Declining species lose fitness through the accumulation of genetic load and its increased exposure due to inbreeding. This basic principle of the conservation genetics paradigm is well accepted, yet the underlying mechanisms haven’t been fully disentangled, mostly because studies of adaptive variation in declining populations have been limited. In this thesis, we will take advantage of the wide array of genomic resources generated by the recently completed Iberian lynx (Lynx pardinus) genome project, in order to directly assess functional variation in the populations of this highly endangered species. We’ll be focusing on the evaluation of selection against deleterious traits (purifying selection), whose efficiency, theory predicts, is weakened in smaller populations. We will use genome-wide patterns of non-synonymous –potentially deleterious– polymorphisms and substitutions to assess the interaction of purifying selection with locally heterogeneous processes like recombination and mutation. The genomic scan for major effect mutations could reveal the genetic basis of inbreeding depression, including those deleterious traits already identified in the captive population (e.g. juvenile epilepsy, cryptorchidism). Finally, we will try to improve current genetic management by fine-tuning the genomic tools at our disposal, including the selection of an extensively curated set of SNP markers and, eventually, by including selection against deleterious alleles in the conservation programme.