RESUMEN
Trait diversity is crucial in undertaking the domestication of useful species such as Vitellaria paradoxa which makes a significant contribution to the rural household economy in Africa. This study aims to document the criteria farmers use to distinguish shea trees; how they vary according to age, education level and sociolinguistic group; and their perception of trees' abundance and production. We surveyed 405 respondents across shea parklands in Benin using a semi-structured questionnaire. We used the Kruskal-Wallis test to evaluate the influence of sociodemographic attributes on relative criteria citation frequency and principal components analysis to characterize farmers' perception on morphotypes' abundance, fruits, and butter yields. The five most cited criteria were fruit size (55.5%), tree fertility (15.40%), bark colour (10.51%), timing of production (5.38%), and pulp taste (3.42%). The citation frequency of criteria varied significantly depending on the sociodemographic factors considered. Trees having small fruit ('Yanki') were reported to be widespread and high fruit/nuts and butter producers. Farmers perceived five important traits with variable importance depending on the sociocultural factors studied. This finding is a key step toward the development of a shea improvement program that could focus on the morphotype Yanki reported to potentially be a high fruit and butter producer.
RESUMEN
A defining component of agroforestry parklands across Sahelo-Sudanian Africa (SSA), the shea tree (Vitellaria paradoxa) is central to sustaining local livelihoods and the farming environments of rural communities. Despite its economic and cultural value, however, not to mention the ecological roles it plays as a dominant parkland species, shea remains semi-domesticated with virtually no history of systematic genetic improvement. In truth, shea's extended juvenile period makes traditional breeding approaches untenable; but the opportunity for genome-assisted breeding is immense, provided the foundational resources are available. Here we report the development and public release of such resources. Using the FALCON-Phase workflow, 162.6 Gb of long-read PacBio sequence data were assembled into a 658.7 Mbp, chromosome-scale reference genome annotated with 38,505 coding genes. Whole genome duplication (WGD) analysis based on this gene space revealed clear signatures of two ancient WGD events in shea's evolutionary past, one prior to the Astrid-Rosid divergence (116-126 Mya) and the other at the root of the order Ericales (65-90 Mya). In a first genome-wide look at the suite of fatty acid (FA) biosynthesis genes that likely govern stearin content, the primary determinant of shea butter quality, relatively high copy numbers of six key enzymes were found (KASI, KASIII, FATB, FAD2, FAD3, and FAX2), some likely originating in shea's more recent WGD event. To help translate these findings into practical tools for characterization, selection, and genome-wide association studies (GWAS), resequencing data from a shea diversity panel was used to develop a database of more than 3.5 million functionally annotated, physically anchored SNPs. Two smaller, more curated sets of suggested SNPs, one for GWAS (104,211 SNPs) and the other targeting FA biosynthesis genes (90 SNPs), are also presented. With these resources, the hope is to support national programs across the shea belt in the strategic, genome-enabled conservation and long-term improvement of the shea tree for SSA.