Vulnerability detection and error minimization in bioassay sample mixing and droplet routing for digital microfluidic biochips

Errors may be intruded in various assay operations because of parametric and physical defects of a biochip, or through undesired external manipulation of the assay operation. In this paper, an approach for error detection and minimization for digital microfluidic biochip is described. Some pathological samples and their corresponding sequence models are chosen as standards to propose possible routing paths/graphs on a biochip. These standard sequence graphs are considered as Golden Execution Models, and any deviation from these models is detected as error. Three possible error models are proposed guided by the fluidic and parametric constraints of the droplet routing. Our aim is to detect any possible error, followed by minimization of these errors. The proposed heuristic tries to minimize (i) electrode usages, (ii) sample execution time/routing time, and (iii) error overhead in subsequent phases of droplet routing. Proposed model is executed on some recent benchmarks and experimental analysis is quite encouraging. The method results in error free routing paths for different bio-assays on account of nominal increase in execution time.