The sequencing data sets comprise two sets of different lengths (6331 bp/4811 bp) surrounding the region of the IGF1 gene. A central result highlighted in the article abstract based on the sequence data is that Middle Eastern grey wolf haplotypes show higher nucleotide diversity, and that these haplotypes therefore originated in this region (Table 2, ). There are several major problems with this interpretation of the data sets.
Firstly, a 'large sample of grey wolves'  for the DNA sequence data is in fact a scarce sample of 28 grey wolves (data set: 4811 bp; Table 2, ), of which 8 (28.6%) are from Israel. Only two samples from China, a region suggested as the origin of dog domestication  were sequenced. Obviously, this is not enough to evaluate if Chinese wolves have a higher or lower genetic diversity than Middle Eastern wolves and they could by pure chance be unrepresentative of their respective population (for example, closely related, immigrant from another population or an inbred/outcrossed specimen).
Secondly, regarding the point 'grey wolf samples were chosen to be globally distributed and representative of all major populations' , the authors omit to discuss that grey wolves were once distributed all over the Northern hemisphere  but are today distributed in patches throughout their former distribution range as the result of extermination, restricting the comparisons that can be made. It is therefore questionable whether a comparison to the current wolf population can give any tangible insight as to the diversity of wolf haplotypes at the time of domestication or, as the authors write, 'early in the history of domestic dogs'.
Thirdly, the largest remaining wolf populations of the Old World in North-East Asia (Kazakhstan, Russia, Siberia and China), which cover approximately 50% of Eurasia, are represented by only two samples in the sequence data set (Figure 1; four in the SNP data, ). However, this area should be well sampled in a phylogeographical study aiming at identifying the origin of the SDH.
Fourthly, Chinese wolves, like Middle Eastern wolves, are of small stature. In addition, they have a morphological feature ('"turned-back" apex of the coronoid process of the ascending ramus', ) typical of dogs. Chinese wolves must consequently be considered as a potential source for the SDH and should therefore be sampled in numbers equal to other wolf populations.
Finally, in a previous study (Additional file 2), Spanish wolves had the largest number (5) of haplotypes with greatest similarity to the SDH whereas Israeli wolves had only a single haplotype closely related to the SDH. However, only two Spanish wolves but eight Israeli wolves were chosen for sequencing. The reason for excluding the Spanish samples is not given in Gray et al. . We therefore question how the samples were chosen for the analysis.