Anna Diana Email and Phone Number

Bridge genomics and genetics with predictive analytics.• Manipulate and analyze high-volume, high-dimensionality molecular marker data from next generation sequencing, and investigating genetic variability among hybrids using unsupervised learning techniques including hierarchical clustering, and PCA analysis.• Systems genetics/genomics and bioinformatics research for diagnostic marker and causal gene discovery for quantitative traits, and for prediction of quantitative traits with marker data.• Incorporation of engineered genetic features into seed placement predictive models to improve recommendation confidence.• Prediction of genetic values using mixed linear models accounting for genetic relationships using either pedigree-based matrices (P-BLUP) or marker- based kinship matrices (G-BLUP).• Cross validation of multiple prediction models using different metrics, e.g. Rank corr & AUC• Present findings and recommendations to key decision makers at various management levels using story-telling and compelling visualizations.

• Demonstrated the impact of the number of crosses, number of progenies per cross selection pressure, and breeding strategies on yield by building simulation-based approaches for each maturity and trait, and provided insights into various trends.• Simulated the entire soybean breeding process and Compared the efficiency of different breeding strategies. Demonstrated through validation studies that simulated outcomes were very close to actual observations.• Prepared, summarized, and presented reports for soybean breeders.

Comparative effectiveness of disease management strategies using cover crops• Designed experiments to compare different treatments of cover crops on suppressing soybean diseases. • Developed hypothesis testing and linear mixed model to determine fixed and random effects in different experiment design settings (RCBD, CRD, etc)• Generated generalized Monte Carlo simulation function to determine optimal sample size and experiment power.