Statistics and Its Interface

Volume 9 (2016)

Number 1

Identification of significant B cell associations with undetected observations using a Tobit model

Pages: 79 – 91

DOI: https://dx.doi.org/10.4310/SII.2016.v9.n1.a8

Authors

Tian Chen (Department of Biostatistics and Computational Biology, University of Rochester Medical Center, Rochester, New York, U.S.A.)

Shujie Ma (Department of Statistics, University of California at Riverside)

James Kobie (Division of Infectious Diseases, Department of Medicine, University of Rochester Medical Center, Rochester, New York, U.S.A.)

Alexander Rosenberg (Division of Allergy, Immunology, and Rheumatology, Department of Medicine, University of Rochester Medical Center, Rochester, New York, U.S.A.)

Ignacio Sanz (Division of Rheumatology and The Lowance Center for Human Immunology, Emory University, Atlanta, Georgia, U.S.A.)

Hua Liang (Department of Statistics, George Washington University, Washington, District of Columbia, U.S.A.)

Abstract

To study the relationship of serum antibody neutralization activity (determined by IC50) and the B cell immune response, we face two challenges: (i) IC50 values can not be observed when they are below the detected limitation, and (ii) the number of factors is larger than the number of observations. To address these two challenges, we propose a Tobit model for the analysis of the study, and an adaptive LASSO penalized variable selection procedure to identify important factors. Furthermore, we suggest extended Bayesian information criterion for selecting the tuning parameter. Our analysis indicates that three measured B cells, specifically the frequency of CD19+CD20+, CD19-CD20+, and IgD-B220-CD27- peripheral blood B cell subsets have significant effects on IC50.We have also run simulation studies to evaluate the numerical performance of the proposed method.

Keywords

extended Bayesian information criterion, LASSO, penalized likelihood, high-dimensional Tobit model

Full Text (PDF format)

Published 22 October 2015