Roxana Alexandridis, Ph.D


Roxana Alexandridis, Ph.D
Senior Principal Scientist
Stemina Biomarker Discovery

We present an overview of projects that involve statistical analysis of data
from biological and biomedical applications. For example, while the discovery of
putative classes and classification to known classes have been often treated as
separate problems, we introduce a unified approach to class discovery and
classification based on multivariate mixture modeling with an unknown number of
components. We apply this method to analyze tumor samples of acute leukemia and
colon cancer. Also, we investigate fluid biomarkers as predictors of conversion to
Alzheimer’s disease (AD) using variable selection by elastic net regularization and
machine learning techniques. Current cerebrospinal fluid or imaging-based single
biomarkers for AD offer reliable identification of patients with underlying AD but
insufficient prediction of the rate of AD progression. We propose a combined plasma
and cerebrospinal fluid signature for the prediction of midterm progression from mild
cognitive impairment to Alzheimer’s disease.
Monday, December 19, 2016 - 3:30am to 4:30am
HSLC, Room 1244
Seminar Abstract: