Biclustered independent component analysis for complex biomarker and subtype identification from structural magnetic resonance images in schizophrenia

Cota Navin Gupta, Eduardo Castro, Srinivas Rachkonda, Theo G.M. van Erp, Steven Potkin, Judith M. Ford, Daniel Mathalon, Hyo Jong Lee, Bryon A. Mueller, Douglas N. Greve, Ole A. Andreassen, Ingrid Agartz, Andrew R. Mayer, Julia Stephen, Rex E. Jung, Juan Bustillo, Vince Daniel Calhoun, Jessica A. Turner

Research output: Contribution to journalArticle

Abstract

Clinical and cognitive symptoms domain-based subtyping in schizophrenia (Sz) has been critiqued due to the lack of neurobiological correlates and heterogeneity in symptom scores. We, therefore, present a novel data-driven framework using biclustered independent component analysis to detect subtypes from the reliable and stable gray matter concentration (GMC) of patients with Sz. The developed methodology consists of the following steps: source-based morphometry (SBM) decomposition, selection and sorting of two component loadings, subtype component reconstruction using group information-guided ICA (GIG-ICA). This framework was applied to the top two group discriminative components namely the insula/superior temporal gyrus/inferior frontal gyrus (I-STG-IFG component) and the superior frontal gyrus/middle frontal gyrus/medial frontal gyrus (SFG-MiFG-MFG component) from our previous SBM study, which showed diagnostic group difference and had the highest effect sizes. The aggregated multisite dataset consisted of 382 patients with Sz regressed of age, gender, and site voxelwise. We observed two subtypes (i.e., two different subsets of subjects) each heavily weighted on these two components, respectively. These subsets of subjects were characterized by significant differences in positive and negative syndrome scale (PANSS) positive clinical symptoms (p = 0.005). We also observed an overlapping subtype weighing heavily on both of these components. The PANSS general clinical symptom of this subtype was trend level correlated with the loading coefficients of the SFG-MiFG-MFG component (r = 0.25; p = 0.07). The reconstructed subtype-specific component using GIG-ICA showed variations in voxel regions, when compared to the group component. We observed deviations from mean GMC along with conjunction of features from two components characterizing each deciphered subtype. These inherent variations in GMC among patients with Sz could possibly indicate the need for personalized treatment and targeted drug development.

Original languageEnglish (US)
Article number179
JournalFrontiers in Psychiatry
Volume8
Issue numberSEP
DOIs
StatePublished - Sep 26 2017
Externally publishedYes

Fingerprint

Schizophrenia
Magnetic Resonance Spectroscopy
Biomarkers
Prefrontal Cortex
Neurobehavioral Manifestations
Temporal Lobe
Pharmaceutical Preparations
Gray Matter
Therapeutics

Keywords

  • Biclustering
  • Gray matter concentration
  • Group information-guided independent component analysis
  • Independent component analysis
  • Positive and negative syndrome scale symptoms
  • Subtypes

ASJC Scopus subject areas

  • Psychiatry and Mental health

Cite this

Biclustered independent component analysis for complex biomarker and subtype identification from structural magnetic resonance images in schizophrenia. / Gupta, Cota Navin; Castro, Eduardo; Rachkonda, Srinivas; van Erp, Theo G.M.; Potkin, Steven; Ford, Judith M.; Mathalon, Daniel; Lee, Hyo Jong; Mueller, Bryon A.; Greve, Douglas N.; Andreassen, Ole A.; Agartz, Ingrid; Mayer, Andrew R.; Stephen, Julia; Jung, Rex E.; Bustillo, Juan; Calhoun, Vince Daniel; Turner, Jessica A.

In: Frontiers in Psychiatry, Vol. 8, No. SEP, 179, 26.09.2017.

