Robust hidden semi-Markov modeling of array CGH data

Abstract

As an extension to hidden Markov models, the hidden semi-Markov models allow the probability distribution of staying in the same state to be a general distribution. Therefore, hidden semi-Markov models are good at modeling sequences with succession of homogenous zones by choosing appropriate state duration distributions. Hidden semi-Markov models are generative models. Most times they are trained by maximum likelihood estimation. To compensate model misspecification and provide protection against outliers, hidden semi-Markov models can be trained discriminatively given a labeled training set at the expense of increased training complexity. As an alternative to discriminative training, in this paper, we consider model mis-specification and outliers by adopting robust methods. Specifically, we use Student's t mixture models as the emission distributions of hidden semiMarkov models. The proposed robust hidden semi-Markov models are used to model array based comparative genomic hybridization data. Experiments conducted on the benchmark data from the Coriell cell lines, and the glioblastoma multiforme data illustrate the reliability of the technique.

Approximation algorithms for speeding up dynamic programming and denoising aCGH data

“…Our formulation has already proved to be valuable in numerous settings [Ding and Shah 2010]. We show a basic exact dynamic programming algorithm which runs in O(n 2 ) time and performs excellently on both synthetic and real data.…”

biomvRhsmm:Genomic Segmentation with Hidden Semi-Markov Model

“…Various models and computational tools have been developed to handle either the general segmentation problem or particular types of partitioning. Most commonly, the approaches address the detection of chromosomal alterations with array-based comparative genomic hybridization (aCGH) [ 9 – 18 ] or SNP array [ 19 – 23 ], transcript [ 24 , 25 ] and protein-binding site detection [ 26 , 27 ] with tiling array, and the identification of gene expression domains [ 28 , 29 ]. In recent years, more effort has been devoted to the development of computational tools to deal with read-count data generated from next-generation sequencing (NGS) [ 30 – 35 ].…”

“…On the other hand, a hidden semi-Markov model (HSMM), a more generalized form of HMM, could be applied to utilize positional information. Indeed, HSMM was proposed for modeling aCGH data [ 18 ], but the tool did not actually utilize positional information and the implementation is no longer publicly available.…”