An MCMC algorithm for parameter estimation in stochastically modeled real-time biosensor arrays

Biosensor arrays rely on affinity between biomolecules of interest (so-called target analytes) and their molecular complements (so-called probes) to detect the presence and quantify the amounts of various biomolecules. For instance, nucleic acid probes capture their Watson-Crick complements, antibody probes capture antigens, cell receptor probes capture ligands, etc. Real-time affinity-based biosensors are capable of acquiring the kinetics of the molecular binding, which is a random process modeled by a stochastic differential equation. The amounts of the captured targets are observed at discrete points in time, and those measurements are corrupted by noise. A Markov Chain Monte Carlo algorithm for target analyte quantification in stochastically modeled real-time biosensors is derived in this paper. In simulation studies where we test the robustness with respect to the measurement noise, the proposed technique significantly outperforms previously proposed methods.