Open AccessDiscovery Proteomics and Nonparametric Modeling Pipeline in the Development of a Candidate Biomarker Panel for Dengue Hemorrhagic Fever
Author(s) -
Brasier Allan R.,
Garcia Josefina,
Wiktorowicz John E.,
Spratt Heidi M.,
Comach Guillermo,
Ju Hyunsu,
Recinos Adrian,
Soman Kizhake,
Forshey Brett M.,
Halsey Eric S.,
Blair Patrick J.,
Rocha Claudio,
Bazan Isabel,
Victor Sundar S.,
Wu Zheng,
Stafford Susan,
Watts Douglas,
Morrison Amy C.,
Scott Thomas W.,
Kochel Tadeusz J.
Publication year - 2012
Publication title -
clinical and translational science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.303
H-Index - 44
eISSN - 1752-8062
pISSN - 1752-8054
DOI - 10.1111/j.1752-8062.2011.00377.x
Subject(s) - dengue fever , biomarker , medicine , nonparametric statistics , receiver operating characteristic , prospective cohort study , intensive care medicine , immunology , statistics , biology , biochemistry , mathematics
Secondary dengue viral infection can produce capillary leakage associated with increased mortality known as dengue hemorrhagic fever (DHF). Because the mortality of DHF can be reduced by early detection and intensive support, improved methods for its detection are needed. We applied multidimensional protein profiling to predict outcomes in a prospective dengue surveillance study in South America. Plasma samples taken from initial clinical presentation of acute dengue infection were subjected to proteomics analyses using ELISA and a recently developed biofluid analysis platform. Demographics, clinical laboratory measurements, nine cytokines, and 419 plasma proteins collected at the time of initial presentation were compared between the DF and DHF outcomes. Here, the subject's gender, clinical parameters, two cytokines, and 42 proteins discriminated between the outcomes. These factors were reduced by multivariate adaptive regression splines (MARS) that a highly accurate classification model based on eight discriminant features with an area under the receiver operator curve (AUC) of 0.999. Model analysis indicated that the feature–outcome relationship were nonlinear. Although this DHF risk model will need validation in a larger cohort, we conclude that approaches to develop predictive biomarker models for disease outcome will need to incorporate nonparametric modeling approaches. Clin Trans Sci 2012; Volume #: 1–13