Research Interests
Unsupervised learning methods
Unsupervised dimension reduction and model-based clustering for high-dimensional data with applications in molecular pattern discovery, biomedical informatics, imaging and neuroscience
Assessment of technical reproducibility and probability-based methods for outlier detection in large-scale biological data
Supervised and semi-supervised learning methods
Analysis of censored survival data
Feature selection and predictive modeling for large-scale genomic data in the presence of censored survival outcomes
Integrative genomics analysis investigating the association between digital gene expression, single nucleotide polymorphisms, copy number variation, methylation and censored survival outcomes
Statistical machine learning methods for biomarker discovery