A Segmentation/Clustering Model for the Analysis of Array CGH Data

• Published in: Biometrics Vol. 63; no. 3; pp. 758 - 766
• Main Authors: Picard, F., Robin, S., Lebarbier, E., Daudin, J.-J.
• Format: Journal Article
• Language: English
• Published: Malden, USA Blackwell Publishing Inc 01.09.2007; International Biometric Society; Blackwell Publishing Ltd; Wiley
• Subjects: Algorithms; Array CGH; Artificial Intelligence; Biomedical research; Biometrics; biometry; chromosome mapping; Chromosome Mapping - methods; Cluster Analysis; Comparative genomic hybridization; Computer Simulation; Coordinate systems; data collection; Data Interpretation, Statistical; Databases, Genetic; Datasets; Dynamic programming; EM algorithm; Gene Dosage; Gene Dosage - genetics; genetics; genome; Genomes; Genomics; Heuristics; Information Storage and Retrieval; Information Storage and Retrieval - methods; Life Sciences; Markov chain; methods; Mixture models; Modeling; Models, Genetic; Models, Statistical; Oligonucleotide Array Sequence Analysis; Oligonucleotide Array Sequence Analysis - methods; Other; Parametric models; Pattern Recognition, Automated; Pattern Recognition, Automated - methods; Research methodology; Segmentation; Sequence Alignment; Sequence Alignment - methods; Sequence Analysis, DNA; Sequence Analysis, DNA - methods; Statistical analysis; Supernova remnants
• ISSN: 0006-341X; 1541-0420; 1541-0420
• DOI: 10.1111/j.1541-0420.2006.00729.x

Microarray-CGH (comparative genomic hybridization) experiments are used to detect and map chromosomal imbalances. A CGH profile can be viewed as a succession of segments that represent homogeneous regions in the genome whose representative sequences share the same relative copy number on average. Segmentation methods constitute a natural framework for the analysis, but they do not provide a biological status for the detected segments. We propose a new model for this segmentation/clustering problem, combining a segmentation model with a mixture model. We present a new hybrid algorithm called dynamic programming-expectation maximization (DP-EM) to estimate the parameters of the model by maximum likelihood. This algorithm combines DP and the EM algorithm. We also propose a model selection heuristic to select the number of clusters and the number of segments. An example of our procedure is presented, based on publicly available data sets. We compare our method to segmentation methods and to hidden Markov models, and we show that the new segmentation/clustering model is a promising alternative that can be applied in the more general context of signal processing.