MCMC-based peak template matching for GCxGC

Comprehensive two-dimensional gas chromatography (GCxGC) is a new technology for chemical separation. Peak template matching is a technique for automatic chemical identification in GCxGC analysis. Peak template matching can be formulated as a largest common point set problem (LCP). Minimizing Hausdorff distances is one of the many techniques proposed for solving the LCP problem. This paper proposes two novel strategies to search the transformation space based on Markov chain Monte Carlo (MCMC) methods. Experiments on seven real data sets indicate that the transformations found by the new algorithms are effective and searching with two Markov chains is much faster than searching with one Markov chain.