On Discovery of Brain-Phenotype Relationships: Detection, Estimation, and Prediction

      A central goal of basic and clinical human neuroscience is to understand how individual differences in phenotypic features relate to variation in neural features, which is vital to parse patient heterogeneity within and across diagnostic categories. Data-driven discovery of linear latent brain-phenotype relationships can be achieved through Canonical Correlation Analysis (CCA) and Partial Least Squares (PLS). Analysis in the modern “big data” regime, with very high-dimensional feature spaces, raises complex computational issues such as overfitting of statistical models. It is therefore essential to establish under which circumstances CCA/PLS methods yield reliable and interpretable brain-phenotype relationships.
      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'


      Subscribe to Biological Psychiatry
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect