Dutta's research interests include spatial analysis, machine learning, high-dimensional methods for biological data and predictive plant phenomics. Dutta is working to develop methods and fast matrix-free computations for big data that arise from field trials, gene sequencing, and high-throughput plant phenotyping.

Modern advances in technology have facilitated the rapid and cost-effective collection of data from agricultural experiments. These data include both phenotypic measurements on plants and their genotypic information as well as meteorological and other environmental variables. Often these data are high-dimensional and have complex dependent structure, presenting substantial challenges for statistical analysis and predictive modeling.

Dutta's research group has been developing new statistical methodologies that take into account the biological and environmental information and adjust for spatial and other nuisance sources of variabilities in order to enhance the scientific understanding of complex biological traits as well as increase the accuracy of phenotypic predictions. Specific projects include developing high-dimensional regression methods for genome-wide associations with non-Gaussian responses, deciphering the relationships among multivariate phenotypic traits, and improving the genotype-and-environment inference using spatial adjustments.

Qiu's interests include high-dimensional data analysis and applications in genetic and phenomic studies, causal inference for genetic association studies, and analysis for high-throughput phenotyping data.

High-throughput phenotyping technologies provide large scale image data for plant growth. One area of Qiu's research focuses on building a computation and analysis pipeline for high-throughput phenotyping data. The pipeline performs pre-processing for image data, extracts plant traits and conducts data cleaning in an automatic and efficient way. Another research area is to explore the relationship between plant traits, genetic variables and environmental factors through high-dimensional analysis, and build predictive models for plant growth dynamics.

Krishnamurthy has courtesy appointments in Electrical and Computer Engineering and Human-Computer Interaction at Iowa State. His research interests include geometric modeling, biomechanics, geometric machine learning, computer-aided design (CAD), cybermanufacturing, finite element analysis, and parallel GPU computing. Krishnamurthy is also a member of the AI Institute for Resilient Agriculture (AIIRA), an NSF- and USDA-funded institute that was launched in 2021 and is housed at Iowa State.

Krishnamurthy's collaboration with PSI will focus on creating novel computational methods and tools to geometrically model different plants and develop advanced multiscale methods to model the utilization of environmental resources (sunlight, fertilizer, water, etc.) and gauge their impact on plants. The data from these simulations will provide valuable feedback to plant breeders to assist them in developing more resilient and productive crops.

Dai's research focuses on developing statistical methods for analyzing complex data objects generated across science and engineering, specifically, functional and longitudinal data that are measured over space or time, as well as geometrical data such as directions and shape. He has extensive collaborative experience modeling these types of data, both with Iowa State scholars in plant phenomics, remote sensing, kinesiology, and with outside collaborators in developmental neuroscience.

Using analytic toolkits that he has developed, Dai will investigate problems involving geometrical data such as 2D and 3D orientations of leaves and roots and the plant ionome (the mineral nutrient and trace element composition of an organism or tissue). This research will integrate function-valued traits, longitudinal measurements and geometrical data in predictive phenomics to reveal valuable information that might otherwise go undetected.