Performing Department
(N/A)
Non Technical Summary
As climate change and other downstream impacts of human activities continue to cause increased drought and temperature stress, it is essential to find new ways to ensure food security. One innovative approach is to study how feral crops - plants that have escaped from cultivation and successfully established - respond to climate change and compare their genetic makeup to cultivated crops. This information can help identify populations of feral crops that can be sampled and used to supplement existing germplasm collections as well as to identify useful candidate genes for future food production. Feral populations may have unique adaptations to various climatic challenges, such as heat stress or soil salinity, that can benefit food production. This project proposes to use herbarium collections to inform and improve future food security by identifying feral crop populations for germplasm collection and conservation within the National Plant Germplasm System. This approach will highlight the potential of herbaria as a valuable resource for both protecting biodiversity and improving crops in novel ways. By identifying and conserving feral populations of crops, we can tap into untapped genetic diversity and ensure food security in the face of climate change.
Animal Health Component
5%
Research Effort Categories
Basic
95%
Applied
5%
Developmental
0%
Goals / Objectives
The PD's overarching career goal is to become a leading agricultural science researcher at a Research-2 university while concomitantly being an excellent teacher and mentor, affording a non-traditional pool of undergraduate students with unique training opportunities. The proposed project aims to expand our understanding of feralization in vegetable crops, specifically Brassica oleracea and B. rapa, to determine the potential of feral crop populations as sources of genetic diversity for combating persistent and increasing threats to food security, such as climate change. The project aims to investigate the frequency and timing, genetic architecture, potential breeding applications, and environmental conditions that drive feralization. To achieve these goals, the project will utilize natural history collections, ecological niche modeling, artificial intelligence methods, and genomics to identify the environmental and genomic signatures of ferality. The results and materials created will act as starting resources for the PD's own research lab and will ultimately contribute to new ways of thinking about paths to innovation for future food security.Objectives - The PD will devote full-time effort (100%) to the project during the course of the fellowship.1. Defining the ecological niche of feral Brassica oleracea and B. rapa.1.1 - Identification of feral populations of crop species.1.2 - Development of ecological niche models.1.3 - Leveraging AI to capture traits not in environmental datasets.2. Comparative genomics of feralB. oleraceaandB. rapa.2.1 - Re-sequencing feral populations.2.2 - Frequency and pathways to ferality.2.3 - Identifying loci associated with feralization and local adaptation.
Project Methods
The methods described in Aim 1 of the research proposal involve defining the ecological niche of feral Brassica oleracea and B. rapa to better understand ferality and its implications for climate change adaptation. In aim 1.1, locality data from digitized herbarium specimens of B. oleracea and B. rapa will be filtered based on known historic and current cultivation zones, including botanical gardens and farms. Cultivation zones will be determined using text searching of label data and Landsat imagery. In aim 1.2, environmental variables such as elevation, soil characteristics, land use, and bioclimatic variables will be assembled from herbarium records and other sources. Ecological niche models for feral B. oleracea and B. rapa will be produced using Maxent software, which uses machine learning algorithms to predict species distributions based on environmental variables. Models will be projected onto maps of current and future environmental conditions using WorldClim databases. In aim 1.3, AI methods will be used to detect changes in phenology (e.g., flowering and fruiting time) of feral populations using herbarium specimens. The ResNet50 convolutional neural network, which is pre-trained on millions of images, will be used to classify reproductive vs. non-reproductive specimens with high accuracy.The methods proposed in Aim 2 of the study include re-sequencing feral populations of B. oleracea and B. rapa, analyzing the frequency and pathways to ferality, and identifying loci associated with feralization and local adaptation. In Aim 2.1, leaf tissue will be collected from feral populations of B. oleracea and B. rapa across the US, and genomic DNA will be extracted and sequenced using whole genome re-sequencing at Rapid Genomics LLC. The sequencing will result in paired-end reads at a depth of 11x. The sequence data will then be aligned to reference genomes of B. oleracea and B. rapa using the Burrows-Wheeler Aligner (BWA) software, and single nucleotide polymorphisms (SNPs) and insertions/deletions will be called using the Genome Analysis Toolkit (GATK) software. All analyses will be conducted on the HiPerGator server maintained by the University of Florida Research Computing Facility. In Aim 2.2, the newly generated sequence data from feral populations will be combined with previous sequence data from cultivars and wild relatives to assess the relationships between different populations and the frequency of feralization. Two phylogenetic programs, SNPhylo and TreeMix, will assess individual sample relationships and introgression to determine the pathways to ferality (endoferal or exoferal). Four-population (f4) tests will also be used to test the support of inferred migration edges from Treemix. Ancestry proportions and genetic structure of populations will be assessed using fastSTRUCTURE, and population structure, allele frequencies, and individual inbreeding coefficients will be evaluated using PCAangsd. In Aim 2.3, loci associated with feralization and local adaptation will be identified. Differentiation outliers and genetic-environment association tests will be used to identify loci with high differentiation between populations. FST methods will be used to calculate genetic differentiation, and the Bayenv program will be used to analyze correlation tests between allele frequencies and the environmental variables which best describe the niche of feral crops as identified in Aim 1. Variants identified in both methods will be further investigated as potentially valuable for targeted modifications enabling crops to be more locally adapted. Additionally, known genes that regulate leaf size, flowering time, and pathogen resistance will be analyzed using genome-wide association studies (GWAS) to identify potential candidate genes associated with feralization and local adaptation.The main advisor group will be primary mentor Dr. Pam Soltis and collaborating mentors Drs. Doug Soltis and Alex McAlvay. They will assess and evaluate the quality of research and professional development. They will critique the expected outcomes by preparing manuscripts and presentations. The rationale for their role is to ensure that robust and reproducible science is being conducted and all data produced will be stored using FAIR principles (Findable, Accessible, Interoperable, Re-Usable) and rooted in CARE (Collective Benefit, Authority to Control, Responsibility, and Ethics) best practices. Beyond the project mentors, Dr. Lucas Majure (curator of the Florida Museum of Natural History Herbarium) and Dr. Zachary Stansell (curator of the PGRU Vegetable and Hemp Crop Collections) will act as an additional advisory group. The rationale for their role is to ensure that herbarium vouchers and germplasm are deposited promptly and with all associated metadata for long-term storage and community use. Peers at the Florida Museum of Natural History, University of Florida, and New York Botanical Garden will also act as a supplementary advisory group. Presentations at internal seminars will allow for feedback on the proposed research's quality, expected measurable outcomes, and impacts. The rationale for their role is for feedback from different experts. Finally, to ensure that the research is being evaluated by different types of expertise and experience, attendees at national meetings (National Association of Plant Breeders and Plant and Animal Genome) and the Project Directors meeting and Foundational Area meeting are other opportunities for evaluation of the proposed research. The attendees at the meetings are composed of peers and government officials who will give different forms of feedback on the research. Finally, attendance at the Biology and Mathematics Educators (BIOME) Institute will serve as a way to assess and evaluate the quality of the PD's teaching strategies and diversity and inclusion efforts.