1. Introduction
Microfluidic biochips are widely used for high-throughput analysis of mutations, expression profiles, and in identification of microorganisms [1] and are based on highly selective solid-phase nucleic acid hybridization [2]. The sensitivity of these biochips strongly depends on hybridization kinetics, temperature of hybridization, target concentration, surface probe concentration, washing protocol, and other factors [3,4,5]. The role and optimal selection of these parameters is critical for developing efficient hybridization protocols for microfluidic biochips [6,7]. Numerous experimental investigations and theoretical models have been reported for describing the kinetics and thermodynamics of solid-phase nucleic acid hybridization [8,9,10]. However, besides the pioneering work reported on computer simulations of reaction–diffusion kinetics [11,12] providing a sound theoretical understanding of these phenomenon, modeling and experimental validation of DNA:DNA hybridization in biochips is generally lacking. 
In this work, a three-dimensional mathematical model for selective nucleic acid hybridization in a microfluidic biochip was developed followed by comparison of theoretical predictions with experimental data. The biochip used was developed by Xeotron, now owned by Invitrogen (Carlsbad, CA, USA) and has been previously described [13]. To model and simulate hybridization kinetic curves, a finite element software package (FEMLAB, a package for COMSOL Multiphysics; COMSOL, Inc., Burlington, MA, USA) was used. The model accounts for fluid flow, convection and diffusion in the channel as well as diffusion on the reaction surface. The model was used to predict hybridization kinetics of eighteen 20-mer probes targeting vancomycin resistant genes. This gene was chosen as a model because emergence of antibiotic resistance associated with pathogenic bacteria is an alarming global issue [14,15] and numerous hybridization-based biochip platforms are being developed to diagnose multi-drug resistant bacteria [16,17,18,19,20]. Effect of target concentration, temperature, recirculation flow rate and rate constants on the hybridization kinetics, with target concentration, rate constants, and temperature being the most important parameters affecting the rate of hybridization were investigated. Recirculation flow rate was found to play a vital role in determining whether the hybridization system was transport- or reaction-limited. The predictive capability and performance of the model and its potential for application in optimizing hybridization protocols and chip geometry were considered adequate with some exceptions.