 1st International Workshop, GRAIL 2017 6th International Workshop, MFCA 2017 and 3rd International Workshop, MICGen 2017 Held in Conjunction with MICCAI 2017, Proceedings

PB - Springer Verlag

ER -