Abstract: Recent technological advances in biomedical research, such as the next generation sequencing (NGS) technology, have opened more opportunities for scientific discovery of genetic information. The NGS technology is particularly useful in profiling a genome for the analysis of DNA copy number variants (CNVs).   CNVs that are more than 50 base pairs (bps) long are also sometimes referred to as structural changes or structural variants (SVs).  Cancer development, genetic disorders, and many other diseases are usually relevant to SVs on the genome.  

The short sequencing reads data resulting from NGS are massive and information rich.  How to retrieve information from reads of the tumor and reference samples for accurate SV detection has become a computational and statistical challenge.  Interestingly, detecting boundaries of SV regions on a chromosome or a genome can be viewed as a change point problem of detecting sequencing reads or reads ratio changes presented in the NGS data.  We provide statistical change point models to help detect SVs using the sequencing data.  We use a Bayesian approach to incorporate possible parameter changes in the underlying distribution of the sequencing data.  Simulation studies have shown the effectiveness of the methods.  The methods are applied to some publically available NGS datasets, and SV regions on breast tumor cell lines are successfully identified.  

Bio: Dr. Chen is professor of statistics and the chair of the Department of Mathematics and Statistics at the University of Missouri-Kansas City.  She is currently on research leave from her home institution and is a Visiting Scientist at the Stowers Institute for Medical Research in Kansas City.  Her research interests include change point analysis, model selection criteria, applied statistics, statistical genetics, and modeling gene expression (microarray, and sequencing) data. She is the leading author of the book “Parametric Statistical Change Point Analysis” (Birkhaüser, 2000) and “Parametric Statistical Change Point Analysis with Applications to Genetics, Medicine, and Finance” (Birkhaüser, 2012).