Joint sparse canonical correlation analysis for detecting differential imaging genetics modules

Jian Fang, Dongdong Lin, S. Charles Schulz, Zongben Xu, Vince Daniel Calhoun, Yu Ping Wang

Research output: Contribution to journalArticle

Abstract

Motivation: Imaging genetics combines brain imaging and genetic information to identify the relationships between genetic variants and brain activities. When the data samples belong to different classes (e.g. disease status), the relationships may exhibit class-specific patterns that can be used to facilitate the understanding of a disease. Conventional approaches often perform separate analysis on each class and report the differences, but ignore important shared patterns. Results: In this paper, we develop a multivariate method to analyze the differential dependency across multiple classes. We propose a joint sparse canonical correlation analysis method, which uses a generalized fused lasso penalty to jointly estimate multiple pairs of canonical vectors with both shared and class-specific patterns. Using a data fusion approach, the method is able to detect differentially correlated modules effectively and efficiently. The results from simulation studies demonstrate its higher accuracy in discovering both common and differential canonical correlations compared to conventional sparse CCA. Using a schizophrenia dataset with 92 cases and 116 controls including a single nucleotide polymorphism (SNP) array and functional magnetic resonance imaging data, the proposed method reveals a set of distinct SNP-voxel interaction modules for the schizophrenia patients, which are verified to be both statistically and biologically significant.

Original languageEnglish (US)
Pages (from-to)3480-3488
Number of pages9
JournalBioinformatics
Volume32
Issue number22
DOIs
StatePublished - Nov 15 2016
Externally publishedYes

Fingerprint

Canonical Correlation Analysis
Nucleotides
Polymorphism
Brain
Joints
Imaging
Imaging techniques
Module
Data fusion
Single nucleotide Polymorphism
Single Nucleotide Polymorphism
Schizophrenia
Canonical Correlation
Neuroimaging
Lasso
Functional Magnetic Resonance Imaging
Data Fusion
Voxel
Magnetic Resonance Imaging
Penalty

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Computational Theory and Mathematics
  • Computational Mathematics

Cite this

Joint sparse canonical correlation analysis for detecting differential imaging genetics modules. / Fang, Jian; Lin, Dongdong; Schulz, S. Charles; Xu, Zongben; Calhoun, Vince Daniel; Wang, Yu Ping.

In: Bioinformatics, Vol. 32, No. 22, 15.11.2016, p. 3480-3488.

Research output: Contribution to journalArticle

Fang, Jian ; Lin, Dongdong ; Schulz, S. Charles ; Xu, Zongben ; Calhoun, Vince Daniel ; Wang, Yu Ping. / Joint sparse canonical correlation analysis for detecting differential imaging genetics modules. In: Bioinformatics. 2016 ; Vol. 32, No. 22. pp. 3480-3488.
@article{e5e2ebc323b7434fb00f81740579e7c1,
title = "Joint sparse canonical correlation analysis for detecting differential imaging genetics modules",
abstract = "Motivation: Imaging genetics combines brain imaging and genetic information to identify the relationships between genetic variants and brain activities. When the data samples belong to different classes (e.g. disease status), the relationships may exhibit class-specific patterns that can be used to facilitate the understanding of a disease. Conventional approaches often perform separate analysis on each class and report the differences, but ignore important shared patterns. Results: In this paper, we develop a multivariate method to analyze the differential dependency across multiple classes. We propose a joint sparse canonical correlation analysis method, which uses a generalized fused lasso penalty to jointly estimate multiple pairs of canonical vectors with both shared and class-specific patterns. Using a data fusion approach, the method is able to detect differentially correlated modules effectively and efficiently. The results from simulation studies demonstrate its higher accuracy in discovering both common and differential canonical correlations compared to conventional sparse CCA. Using a schizophrenia dataset with 92 cases and 116 controls including a single nucleotide polymorphism (SNP) array and functional magnetic resonance imaging data, the proposed method reveals a set of distinct SNP-voxel interaction modules for the schizophrenia patients, which are verified to be both statistically and biologically significant.",
author = "Jian Fang and Dongdong Lin and Schulz, {S. Charles} and Zongben Xu and Calhoun, {Vince Daniel} and Wang, {Yu Ping}",
year = "2016",
month = "11",
day = "15",
doi = "10.1093/bioinformatics/btw485",
language = "English (US)",
volume = "32",
pages = "3480--3488",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "22",

}

TY - JOUR

T1 - Joint sparse canonical correlation analysis for detecting differential imaging genetics modules

AU - Fang, Jian

AU - Lin, Dongdong

AU - Schulz, S. Charles

AU - Xu, Zongben

AU - Calhoun, Vince Daniel

AU - Wang, Yu Ping

PY - 2016/11/15

Y1 - 2016/11/15

N2 - Motivation: Imaging genetics combines brain imaging and genetic information to identify the relationships between genetic variants and brain activities. When the data samples belong to different classes (e.g. disease status), the relationships may exhibit class-specific patterns that can be used to facilitate the understanding of a disease. Conventional approaches often perform separate analysis on each class and report the differences, but ignore important shared patterns. Results: In this paper, we develop a multivariate method to analyze the differential dependency across multiple classes. We propose a joint sparse canonical correlation analysis method, which uses a generalized fused lasso penalty to jointly estimate multiple pairs of canonical vectors with both shared and class-specific patterns. Using a data fusion approach, the method is able to detect differentially correlated modules effectively and efficiently. The results from simulation studies demonstrate its higher accuracy in discovering both common and differential canonical correlations compared to conventional sparse CCA. Using a schizophrenia dataset with 92 cases and 116 controls including a single nucleotide polymorphism (SNP) array and functional magnetic resonance imaging data, the proposed method reveals a set of distinct SNP-voxel interaction modules for the schizophrenia patients, which are verified to be both statistically and biologically significant.

AB - Motivation: Imaging genetics combines brain imaging and genetic information to identify the relationships between genetic variants and brain activities. When the data samples belong to different classes (e.g. disease status), the relationships may exhibit class-specific patterns that can be used to facilitate the understanding of a disease. Conventional approaches often perform separate analysis on each class and report the differences, but ignore important shared patterns. Results: In this paper, we develop a multivariate method to analyze the differential dependency across multiple classes. We propose a joint sparse canonical correlation analysis method, which uses a generalized fused lasso penalty to jointly estimate multiple pairs of canonical vectors with both shared and class-specific patterns. Using a data fusion approach, the method is able to detect differentially correlated modules effectively and efficiently. The results from simulation studies demonstrate its higher accuracy in discovering both common and differential canonical correlations compared to conventional sparse CCA. Using a schizophrenia dataset with 92 cases and 116 controls including a single nucleotide polymorphism (SNP) array and functional magnetic resonance imaging data, the proposed method reveals a set of distinct SNP-voxel interaction modules for the schizophrenia patients, which are verified to be both statistically and biologically significant.

UR - http://www.scopus.com/inward/record.url?scp=85024486523&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85024486523&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/btw485

DO - 10.1093/bioinformatics/btw485

M3 - Article

C2 - 27466625

AN - SCOPUS:85024486523

VL - 32

SP - 3480

EP - 3488

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 22

ER -