Research output: Contribution to journalArticle

Gupta, CN, Castro, E, Rachkonda, S, van Erp, TGM, Potkin, S, Ford, JM, Mathalon, D, Lee, HJ, Mueller, BA, Greve, DN, Andreassen, OA, Agartz, I, Mayer, AR, Stephen, J, Jung, RE, Bustillo, J, Calhoun, VD & Turner, JA 2017, 'Biclustered independent component analysis for complex biomarker and subtype identification from structural magnetic resonance images in schizophrenia', Frontiers in Psychiatry, vol. 8, no. SEP, 179. https://doi.org/10.3389/fpsyt.2017.00179
Gupta, Cota Navin ; Castro, Eduardo ; Rachkonda, Srinivas ; van Erp, Theo G.M. ; Potkin, Steven ; Ford, Judith M. ; Mathalon, Daniel ; Lee, Hyo Jong ; Mueller, Bryon A. ; Greve, Douglas N. ; Andreassen, Ole A. ; Agartz, Ingrid ; Mayer, Andrew R. ; Stephen, Julia ; Jung, Rex E. ; Bustillo, Juan ; Calhoun, Vince Daniel ; Turner, Jessica A. / Biclustered independent component analysis for complex biomarker and subtype identification from structural magnetic resonance images in schizophrenia. In: Frontiers in Psychiatry. 2017 ; Vol. 8, No. SEP.
@article{e32782d7109f40f0b4efd289d2316cbc,
title = "Biclustered independent component analysis for complex biomarker and subtype identification from structural magnetic resonance images in schizophrenia",
abstract = "Clinical and cognitive symptoms domain-based subtyping in schizophrenia (Sz) has been critiqued due to the lack of neurobiological correlates and heterogeneity in symptom scores. We, therefore, present a novel data-driven framework using biclustered independent component analysis to detect subtypes from the reliable and stable gray matter concentration (GMC) of patients with Sz. The developed methodology consists of the following steps: source-based morphometry (SBM) decomposition, selection and sorting of two component loadings, subtype component reconstruction using group information-guided ICA (GIG-ICA). This framework was applied to the top two group discriminative components namely the insula/superior temporal gyrus/inferior frontal gyrus (I-STG-IFG component) and the superior frontal gyrus/middle frontal gyrus/medial frontal gyrus (SFG-MiFG-MFG component) from our previous SBM study, which showed diagnostic group difference and had the highest effect sizes. The aggregated multisite dataset consisted of 382 patients with Sz regressed of age, gender, and site voxelwise. We observed two subtypes (i.e., two different subsets of subjects) each heavily weighted on these two components, respectively. These subsets of subjects were characterized by significant differences in positive and negative syndrome scale (PANSS) positive clinical symptoms (p = 0.005). We also observed an overlapping subtype weighing heavily on both of these components. The PANSS general clinical symptom of this subtype was trend level correlated with the loading coefficients of the SFG-MiFG-MFG component (r = 0.25; p = 0.07). The reconstructed subtype-specific component using GIG-ICA showed variations in voxel regions, when compared to the group component. We observed deviations from mean GMC along with conjunction of features from two components characterizing each deciphered subtype. These inherent variations in GMC among patients with Sz could possibly indicate the need for personalized treatment and targeted drug development.",
keywords = "Biclustering, Gray matter concentration, Group information-guided independent component analysis, Independent component analysis, Positive and negative syndrome scale symptoms, Subtypes",
author = "Gupta, {Cota Navin} and Eduardo Castro and Srinivas Rachkonda and {van Erp}, {Theo G.M.} and Steven Potkin and Ford, {Judith M.} and Daniel Mathalon and Lee, {Hyo Jong} and Mueller, {Bryon A.} and Greve, {Douglas N.} and Andreassen, {Ole A.} and Ingrid Agartz and Mayer, {Andrew R.} and Julia Stephen and Jung, {Rex E.} and Juan Bustillo and Calhoun, {Vince Daniel} and Turner, {Jessica A.}",
year = "2017",
month = "9",
day = "26",
doi = "10.3389/fpsyt.2017.00179",
language = "English (US)",
volume = "8",
journal = "Frontiers in Psychiatry",
issn = "1664-0640",
publisher = "Frontiers Research Foundation",
number = "SEP",

}

TY - JOUR

T1 - Biclustered independent component analysis for complex biomarker and subtype identification from structural magnetic resonance images in schizophrenia

AU - Gupta, Cota Navin

AU - Castro, Eduardo

AU - Rachkonda, Srinivas

AU - van Erp, Theo G.M.

AU - Potkin, Steven

AU - Ford, Judith M.

AU - Mathalon, Daniel

AU - Lee, Hyo Jong

AU - Mueller, Bryon A.

AU - Greve, Douglas N.

AU - Andreassen, Ole A.

AU - Agartz, Ingrid

AU - Mayer, Andrew R.

