Introgression in wolves & dogs

Apropos of our discussion of evolution and dogs, and introgression, here is a new paper I stumbled upon in Molecular Ecology, Detecting introgressive hybridization between free-ranging domestic dogs and wild wolves (Canis lupus) by admixture linkage disequilibrium analysis. Linkage disequilibrium is basically the non-random association of alleles across loci. For example, imagine that you have alleles A1 and A2 at locus A, and B1 and B2 at locus B. Imagine that these two locii are on separate chromosomes (just to make it clearer, though they don't have to be). In a randomly mating population A1 and B1 or B2 should not be correlated together in the same organism's genome anymore than you would expected based on their frequencies. But sometimes gene-gene interactions result in the coadapted fitness of two alleles, so that A1 + B1 is more fit than A1 + B2 (imagine, if you will, higher spontaneous abortion of fetuses with A1 + B2). Nevertheless, today linkage disequilibrium is more often used to study the impact of selective and demographic forces operant upon the genome across the span of history than coadapted complexes. A strong powerful selective event on one locus, say Z, might drag adjacent regions of the genome along via a "hitch-hiking" event during the selection sweep (this is why researchers tend to look around the region coding for lactase to confirm that their technique works). As time passes recombination should break apart linkage disequilibrium, so the extent of the current associations can be highly informative. And so it is with this paper.

Here is the relevant portion of the abstract:

Occasional crossbreeding between free-ranging domestic dogs and wild wolves (Canis lupus) has been detected in some European countries by mitochondrial DNA sequencing and genotyping unlinked microsatellite loci. Maternal and unlinked genomic markers, however, might underestimate the extent of introgressive hybridization, and their impacts on the preservation of wild wolf gene pools. In this study, we genotyped 220 presumed Italian wolves, 85 dogs and 7 known hybrids at 16 microsatellites belonging to four different linkage groups...Results indicate that (i) linkage disequilibrium was higher in wolves than in dogs; (ii) 11 out of 220 wolves (5.0%) were likely admixed, a proportion that is significantly higher than one admixed genotype in 107 wolves found previously in a study using unlinked markers; (iii) posterior maximum-likelihood estimates of the recombination parameter r revealed that introgression in Italian wolves is not recent, but could have continued for the last 70 (± 20) generations, corresponding to approximately 140-210 years...despite some admixture, wolf and dog gene pools remain sharply distinct (the average proportions of membership to wolf and dog clusters were Qw = 0.95 and Qd = 0.98, respectively), suggesting that hybridization was not frequent, and that introgression in nature is counteracted by behavioural or selective constraints.

The important points here are that more than on locus needs to be looked at, and, hybridization does not imply that the two populations lose their distinctive characters.