Alladi Ramakrishnan Hall

Multivariate methods for fusion of multimodal imaging and genetic data in schizophrenia

C. Navin Gupta

Imaging Genetics and Informatics Lab, Georgia State University, Atlanta, USA

Mental illnesses, such as schizophrenia currently lack definitive biological markers and rely primarily on symptom assessments for diagnosis. It is one of the most cryptic and costly mental disorders in terms of human suffering and societal expenditure. This two-fold talk presents some recent results using multivariate technique in a unimodal and multimodal datasets for schizophrenia.
The first part introduces multivariate approach (namely independent component analysis) and applies it to a multisite structural magnetic resonance imaging datasets (C.N.Gupta et al, Schizophrenia Bulletin 2015). This mega-analysis confirmed that the commonly found gray matter loss for patients with schizophrenia in the anterior temporal lobe, insula and medial frontal lobe form a single consistent spatial pattern even across a diverse aggregated and multiple datasets. I will also briefly introduce the ongoing work of clinical subtyping from sMRI data in a single modality using biclustered ICA.
Collecting multiple types of data from the same individual (like sMRI, EEG, FA, genetics, etc) is becoming popular. However combining imaging and genetic data is not easy. The second part of the talk will discuss a multivariate approach called parallel ICA for multimodal fusion analysis of fractional anisotropy (FA) images and single nucleotide polymorphisms (SNP) (C.N.Gupta et al, Frontiers in Human Neuroscience, 2015). The parallel ICA framework identified an FA component which reflected decreased white matter integrity in the forceps major for patients with schizophrenia. The associated SNP component was overrepresented in genes whose products are involved in corpus callosum morphology (e.g., CNTNAP2, NPAS3 and NFIB) as well as canonical pathways of synaptic long term depression and protein kinase A signaling.