Sometimes timing is funny. In November 2017 I wrote a post defending the Przewalski's horse's status as a wild animal. A new study by Orlando et al. has challenged the status of the Przewalski's horse as the last living genuine wild horse.
The earliest archaeological evidence of horse husbandry is from the Botai culture of Kazakstan from 5.500 years ago. It has been assumed previously that these Botai horses belong to the earliest strain of domestic horses of the caballine lineage. Surprisingly, the authors found only about 2,7% Botai-related ancestry for all domestic horses from 4.000 years ago, while the authors claim the Botai horses turned out to be the ancestral stock of the modern Przewalski's horses population. I have not read the paper yet because it is behind a paywall, thus I cannot see what led the authors to the conclusion that all modern Przewalski's horses descend from the Botai population and not just that the Botai population was part of the przewalskii clade. But let us assume the former is the case for now.
The authors thus write that Przewalski's horses are feral descendants of these early domestic horses, and not true wild horses. I have problems with the latter part of the conclusion. First of all, the horses of the Botai culture must have been in a very early state of domestication, and not for a long time. After escaping human husbandry, those early domestic horses have been exposed to natural selection for four millennia again. Any changes that might have occurred in the early state of domestication must have adapted to the requirements of living as a wild animal again. Thus, I don't regard the short (in evolutionary terms) episode of domestication as substantial enough to categorize the Przewalski's horse as feral instead of a wild animal. In fact, ever since its discovery, the Przewalski's horse has been used as a model for a wild equine and numerous differences in physiology, development and behaviour have been noted between the Przewalski's horse and domestic caballine horses as much as feral caballine horses. Przewalski's horses are way less tamable and more aggressive than domestic caballine horses, and genomic studies have shown that numerous genes for physiological aspects have been altered through human utilization while that is not the case in the Przewalki's horse. The recent study by Orlando et al., or to be precise the short period of early domestication, does not alter this fact. The situation is not comparable to mustangs and other feral caballine populations at al.
Thus I do not regard the Przewalski's horse as any less wild as before. One could use it as an example for a "post-domestic wildtype", a term I tried to introduce in my posts on dedomestication as opposed to "pre-domestic", a term that is already used. However, I am rather confident that the majority of authors will begin to list the Przewalski's horse as a feral horse now, as the 19th century conception of nature and evolution as something static and humans as an irreversible altering factor is still prevalent in a lot of experts heads. Unfortunately, in my opinion.
Interestingly, the authors found the alleles for the Leopard spotted colour present in the Botai horses. This colour variant has already been found in Pleistocene wild horses (see earlier studies of Orlando et al.).
Orlando et al. 2018: Ancient genomes revisit the ancestry of domestic and Przewalski's horses.
 Schubert et al.: Prehistoric genomes reveal the genetic foundation and costs of horse domestication. 2014.