AU - Stephen, Julia

AU - Jung, Rex E.

AU - Bustillo, Juan

AU - Calhoun, Vince Daniel

AU - Turner, Jessica A.

PY - 2017/9/26

Y1 - 2017/9/26

N2 - Clinical and cognitive symptoms domain-based subtyping in schizophrenia (Sz) has been critiqued due to the lack of neurobiological correlates and heterogeneity in symptom scores. We, therefore, present a novel data-driven framework using biclustered independent component analysis to detect subtypes from the reliable and stable gray matter concentration (GMC) of patients with Sz. The developed methodology consists of the following steps: source-based morphometry (SBM) decomposition, selection and sorting of two component loadings, subtype component reconstruction using group information-guided ICA (GIG-ICA). This framework was applied to the top two group discriminative components namely the insula/superior temporal gyrus/inferior frontal gyrus (I-STG-IFG component) and the superior frontal gyrus/middle frontal gyrus/medial frontal gyrus (SFG-MiFG-MFG component) from our previous SBM study, which showed diagnostic group difference and had the highest effect sizes. The aggregated multisite dataset consisted of 382 patients with Sz regressed of age, gender, and site voxelwise. We observed two subtypes (i.e., two different subsets of subjects) each heavily weighted on these two components, respectively. These subsets of subjects were characterized by significant differences in positive and negative syndrome scale (PANSS) positive clinical symptoms (p = 0.005). We also observed an overlapping subtype weighing heavily on both of these components. The PANSS general clinical symptom of this subtype was trend level correlated with the loading coefficients of the SFG-MiFG-MFG component (r = 0.25; p = 0.07). The reconstructed subtype-specific component using GIG-ICA showed variations in voxel regions, when compared to the group component. We observed deviations from mean GMC along with conjunction of features from two components characterizing each deciphered subtype. These inherent variations in GMC among patients with Sz could possibly indicate the need for personalized treatment and targeted drug development.

AB - Clinical and cognitive symptoms domain-based subtyping in schizophrenia (Sz) has been critiqued due to the lack of neurobiological correlates and heterogeneity in symptom scores. We, therefore, present a novel data-driven framework using biclustered independent component analysis to detect subtypes from the reliable and stable gray matter concentration (GMC) of patients with Sz. The developed methodology consists of the following steps: source-based morphometry (SBM) decomposition, selection and sorting of two component loadings, subtype component reconstruction using group information-guided ICA (GIG-ICA). This framework was applied to the top two group discriminative components namely the insula/superior temporal gyrus/inferior frontal gyrus (I-STG-IFG component) and the superior frontal gyrus/middle frontal gyrus/medial frontal gyrus (SFG-MiFG-MFG component) from our previous SBM study, which showed diagnostic group difference and had the highest effect sizes. The aggregated multisite dataset consisted of 382 patients with Sz regressed of age, gender, and site voxelwise. We observed two subtypes (i.e., two different subsets of subjects) each heavily weighted on these two components, respectively. These subsets of subjects were characterized by significant differences in positive and negative syndrome scale (PANSS) positive clinical symptoms (p = 0.005). We also observed an overlapping subtype weighing heavily on both of these components. The PANSS general clinical symptom of this subtype was trend level correlated with the loading coefficients of the SFG-MiFG-MFG component (r = 0.25; p = 0.07). The reconstructed subtype-specific component using GIG-ICA showed variations in voxel regions, when compared to the group component. We observed deviations from mean GMC along with conjunction of features from two components characterizing each deciphered subtype. These inherent variations in GMC among patients with Sz could possibly indicate the need for personalized treatment and targeted drug development.

KW - Biclustering

KW - Gray matter concentration

KW - Group information-guided independent component analysis

KW - Independent component analysis

KW - Positive and negative syndrome scale symptoms

KW - Subtypes

UR - http://www.scopus.com/inward/record.url?scp=85030104364&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030104364&partnerID=8YFLogxK

U2 - 10.3389/fpsyt.2017.00179

DO - 10.3389/fpsyt.2017.00179

M3 - Article

C2 - 29018368

AN - SCOPUS:85030104364

VL - 8

JO - Frontiers in Psychiatry

JF - Frontiers in Psychiatry

SN - 1664-0640

IS - SEP

M1 - 179

ER -