Predicting phenotypes from genotypes holds the great promise to improve the health management in humans and animals, and breeding efficiency in animals and plants. Although many prediction methods have been developed, the challenge remains that the best method shifts due to any of the many factors, including species, environments, populations, and traits of interest. Studies demonstrated that the number of genes underlying a trait and its heritability are the two key factors that determine which methods fit the trait better than others. In most of the cases, however, these two factors are unknown for specific traits of interest. We developed an algorithm with efficient computing time for mining the Maximum Accuracy of Prediction (MMAP) using unsupervised learning on public available real phenotype and genotype data, and simulated phenotype data. MMAP was implemented as a cloud computing platform, which provides the user interface to upload input data, manage projects of analyses, and download the output results. The platform is free for the public to manage phenotypic and genotypic data, and conduct the computation for predicted phenotypes and genetic merit using the best prediction method optimized from many available ones, including Ridge Regression, gBLUP, compressed BLUP, Bayesian LASSO, Bayes A, B, Cpi and many more. Users can also use the platform to conduct data analyses with any methods of their choices. It is expected that extensive usage of MMAP would enrich the training data which in turn rewards the continuously improving the identification of the best method on a specific trait.

Login: MMAP, User Manual, and Demo data.