A particle filtering algorithm for parameter estimation in real-time biosensor arrays

Biosensor arrays (e.g., DNA microarrays, protein arrays) detect the presence and quantify the amounts of various biomolecules, including nucleic acids, antibodies, and cell receptors. They rely on chemical attraction between the biomolecules of interest (targets) and their molecular complements which serve as biological sensing elements (probes). The attraction between biomolecules leads to binding, in which probes capture target analytes. Recently developed real-time affinity-based biosensors are capable of acquiring the kinetics of the binding process. Molecular binding is a random process which, in this paper, is modeled by a stochastic differential equation observed at discrete points in time, where the observations are corrupted by an additive noise. The target analyte quantification is posed as a parameter estimation problem, and solved using numerical techniques - Markov Chain Monte Carlo (MCMC) method and Particle Filtering (PF). Simulation studies show that these methods complement each other to provide accurate quantification of the targets over a wide range of measurement noise powers.