Variance Effective Population Size under Mixed Self and Random Mating with Applications to Genetic Conservation of Species

Abstract

Listen

Translate article

When collecting and regenerating genetic resources, genetic drift affects the representation of a population and occurs at two stages: when sampling the parents and when gametes are sampled from these parents. The variance effective population size [Ne(v)] quantifies genetic drift. In this study, a model for calculating Ne(v), that considers the two‐stage sampling of mixed self and random mating species, is developed. For germplasm collection, as the rate of natural or artificial self‐fertilization (s) increases, Ne(v) is reduced and becomes increasingly dependent on the number of seed parents (P) and is less influenced by the number of seeds sampled per parent (n/P). Female gametic control (GC) leads to higher Ne(v) than with random sampling of seeds (RS), but its effect is tangible only when n/P is small. For accession regeneration, maintaining accession integrity (the proportion of functional parents, u) at an adequately high level and adopting GC are required for assuring Ne(v) equal to or greater than the actual size of the accession (Ne(v) ≥ n). The importance of these two factors is enhanced as s increases. For arbitrary rates of selfing (0 ≤ s ≤ 1), under inbreeding equilibrium (IE) and with constant population size (n = N), Ne(v) can be adequately maintained through GC with a loss of ≤20% within accessions. For large sample size (n → ∞), an accession loss of ≤33% can be recovered. For maintaining adequate Ne(v), artificial selfing followed by GC is more efficient than accession regeneration by natural reproduction. For achieving appropriate Ne(v)s, increasing the rate of self‐fertilization in polymorphic materials makes collection more difficult but regeneration easier for minimal loss (≤20%) within accessions.