Zoo populations of threatened species are a valuable resource for the restoration of wild populations. However, their small effective population size poses a risk to long-term viability, especially in species with high genetic load. Recent bioinformatic developments can identify harmful genetic variants in genome data. Here, we advance this approach, analysing the genetic load in the threatened pink pigeon (Nesoenas mayeri). We lift-over the mutation-impact scores that had been calculated for the chicken (Gallus gallus) to estimate the genetic load in six pink pigeons. Additionally, we perform in-silico crossings to predict the genetic load and realised load of potential offspring. We thus identify the optimal mate pairs that are theoretically expected to reproduce offspring with the least inbreeding depression. We use computer simulations to show how genomics-informed conservation can reduce the genetic load and maintain genome-wide diversity, arguing this will become instrumental in maintaining the long-term viability of zoo populations.