mattpedone Goto Github PK

invent

Nonlinear Interactions Varying Coefficient Regression Model (INteractions Varying coefficiENTs)

nlprova

reproduce-tppmx

subject-specific-dm-reg

Building upon the Dirichlet-multinomial regression framework, we develop an high-dimensional Bayesian hierarchical model for microbiota analysis that exploits subject-specific regression coefficients to simultaneously borrow strength across districts and include complex interactions between diet and clinical factors if supported by the data.

treatppmx

Bayesian predictive model for personalized treatment selection for new untreated patients, which leverages known predictive and prognostic biomarkers. In particular, predictive biomarkers are exploited to inform a product partition model with covariates (PPMx) to obtain homogeneous clusters.

vweights

Computes weights for NGG measure