This book is the first comprehensive compilation of deliberations on whole genome sequencing of the diploid and tetraploid alfalfa genomes including sequence assembly, gene annotation, and comparative genomics with the model legume genome, functional genomics, and genomics of important agronomic characters. Other chapters describe the genetic diversity and germplasm collections of alfalfa, as well as development of genetic markers and genome-wide association and genomic selection for economical important traits, genome editing, genomics, and breeding targets to address current and future needs. Altogether, the book contains about 300 pages over 16 chapters authored by globally reputed experts on the relevant field in this crop. This book is useful to the students, teachers, and scientists in the academia and relevant private companies interested in genetics, breeding, pathology, physiology, molecular genetics and breeding, biotechnology, and structural and functional genomics. The work is also useful to seed and forage industries.

The genetic structure of complex agronomic traits in alfalfa (Medicago sativa) is not well understood. By crossing the subspecies M. sativa subsp. falcata and M. sativa subsp. sativa, a fullsib F1 population was created from which a genetic linkage map of each parental genome was developed using RFLP and SSR markers. These maps include simplex, duplex, and simplex-simplex alleles along with a number of alleles exhibiting segregation distortion. The inclusion of these more complicated segregation ratios resulted in greater saturation of the genome, a better convergence to eight consensus linkage groups, and a more realistic view of regions of the genome that may not behave normally due to segregation distortion than would have been possible by only using simplex alleles as has been done previously. The population was clonally propagated and grown at three field locations with phenotypic data collected over three years for various agronomic traits, including biomass production, forage height, and forage regrowth. Combining the marker data with the phenotypic data, markers were identified from each parental genome that were associated with these traits, suggesting that both major germplasm sources of cultivated alfalfa contain alleles that may contribute to improved alfalfa cultivars. These results provide a much better understanding of the genomic regions underlying these traits and are an important start in efforts aimed at the use of marker-assisted selection for the improvement of alfalfa cultivars.

The primary goal of the research presented in the following chapters was to assess the genomic basis of the adaptive trait of autumn dormancy in alfalfa using high resolution genetic mapping approaches in diverse germplasm. Previous work in the field has developed protocols to generate vast amounts of sequence based marker data for this species using GBS. However, opportunity remains to extract more information from such data than had been previously possible. In chapter 1, we review some concepts relevant to the rest of this work including basic biology and improvement of alfalfa, the trait of autumn dormancy, and opportunities for molecular markers in alfalfa research. In chapter 2, we employ a recent SNP calling methodology and develop a framework for estimating allele frequencies from pooled sequencing. We use that framework to screen the non-dormant cultivar CUF 101 and populations developed by three cycles of selection for taller and shorter plants in autumn. We validated the robustness of our GBS-derived, population-specific allele frequency estimates using an analytic approach. In chapter 3, we analyze pre- and post-selection populations from an additional five backgrounds together with CUF 101 to seek evidence for loci under selection in germplasm expressing a range of dormancy levels. In chapter 4, we build on recent scientific reports to propose a method for conducting genome-wide association scans from low-coverage autotetraploid sequencing data that is able to account for uncertainty regarding genotypes. We apply this method to a panel of semi-dormant individuals from a commercial breeding program. Using simulations, we show this method to be more powerful than an existing method.

Alfalfa (Medicago sativa L.) is a perennial, outcrossing legume crop predominantly grown for hay, silage, or pasture. Genetic improvement in Alfalfa in terms of hay yield is still comparable to 30 years ago. Under a variety of growing conditions, forage yield in Alfalfa is stymied by biotic and abiotic stresses including heat, salt, drought, and disease. To overcome such stresses, Alfalfa uses a differential gene expression pathway which is under the control of transcription factors that contribute to tolerance of stresses. The Alfalfa breeding program is mainly focused on developing synthetic varieties through recurrent phenotypic selection exploiting additive genetic effects. The production of hybrid Alfalfa breeding programs uses synthetic varieties as the most feasible means for genetic gain. High heterozygosity of the plants and severe inbreeding depression upon selfing precludes the development of inbred lines for hybrid production. However, quantifying inbreeding depression through fitness and vigor traits expressed as weak and strong plants can help map these traits using association study. Identifying these genetic variants paves the way for the elimination of deleterious alleles and eventually the development of inbred alfalfa lines for hybrid production. However, genetic regions identified through association study do not always translate to actual functional proteins as they are not always linked to genes or genetic variants responsible for traits of interest. As the protein's biological function is strongly dependent on its 3D structure, associating proteins directly with phenotype could help assess the effect of mutation on protein function. To understand the role of transcription factors in stress tolerance, we identified and performed transcriptome analysis of Basic-leucine zipper (bZIP) transcription factors that have played a critical role in regulating growth and development and mediating the responses to abiotic stress in several species, including Arabidopsis thaliana, Oryza sativa, Lotus japonicus, and Medicago truncatula. We identified 237 bZIP genes that were differentially expressed in response to ABA, cold, drought, and salt stresses, indicating a likely role in abiotic stress signaling and/or tolerance. These expressions were further validated through RT-qPCR analysis. Next, a genome-wide association study was performed to map genetic loci associated with Alfalfa for plant vigor trait using 534 plants collected from three locations (Washington, Wisconsin, and Utah) over three generations of selfing. These plants were selected based on plant health of strong and weak within the same line. A total of 11 genetic loci were identified using 588,136 Single nucleotide polymorphisms (SNPs). Gene ontology analysis of significant loci associated them with genes involved in stress response, defense responses against pathogens, and plant reproduction. Finally, we attempted the first-ever association study between features from alphafold predicted 3D structure of protein and phenotype, to link non-synonymous mutation to phenotypes. We used 154 genes, including significant genes from the GWAS study, after filtering 591,919 SNPs, to predict protein 3D structures that identified the five significant GWAS hits. However, two more genes with the lowest p-values (Nod 19, Cytochrome P450) were also identified which play key roles in plant growth and development and also in stress tolerance. This association study is a promising way to narrow down causal mutations from SNP GWAS through stringent filtering of SNPs.