Collaborative Research: TRTech-PGR Developing computational tools for single-cell genomic data analysis and annotation across diverse plant species.

Recently, the plant genomics field has witnessed a rapid growth of the applications of single-cell RNA sequencing technology--a method that can quantify gene expression in individual cells--in a variety of plant species. To gain insight from this vast amount of expression data on genes and their function in the cells from where it was sampled, this project team will develop new computational tools for mining and analyzing both published and newly generated data.

There are several major challenges in analyzing plant single-cell expression data, including (1) determining known and new cell types and their function in species that are not well studied but important for agriculture, and learning plant biology, (2) comparison across species, and (3) a lack of high-quality curated training data for developing computational tools to analyze plant data. To address these challenges, this project team will develop new computational tools for the analysis of single-cell expression data across diverse range of plant species to assess the conservation and divergence and discovering novel cell-types and gene functions.

Science outreach and training activities in bioinformatics will include developing teaching and training course materials to engage researchers, undergraduate and graduate students, and high school students.

The single-cell transcriptomics is generating a huge amount of data from varied species of plants and is leading the community effort in identifying cellular-level transcriptome events and processes. Many of these events and processes are unique to a cell type, species, or a cell’s developmental stage. The use of wide array of current methods, lack of shared resources, and a common or species-specific set of cell-type markers, is creating a bottleneck for inter and intra-specific comparative analysis and knowledge dissemination.

The team proposed, Aim-1: Develop computational tools for multi-reference-based single-cell/nucleus annotation for plants. Upgrade the orthologous marker gene group and develop a method for cell cluster annotation for non-model species along with a data browser for cell cluster visualization and comparison. Aim-2: Develop generative models to improve the resolution of single-cell sequencing data and appropriately analyze data from cells that are not used for training the model.

To compare and validate predictions, single-nucleus, and single-cell (protoplast) transcriptome data from a control experiment on tomato will be used. Aim-3: Develop teaching and training course materials on single-cell data analytics to engage researchers in both online and in-person workshops. Plant genomics data analytics and bioinformatics skills training activities will also be developed to engage high school students.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.