The capability to screen a large number of samples (M) for specific responses to a library of active agents (N) in a manner which is time- and cost- efficient is of critical importance in application areas such as plant diagnostics, crop genotyping and drug discovery. There is great interest in these areas for the identification of specific genes or plant pathogens in crops using DNA markers, DNA traceability for food safety and identification of a specific response of cells to a specific drug. DNA-based methods in the field of point-of-use devices are critical for on-site testing of samples without expensive instrumentation. However, the high cost of reagents and liquid handling robots required to perform vast numbers of pipetting steps significantly hampers the proliferation of key enabling technologies into smaller laboratories. Centrifugal microfluidic devices have emerged as increasingly useful tools for biomedical applications and diagnostics and can be manufactured using inexpensive materials and low cost instrumentation. The integration of DNA amplification methods are used to rapidly generate large amounts of DNA to reliably detect and identify diseases present in plant material which cause economic losses to agribusiness each year. This work demonstrates three microfluidic centrifugal platforms for automating the combinatorial mixing challenge in a simple instrument and demonstrates a reduction of pipetting steps towards large numbers of samples and reagents. These platforms permit MN combinatorial mixing to generate unique sample/reagent outputs in an autonomous manner. These platforms demonstrate highly scalable automation of the liquid handling protocols required for combinatorial screening methods on a simple, spindle-motor based instrument. Furthermore, by significantly reducing the number of pipetting steps, by lowering reagent costs through miniaturisation as well as by accurate metering and widely eliminating human error in liquid handling, our technology meets the requirements of deployment in decentralized settings.