A parallel independent component analysis approach to investigate genomic influence on brain function

Jingyu Liu, Oguz Demirci, Vince Daniel Calhoun

Research output: Contribution to journalArticle

Abstract

Relationships between genomic data and functional brain images are of great interest but require new analysis approaches to integrate the high-dimensional data types. This letter presents an extension of a technique called parallel independent component analysis (paraICA), which enables the joint analysis of multiple modalities including interconnections between them. We extend our earlier work by allowing for multiple interconnections and by providing important overfitting controls. Performance was assessed by simulations under different conditions, and indicated reliable results can be extracted by properly balancing overfitting and underfitting. An application to functional magnetic resonance images and single nucleotide polymorphism array produced interesting findings.

Original languageEnglish (US)
Pages (from-to)413-416
Number of pages4
JournalIEEE Signal Processing Letters
Volume15
DOIs
StatePublished - 2008
Externally publishedYes

Fingerprint

Overfitting
Independent component analysis
Independent Component Analysis
Magnetic resonance
Nucleotides
Polymorphism
Interconnection
Genomics
Brain
Magnetic Resonance Image
Single nucleotide Polymorphism
High-dimensional Data
Balancing
Modality
Integrate
Simulation
Influence
Relationships

Keywords

  • Entropy
  • FMRI
  • Genetic association
  • Independent component analysis (ICA)
  • Multimodal process
  • Parallel ICA

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing
  • Applied Mathematics

Cite this

A parallel independent component analysis approach to investigate genomic influence on brain function. / Liu, Jingyu; Demirci, Oguz; Calhoun, Vince Daniel.

In: IEEE Signal Processing Letters, Vol. 15, 2008, p. 413-416.

Research output: Contribution to journalArticle

@article{5d58f517cff44fb89dc3b5c38dea78f1,
title = "A parallel independent component analysis approach to investigate genomic influence on brain function",
abstract = "Relationships between genomic data and functional brain images are of great interest but require new analysis approaches to integrate the high-dimensional data types. This letter presents an extension of a technique called parallel independent component analysis (paraICA), which enables the joint analysis of multiple modalities including interconnections between them. We extend our earlier work by allowing for multiple interconnections and by providing important overfitting controls. Performance was assessed by simulations under different conditions, and indicated reliable results can be extracted by properly balancing overfitting and underfitting. An application to functional magnetic resonance images and single nucleotide polymorphism array produced interesting findings.",
keywords = "Entropy, FMRI, Genetic association, Independent component analysis (ICA), Multimodal process, Parallel ICA",
author = "Jingyu Liu and Oguz Demirci and Calhoun, {Vince Daniel}",
year = "2008",
doi = "10.1109/LSP.2008.922513",
language = "English (US)",
volume = "15",
pages = "413--416",
journal = "IEEE Signal Processing Letters",
issn = "1070-9908",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - A parallel independent component analysis approach to investigate genomic influence on brain function

AU - Liu, Jingyu

AU - Demirci, Oguz

AU - Calhoun, Vince Daniel

PY - 2008

Y1 - 2008

N2 - Relationships between genomic data and functional brain images are of great interest but require new analysis approaches to integrate the high-dimensional data types. This letter presents an extension of a technique called parallel independent component analysis (paraICA), which enables the joint analysis of multiple modalities including interconnections between them. We extend our earlier work by allowing for multiple interconnections and by providing important overfitting controls. Performance was assessed by simulations under different conditions, and indicated reliable results can be extracted by properly balancing overfitting and underfitting. An application to functional magnetic resonance images and single nucleotide polymorphism array produced interesting findings.

AB - Relationships between genomic data and functional brain images are of great interest but require new analysis approaches to integrate the high-dimensional data types. This letter presents an extension of a technique called parallel independent component analysis (paraICA), which enables the joint analysis of multiple modalities including interconnections between them. We extend our earlier work by allowing for multiple interconnections and by providing important overfitting controls. Performance was assessed by simulations under different conditions, and indicated reliable results can be extracted by properly balancing overfitting and underfitting. An application to functional magnetic resonance images and single nucleotide polymorphism array produced interesting findings.

KW - Entropy

KW - FMRI

KW - Genetic association

KW - Independent component analysis (ICA)

KW - Multimodal process

KW - Parallel ICA

UR - http://www.scopus.com/inward/record.url?scp=58049149827&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58049149827&partnerID=8YFLogxK

U2 - 10.1109/LSP.2008.922513

DO - 10.1109/LSP.2008.922513

M3 - Article

C2 - 19834575

AN - SCOPUS:58049149827

VL - 15

SP - 413

EP - 416

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

SN - 1070-9908

ER -