Markov Chain Monte Carlo Data Association for Merge and Split Detection in Tracking Protein Clusters

Abstract

Tagging and tracking protein molecules with the help of laser scanning confocal microscope (LSCM) are a key to better understanding of proteomics in diverse aspects. One challenge of tracking multiple green fluorescent protein (GFP) clusters is how to deal with the interaction between multiple objects, namely splitting and merging. In this paper, we propose a framework to track multiple GFP clusters merge and split by using Markov chain Monte Carlo data association (MCMCDA) method combined with asymmetric region matching strategy. The experimental results show that the method is promising.