TY - JOUR

T1 - Detecting associations between intact connectomes and clinical covariates using recursive partitioning object-oriented data analysis

AU - Yang, Dake

AU - Deych, Elena

AU - Shands, Berkley

AU - Campbell, Meghan C.

AU - Perlmutter, Joel S.

AU - Petersen, Steve

AU - Schlaggar, Bradley L.

AU - Shannon, William

N1 - Publisher Copyright:
© 2019 John Wiley & Sons, Ltd.

PY - 2019/12/20

Y1 - 2019/12/20

N2 - Many neuroscientists are interested in how connectomes (graphical representations of functional connectivity between areas of the brain) change in relation to covariates. In statistics, changes like this are analyzed using regression, where the outcomes or dependent variables are regressed onto the covariates. However, when the outcome is a complex object, such as connectome graphs, classical regression models cannot be used. The regression approach developed here to work with complex graph outcomes combines recursive partitioning with the Gibbs distribution. We will only discuss the application to connectomes, but the method is generally applicable to any graphical outcome. The method, called Gibbs-RPart, partitions the covariate space into a set of nonoverlapping regions such that the connectomes within regions are more similar than they are to the connectomes in other regions. This paper extends the object-oriented data analysis paradigm for graph-valued data based on the Gibbs distribution, which we have applied previously to hypothesis testing to compare populations of connectomes from distinct groups (see the work of La Rosa et al).

AB - Many neuroscientists are interested in how connectomes (graphical representations of functional connectivity between areas of the brain) change in relation to covariates. In statistics, changes like this are analyzed using regression, where the outcomes or dependent variables are regressed onto the covariates. However, when the outcome is a complex object, such as connectome graphs, classical regression models cannot be used. The regression approach developed here to work with complex graph outcomes combines recursive partitioning with the Gibbs distribution. We will only discuss the application to connectomes, but the method is generally applicable to any graphical outcome. The method, called Gibbs-RPart, partitions the covariate space into a set of nonoverlapping regions such that the connectomes within regions are more similar than they are to the connectomes in other regions. This paper extends the object-oriented data analysis paradigm for graph-valued data based on the Gibbs distribution, which we have applied previously to hypothesis testing to compare populations of connectomes from distinct groups (see the work of La Rosa et al).

KW - connectome

KW - object-oriented data analysis

KW - recursive partitioning

KW - regression

UR - http://www.scopus.com/inward/record.url?scp=85074598128&partnerID=8YFLogxK

U2 - 10.1002/sim.8374

DO - 10.1002/sim.8374

M3 - Article

C2 - 31650580

AN - SCOPUS:85074598128

SN - 0277-6715

VL - 38

SP - 5486

EP - 5496

JO - Statistics in Medicine

JF - Statistics in Medicine

IS - 29

ER -