Genetics basics for people that don't know anything about it [part 2]๐งต
![Genetics basics for people that don't know anything about it [part 2]๐งต https://t.co/CfAvrIMaQc](https://pbs.twimg.com/media/FTD7Tq2WUAA2yV5.jpg)
[Part 1] can be found here
After [part 1] you should be able to read genetics papers and understand their general ideas and discussions. In [part 2] we will look over how you can do your own analyses and even analyze your own genetic make up
First to analyze your own DNA you must get a testing kit from a company that will give you your results. DNA companies analyze your DNA and compare you to a batch of modern population groups. This is okay but we can analyze it on our own further
Before you choose a company where you order a kit from you should check whether they give you results for Y dna and mtdna (women don't have Y dna so that is irrelevant for them). Another thing to look for is the number of autosomal SNPs they check for. The higher the better
Now it all comes down to price. You can order a whole genome sequence testing kit but this is more reserved for the fanatics and not needed. 23andme (ancestry only) is good enough.
For more information comparing tests you can get an overview here
eupedia.com
eupedia.com
To make new cells, an existing cell divides in two. But first it copies its DNA so the new cells will each have a complete set of genetic instructions. Cells sometimes make mistakes during the copying process - kind of like typos.
These typos lead to variations in the DNA sequence at particular locations, called single nucleotide polymorphisms, or SNPs (pronounced "snips"). DNA is passed from parent to child, so you inherit your SNPs versions from your parents.
You'll be a match with your siblings, grandparents, aunts and uncles at many of these SNPs. You'll have far fewer matches with people to whom you're only distantly related. The number of SNPs where you match another person can therefore be used to tell how closely related you are
Once you have your results back you can now use a third party that can transform your results in to coordinates so we can analyze it further and check what ancient dna you have inherited and what groups you come from.
You can do this by uploading your raw data file to Illustrative DNA (i'm not affiliated) for a reasonable price. If you're not interested in doing your own analysis don't worry they tell you what ancient populations you come from roughly
illustrativedna.com
illustrativedna.com
Once you have your coordinates you can go further to analyze your own DNA or that of other populations. I will explain it step by step. First you need to know where you can find coordinates of ancient and modern populations. You can find them here:
eurogenes.blogspot.com
eurogenes.blogspot.com
These 4 sheets are the most important ones. Clicking on them will lead to a google doc with a list of populations followed by coordinates. For the beginners you can just focus on the averages. For example here we can see genetic coordinates of England during the mesolithic period
Just look through the blog I linked earlier. Don't be intimidated. I will show how easy it is to do the analyses by yourself. First you need the calculators needed this you can find here:
vahaduo.github.io
vahaduo.github.io
You don't need to install anything. It is all online. The first one in the screenshot functions somewhat similarly to a qpadm graph (remember in part 1). The other two are PCAs
First we will look at the first calculator as this can be used to analyze a person or a group to see what populations they descent from. When you open it you will notice it being empty obviously.
First you need to fill the [Source] tab with populations you want to check you or a group of people derive from. You can find this out by reading genetics papers. And you can play around with some guesses yourself.
![First you need to fill the [Source] tab with populations you want to check you or a group of people...](https://pbs.twimg.com/media/FTD7Y4iWYAIA1k9.jpg)
Davidski made a calculator with various general populations that can be used for a basic calculation a while back, but that has been since deleted. Luckily it has been saved and is seen in the tweet above as an example. You can copy it here
pastebin.com
pastebin.com
Or here:
anthrogenica.com
anthrogenica.com
Next up is the [Target]. Meaning whose dna do we want to analyze. In this example I will use the modern average of the English and the Romanians that both can be found in the google doc on the blog mentioned earlier. After that we go to [Multi]. Press run
![Next up is the [Target]. Meaning whose dna do we want to analyze. In this example I will use the mod...](https://pbs.twimg.com/media/FTD7Zc3WAAAehcv.jpg)
![Next up is the [Target]. Meaning whose dna do we want to analyze. In this example I will use the mod...](https://pbs.twimg.com/media/FTD7ZleXEAA5jt_.jpg)
![Next up is the [Target]. Meaning whose dna do we want to analyze. In this example I will use the mod...](https://pbs.twimg.com/media/FTD7ZubXsAY-ANZ.jpg)
And now you have just done a genetic analyses. You can already do this with modern and ancient populations playing around with it. And when you have your coordinates back you can even do it for yourself.
Now some important mistakes to avoid. As you might have noticed you can see that there is a distance given. The lower the distance, so the smaller the number, usually the better. Try aiming for a distance lower than 0.03 by adding and subtracting populations in the [Source] tab
Another important thing to avoid is having too much in your [Source] tab. This can lead to overfitting. Meaning you will have a bunch of populations you supposedly derive from with absurd precentages at like 1 or even 0.1. A good idea is to remove things lower than around 5%
![Another important thing to avoid is having too much in your [Source] tab. This can lead to overfitti...](https://pbs.twimg.com/media/FTD7arUX0AM9qO_.jpg)
As I mentioned earlier the source populations given by Davidski for basic calculations are fine to use if you have no interest in reading genetics paper or finding out which populations are better to use, but it is not ideal.
@threadreaderapp unroll
For a PCA (see part 1) it functions almost the same. Fill in the [Source] tab. Fill in [Projected] what you want to analyze. Go to [PCA Plot]. Click "Run PCA" first then "Plot PCA"
![For a PCA (see part 1) it functions almost the same. Fill in the [Source] tab. Fill in [Projected] w...](https://pbs.twimg.com/media/FTD94PhWAAIzPYD.jpg)
![For a PCA (see part 1) it functions almost the same. Fill in the [Source] tab. Fill in [Projected] w...](https://pbs.twimg.com/media/FTD94b9WYAEdIyf.jpg)
![For a PCA (see part 1) it functions almost the same. Fill in the [Source] tab. Fill in [Projected] w...](https://pbs.twimg.com/media/FTD94koXsAImA_a.jpg)
![For a PCA (see part 1) it functions almost the same. Fill in the [Source] tab. Fill in [Projected] w...](https://pbs.twimg.com/media/FTD94t-XsAAeYx3.jpg)
For the other PCA populations are already added. You just have to pick the population you want by clicking [Choose PCA] then entering the coordinates and name as a whole in "paste single or..." after that you click [Add to PCA]
[END]
Here is a great explanation in French from a friend of mine.
Loading suggestions...