CRISPR-Cas
What you need is CRISPR-Cas. It's a potentially world changing DNA technique. The techniques to change DNA can mostly choose one of the following attributes: fast, cheap, precise. If they are precise, they are often taking a long time and are expensive. CRISPR-Cas, if it's truly working (which seems to be the case) is fast, cheap, very precise and can be done at any moment. Before conception, during pregnancy and any time after birth right up until they die of old age.
CRISPR-Cas works by using the enzymes bacteria make for defence. If a bacteria survives a virus attack, it'll keep part of the viruses DNA within an enzyme. If this enzyme detects that very specific code of DNA, it'll rip it out and replace it with garbage DNA. This will kill the virus or render it harmless.
We can now high-jack this process. We can change the specific DNA that need to be replaced, as well as what it'll be replaced with. This means that, for example, if your eye colour has one specific DNA strain, we can change it for another, or something completely different. Most attributes we get from DNA is more complex and not one specific line of 'code' in the DNA, but even the small things can have huge ramifications.
From DNA we make RNA. RNA is like a bit of code that is mostly used for the creation of enzymes. These enzymes do a lot of work in the body. Now imagine someone with a thyroid gland defect. This can impair growth, metabolism, temperature, cholesterol and much more. In a few CRISPR-CAS therapies you can potentially change the whole function of this gland.
Research is now mostly focused on birth defects. Preventing MS, dementia, immune disease and much more. The list is pretty big.
Potentially you can put grown people's bodies back in a state that the bodies think they're twelve years old, giving them a growth spurt again. Or their brains making rapid connections. Give them bio-luminous skin. Or give them the ability to lose less telomeres and slow down/prevent aging.
For your story, the research might be most interesting for children. Older people might be more stable, but DNA is a complex and tricky thing. Some research might be more suited for children, especially with their still growing immune systems. We can see this with strength against some diseases. They are an inconvenience for children, but potentially deadly for anyone who didn't get the disease when they were kids. It is then much easier to test on kids than to try to adapt grown men first with a technique we aren't even sure will be able to work that miracle at a certain time.
Do keep in mind that you can only work with the available chromosomes. You can't add more chromosomes. I'm not sure if you can lengthen or shorten the DNA strain you replace, but I thought I read somewhere it's possible.
In short is the following what you need:
- A bacteria you can infect with the desired change of DNA to produce the enzyme, or a direct way of producing such enzymes.
- A DNA sequence of the person you want to change the DNA from. Potentially you can have some more meta DNA for a larger group you can change, but if it's only a little bit different the DNA won't be replaced.
- Access to the person to inject the enzyme in enough quantities.
- DNA knowledge. This is something that would be near impossible to do randomly. It would be like trying to break an encrypted password, but instead of working with 0 and 1 you need to work with about 5 values.