A fast-and-robust profiler for improving polymerase chain reaction diagnostics

Taguchi-generalized regression neural network micro-screening for physical and sensory characteristics of bread

Generalized regression neural networks (GRNN) may act as crowdsourcing cognitive agents to screen small, dense and complex datasets. The concurrent screening and optimization of several complex physical and sensory traits of bread is developed using a structured Taguchi-type micro-mining technique. A novel product outlook is offered to industrial operations to cover separate aspects of smart product design, engineering and marketing. Four controlling factors were selected to be modulated directly on a modern production line: 1) the dough weight, 2) the proofing time, 3) the baking time, and 4) the oven zone temperatures. Concentrated experimental recipes were programmed using the Taguchi-type L9(34) OA-sampler to detect potentially non-linear multi-response tendencies. The fused behavior of the master-ranked bread characteristics behavior was smart sampled with GRNN-crowdsourcing and robust analysis. It was found that the combination of the oven zone temperatures to play a highly influential role in all investigated scenarios. Moreover, the oven zone temperatures and the dough weight appeared to be instrumental when attempting to synchronously adjusting all four physical characteristics. The optimal oven-zone temperature setting for concurrent screening-and-optimization was found to be 270-240 °C. The optimized (median) responses for loaf weight, moisture, height, width, color, flavor, crumb structure, softness, and elasticity are: 782 g, 34.8 %, 9.36 cm, 10.41 cm, 6.6, 7.2, 7.6, 7.3, and 7.0, respectively.

Development of a hydrolysis probe-based real-time assay for the detection of tropical strains of Fusarium oxysporum f. sp. cubense race 4

Fusarium oxysporum f. sp. cubense (Foc) is one of the most important threats to global banana production. Strategies to control the pathogen are lacking, with plant resistance offering the only long-term solution, if sources of resistance are available. Prevention of introduction of Foc into disease-free areas thus remains a key strategy to continue sustainable banana production. In recent years, strains of Foc affecting Cavendish bananas have destroyed plantations in a number of countries in Asia and in the Middle East, and one African country. One vegetative compatibility group (VCG), 01213/16, is considered the major threat to bananas in tropical and subtropical climatic conditions. However, other genetically related VCGs, such as 0121, may potentially jeopardize banana cultures if they were introduced into disease-free areas. To prevent the introduction of these VCGs into disease-free Cavendish banana-growing countries, a real-time PCR test was developed to accurately detect both VCGs. A previously described putative virulence gene was used to develop a specific combination of hydrolysis probe/primers for the detection of tropical Foc race 4 strains. The real-time PCR parameters were optimized by following a statistical approach relying on orthogonal arrays and the Taguchi method in an attempt to enhance sensitivity and ensure high specificity of the assay. This study also assessed critical performance criteria, such as repeatability, reproducibility, robustness, and specificity, with a large including set of 136 F. oxysporum isolates, including 73 Foc pathogenic strains representing 24 VCGs. The validation data demonstrated that the new assay could be used for regulatory testing applications on banana plant material and can contribute to preventing the introduction and spread of Foc strains affecting Cavendish bananas in the tropics.