Natural hybridization is being increasingly recognized as a relevant phenomenon for fundamental evolutionary processes such as the introduction of adaptive variation or the origin of new species. Differently, anthropogenic hybridization (AH), following introductions of plants and animals and/or large-scale habitat modification and fragmentation by humans, is being increasingly considered a major threat to local wild populations. Especially if introgressive, AH may lead to genetic flow across genetically distinct populations causing substantial genetic swamping if not extinction. Controversy has traditionally surrounded the definition of appropriate conservation policies to deal with AH, as management measures are complex, inherently limited by insufficient power to detect admixed individuals and, most importantly, are and socially debated. Using the wolf x dog hybridization in Italy as a study case, and based on up-to-date molecular genetic tools coupled with innovative demographic modeling approaches, we aim to assess the extent of hybridization in wild wolf populations through the formal estimation of the proportion of admixed individuals (i.e. prevalence). To this aim, we will collaborate with a network of research and management institutions, both at the national and international scale, and on-going wolf monitoring programs in the northern Apennines, to (i) collect and analyse wolf samples, (ii) assign each sampled individual to the wolf parental population or to the admixed category, (iii) estimate prevalence of admixture within local wolf populations, and (iv) inform management based on our findings. In doing so, we will develop and disseminate robust sampling and estimation procedures that can be used with other taxa also affected by AH. As interventions to control the negative effects of AH can be controversial, we will also run a survey within the scientific community to assess commonalities and divergence regarding management options.
Many of the aspects we intend to address in this project appear to be innovative and relevant, both on theoretical (i.e. modeling) and applied grounds. First, whereas sophisticated genetic tools have been recently developed to detect admixed individuals in hybridizing populations, arbitrary thresholds are still used for their probabilistic assignment to the parental or the admixed category [1,2]. This classification is usually achieved through assignment tests based on Bayesian clustering techniques [3], according to which each individual is probabilistically assigned (q-values, indicating proportion of membership to a given cluster) to each of the reference populations according to its allelic frequencies. However, due to the arbitrary definition of threshold q-values the final proportion of admixed individuals is strongly subjective. Alternatively, threshold values can be obtained through making reference to the blind assignment of synthetic genotypes of known admixture genealogy [4]. Thus, by using a large reference population of wolves and dogs throughout Italy, showing no phenotypic and genetic signature of admixture, we will generate simulated, synthetic genotypes to objectively define critical threshold values to categorize sample individuals into parental and admixed categories. This will be instrumental in enhancing the reliability of any admixture assessment exercise, representing an important methodological innovation for the study of hybridization in animal populations.
Second, no formal modeling procedure has been developed to estimate key parameters in admixed animal populations, whose individuals feature uncomplete detectability and uncertainty of assignment (e.g. pure vs admixed individuals). In this perspective, capture-recapture (CR) models can be integrated to account for uncertainty in state assignment. Specifically, multi-event CR models, allow to include uncertain data in the sample thus increasing the sample size and avoiding the bias expectedly associated with the exclusion of uncertain individuals [5]. By applying this modeling approach to our case study we will make a significant step forward not only for a reliable assessment of this relevant conservation problem, but also for the development and availability of more complex and realistic models, potentially applicable to different animal taxa affected by anthropogenic hybridization.
Third, due to the restrictive conditions above, no reliable estimate of wolf x dog hybridization is currently available both in Italy and elsewhere, thus limiting evaluation and adoption of management interventions, and the same is true also for many other animal taxa affected by anthropogenic hybridization [6]. By (i) collecting representative samples of wolves from wild populations, and (ii) formally estimating the proportion of admixed individuals, we will be able to provide reliable estimates of wolf x dog admixture in local wolf population useful to design, plan, and implement management interventions. In addition, based on simulation work we will also provide indications on the optimal sampling regimes to be adopted by field technicians in order to achieve ideal samples to estimate population size and admixture in wild populations with due accuracy. This is severely needed, both in Italy and elsewhere in Europe, as sampling of wildlife populations is usually conducted with no reference to statistical efficiency.
Forth, by working cooperatively with different research and management institutions (see below), and experts of different expertise and professional background (i.e., genetists, ecologists, conservation biologists, researchers and practitioners), we will constitute for the first time a technical working group that will face and assess the problem of anthropogenic hybridization; although at this time the working group has not official institutional endorsement from regional or central government agencies, it will likely represent an important reference for such institutions in the years to come.
Lastly, by means of publication and dissemination of the results of our Delphi survey on the opinions and attitudes of ecologists, genetists and conservation biologists regarding how to effectively manage anthropogenic hybridization, we will stimulate a wider debate about this issue not only within the scientific community but also in a social and management arena. This is especially important for the development of solid policies that aim to reduce the occurrence of anthropogenic hybridization, as this issue is particularly controversial in a social dimension, being largely associated to value-laden opinions from different social groups.
References
[1] Verardi A et al. 2006. Mol Ecol 15:2845¿2855
[2] Benson JF et al. 2012. Mol Ecol 21:5934¿5954
[3] Falush D et al. 2003. Genetics 164:1567¿1587
[4] vonHoldt BM et al. 2013. Mamm Genome 24:80
[5] Pradel et al. 2007. Can J Stat 35:1¿14
[6] Allendorf F et al. 2001. TREE 16:613¿622