Type
Text
Type
Dissertation
Advisor
Zhu, Wei | Xing, Haipeng | Wu, Song | Xu, Jinfeng.
Date
2013-12-01
Keywords
Bounded Complexity Mixture Approximation, Expectation Maximization, Hidden Markov Model, Recurrent Copy Number Alterations | Statistics
Department
Department of Applied Mathematics and Statistics.
Language
en_US
Source
This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.
Identifier
http://hdl.handle.net/11401/77521
Publisher
The Graduate School, Stony Brook University: Stony Brook, NY.
Format
application/pdf | application/vnd.ms-excel
Abstract
With the recent advances in high resolution microarrays and next generation sequencing, DNA copy number can now be profiled in a high throughput global manner. This has enabled the systematic study of DNA copy number alterations in tumors, as well as the profiling of inherited population-wide copy number variants. Studies of DNA copy number usually involve many samples that fall into different groups, e.g. tumor subtype or ethnic group. It is often of interest to find recurrent alterations within each group. We develop a stochastic segmentation model for detecting recurrent DNA copy number alterations in grouped array-CGH data. In our model, the parameter in each regime is a random variable following specific regime-specific distribution. Explicit formulas for posterior means can be used to estimate the signal directly without performing segmentation. We give a linear-time algorithm for fitting this model and for estimating its parameters by expectation maximization. Simulation studies and applications to real grouped array-CGH data illustrate the advantages of the proposed model. | 117 pages
Recommended Citation
Cai, Ying, "A Stochastic Segmentation Model for Recurrent Copy Number Alterations in Grouped array-CGH Data" (2013). Stony Brook Theses and Dissertations Collection, 2006-2020 (closed to submissions). 3326.
https://commons.library.stonybrook.edu/stony-brook-theses-and-dissertations-collection/3326