I tried to think of a few ways to present the new version of the report, but a PDF would be most useful. I’m just not sure if I know how to put PDF files on here, so I printed out the reports for both Princess Yue Pants and Tommy Gargoyle, and adjusted the photos. Go ahead and click on them, and they will blow up so you can read them. The cats they listed as being in common with my cats has updated, with about half of them being eliminated from this Alpha report. I’m guessing it’s because they grouped my two cats into either a “Western” breed or an “Eastern” breed, and while their percentage was favorable towards Eastern, they were both designated Western, so I guess the other ones drop. I was hoping for more insight into health issues, but they only provide the one top breeds and the issues that those breeds typically present. These reports will be updated at a later time, as they expand their research. My guess is that they have to build a decent database for purebred cats, and then go into the mutts we, which most all of you with cats have as well. Breeding cats started much more recently than the breeding of dogs, so this isn’t as easy as you might think. They also request cats with certain health issues and traits and study their genes as well, like cats with asthma. It’s a slow going process getting your feet off the ground as a company ready to present to consumers.
For reference I am providing a few photos of these cats of mine:
What?
This guy
Tommy favors the Russian Blue breed in attitude and looks, but Yue has more predominance to that breed than he does. Yue favors Ragdoll and Norwegian Forest Cats, but she’s only 10 lbs, and those two breeds were further down in her list.
The vomit Princess
I am looking forward to future updates. P.S. Tom does have FLUTD. He came down with it when he was 7.
You can purchase your CatKit here
In terms of your health results, additional experiments need to be performed on your samples. You can see more about that discussion in the comments on this Instagram Post (https://www.instagram.com/p/Bu1uXAsAI6J/?utm_source=ig_web_copy_link), but the idea is essentially that the ancestry results are based upon 0.5x (Illumina) Sequencing. However, cats are diploid and have two copies of each chromosome. In order to be able to detect a change in 1 copy within the range of a read of Illumina, I believe they are going to use a SNP chip and Sanger sequencing?
In terms of ancestry, I am interested in seeing what they can do with the 0.5x sequencing. When I have my own cat’s results, perhaps I can learn more about the pie chart.
However, in terms of the chromosome painting, I wonder if that can be improved with additional sequencing and/or whatever they use for the health results (I paid $1000 for 15x sequencing, in addition to the $95 for 0.5x sequencing: if they use the same scripts for both, it would be interesting to how that affects the results, and I will certainly share that). While cats have a shorter generation time than people, here is an example of what a chromosome painting result looks like for me (and I hope this helps illustrate what I mean about the frequency of small segments in the cat chromosome painting):
https://github.com/cwarden45/DTC_Scripts/blob/master/23andMe/Ancestry_plus_1000_Genomes/chromosome_painting_conf90_190321.PNG
To be fair, I think there is also a benefit to getting a second opinion from alternative software for human analysis (and that is what you would see if you browsed up from that photo within the GitHub repository).
Also, I’m not sure how often the interface changes, but I also had some comments on this Tweet about a Basepaws article: https://twitter.com/basepaws/status/1109094137179983874
(based upon the example that they provide, being in a higher percentile for a wild cat doesn’t actually mean your cat is more related to that wild cat than other wild cats)
To rephrase the wild cat issue another way, let’s say you calculated distance distances between your cat and the human reference genome. The cat sequences are not all the same, so the distances won’t be exactly the same (you can define a random organism, and get a random distribution of distances, where it doesn’t actually matter if you are in the 5% or 95% of those distances). However, all the wild cats will be more closely related to a domestic cat than a person: so, it is probably not meaningful to say your cat is in the Xth percentile to compared to that sequence meant to represent people (and that is essentially my concern about the wild cats that should have pre-defined overall similarity to domestic cats). I would guess you will encounter issues when talking about populations that differ clearly enough to be defined different species, but I don’t really know as much about comparisons between Felis catus and Felis silvestris, which I think would be more like talking about humans and chimps (although all humans are more closely related to all chimps than gorillas, and humans and chimps are in different genera).
Likewise, even though it is among the Twitter comments, here is a link the cat species phylogenetic tree: https://en.wikipedia.org/wiki/Felidae#Phylogeny. Admittedly, I think it is kind of interesting that pumas are closer to domestic cats than lynx on that tree, but I am not sure whether there are parts of that tree that are more controversial (with varying opinions from different experts). Tree of life defines clusters of species, but doesn’t say whether the domestic cat is more closely related to a puma or a lynx: http://tolweb.org/Felidae/16006 (and “Felidae” in the next highest classification for Family, above the Genus of either “Puma” or “Lynx”)
As I understand it, I think possible limitations to the percentile metric may also be relevant important for the domestic cat breeds: for example, it looks like Tommy has a lot more shared segments for the Ragdoll and Persian, even though it was supposed to have a lower percentile than Turkish Angora (and I would agree that it seems like he should show more similarity to a Russian blue, but I don’t think the UC-Davis VGS even has a test for that specific breed: https://www.vgl.ucdavis.edu/services/cat/).
However, I think the goal is to also eventually provide trait information: for example, Stormy is a Scottish Fold but he can also be tested for the “Colorpoint” trait (which is the marking on the ears and tail, to define a “lynx point” or “siamese point”): http://cdwscience.blogspot.com/2013/01/stormys-feline-dna-test.html
In other words, if the defining characteristic for a breed is defined by one gene (or a small number of genes), then a “trait” result may better match your expectations than an “ancestry” result (at least as I see presented in your report).
Also, while I’m in the mood to share links, there was also this YouTube Review: https://www.youtube.com/watch?v=iXF3Igxpc_Q&index=5&list=LLuYMmBMW1w2Op5ftLKgYIbg&t=0s
Thank you again for sharing your results!
I hope this long comment was interesting / useful…otherwise, I apologize for the information overload.
P.S. I was also not 100% about the formatting for comments in this format. So, I apologize about the hyperlinks that are probably a little messy to read.
Right after posting the comment, I saw a typo:
VGS should be VGL (it stands for “Veterinary Genetics Laboratory”)
Sorry about that!