Reconstructing the domestication process of the white lupin (Lupinus albus) in the Mediterranean basin using pangenome graphs

Domestication was a human-mediated process promoted by the transition from hunter-gatherer to farmers in the Holocene that reshaped the phenotypic and genotypic architecture of many species, including animals, plants and bacteria. The white lupin (Lupinus albus) is an example of such a domestication process in the Old World that was probably started by ancient Greeks ∼4,000 years ago and then conducted locally in all Mediterranean basin, including the Italian peninsula. Fundamental questions related to the domestication of the species (when, where and how many times the domestication process happened) have not been answered yet. Structural variants are an emerging source of genetic variation that may have facilitated the local adaptation of plants to different environments but they are difficult to assess using traditional Whole Genome Sequencing (WGS) approaches due to the intrinsic limitations of short reads but especially to the bias caused by using a single reference genome, an issue that still persist even using long reads. Multiple whole-genomes organized in a pangenome graph are a promising resource to overcome these issues, with the potential to uncover the hidden genomic diversity related to the complex evolutionary changes associated with a domestication process. Using whole genomes of several wild/landraces white lupin accessions and state-of-the-art sequencing and assembly approaches, we propose here a comprehensive project with four major aims:
i) create a pangenome graph serving as a proper reference structure
ii) discover a comprehensive set of genomic variants in a large sample of white lupins, avoiding any form of reference bias
iii) reconstruct the domestication history of the species
iv) identify the genomic loci that were more affected by the artificial selection
We aim to assemble a pangenome graph using 16 de novo assembled genomes at chromosomal level, being representative of the genetic diversity of the white lupin in Europe. The WGS data of ~350 already-sequenced accessions, made available to us by the EU Horizon 2020 INCREASE project, will be then mapped to the pangenome graph and used to discover single nucleotide (SNPs) and presence/absence variants (PAVs) within each genome. Population genetic approaches will be used to characterize the fine-scale population structure, to identify the origin of the domesticated founder population and to estimate the best supported demographic history associated with the domestication events. We will finally perform a selection scan to identify the genomic regions showing the footprints of the artificial selection. Candidate loci will be a crucial resource in the hands of breeders to improve Italian and European varieties for important traits such as adaptation to different agro-environmental conditions. The results will be disseminated to stakeholders (farmers, breeders, citizens) to promote the awareness of the importance of the conservation and use of plant genetic resources.