spBART

How it works

Highlighted claims

• spBART separates low-dimensional demographic covariates from high-dimensional molecular features in a semi-parametric probit BART model. — Semi-Parametric Bayesian Additive Regression Trees for Risk Prediction with High-Dimensional Epigenetic Signatures and Low-Dimensional Covariates
• The model uses a probit latent-variable formulation for binary multiple myeloma status. — Semi-Parametric Bayesian Additive Regression Trees for Risk Prediction with High-Dimensional Epigenetic Signatures and Low-Dimensional Covariates
• The tree component captures nonlinear and interactive associations among 5hmC signatures. — Semi-Parametric Bayesian Additive Regression Trees for Risk Prediction with High-Dimensional Epigenetic Signatures and Low-Dimensional Covariates
• The parametric component estimates additive linear effects for age, sex, race, and BMI. — Semi-Parametric Bayesian Additive Regression Trees for Risk Prediction with High-Dimensional Epigenetic Signatures and Low-Dimensional Covariates