8
$\begingroup$

I am writing a science fiction story with mass produced clones. Since I have a number of clone I need to tell them apart. The ID system must be:

  • Permanent over the lifetime of the clone (call it 150 years)
  • Very difficult to tamper with (radio frequency transmitters can be excised, etched bones can be re-etched -- hard to read anyway)
  • Not overly harmful to the clone (clones are only valuable if they function like humans)

Can I use DNA to do this, by simply adding several codons to an existing DNA strand (assuming quick, basically immediate, and accurate Polymerase Chain Reaction - PCR)? Could you modify the clone by rewriting his/her DNA thus causing a serial number to appear on the skin and still detect the genes responsible with PCR (this would should meet the criteria above - two level verification between skin and genetics seems a secure methodology)? Could either of these techniques be used to modify the clones genetic strucure prior to inception thus changing the DNA of all cells within the cloned organism?

Note: The following wikipedia article might provide a start in responding to this post (https://en.m.wikipedia.org/wiki/Gene_therapy)?

$\endgroup$
  • 4
    $\begingroup$ If your focus is a permanent ID system, then why have it display on the skin? No matter how it got there, if you scar it enough, the text will disappear. DNA does not heal/remove excess scar tissue. Bear in mind that, to date, Man has not invented a marking process that Man cannot obscure. Your DNA solution is inventive and seems to solve the problem itself. So, why the need to display it on skin? $\endgroup$ – JBH Mar 4 '18 at 17:45
  • $\begingroup$ Possible duplicate of How to uniquely identify organism in planet , in which all are biologically same? $\endgroup$ – Amruth A Mar 5 '18 at 12:26
  • 1
    $\begingroup$ @AmruthA Not a duplicate: the other question assumes voluntary buy-in by the organisms; this question is explicitly concerned with preventing fraud. "Name tags" and "tattoos" (answers to the other question) just aren't going to cut it. $\endgroup$ – Azuaron Mar 5 '18 at 15:33
  • $\begingroup$ @Amrutha Not a duplicate: The cited question (worldbuilding.stackexchange.com/questions/100140/…) deals with technology available around the year 1900. This post cites mechanisms not available at this time in history (Polyemerase Chain Reaction, genetic engineering, radio frequency transmitters which would be small enough to implant in a human body). $\endgroup$ – Englishman Bob Mar 5 '18 at 16:11
12
$\begingroup$

Can I use DNA to do this, by simply adding several codons to an existing DNA strand (assuming quick, basically immediate, and accurate Proteinas Chain Reaction - PCR)?

Absolutely. A significant percentage (as high as 20% according to some authors) of DNA has no biological activity and is a leftover from ages past. Another 60% has little direct activity (it is non-coding) but might be useful, so let's leave it alone, but that still leaves a lot to play with.

Locate a unused section of a pseudogene which is very likely to be completely nonfunctional (e.g. the seventh in a run of fifteen useless incomplete repetitions likely due to copying errors in the last twenty million years) and replace it with an equally non-coding sequence. There are some combinations you cannot use to ensure that the sequence won't actully code anything, but you have plenty of space for your needs.

Or if you feel more adventurous, you can use synonymization: several aminoacids are coded by more than one DNA triplet, and in normal human genetic code, you'll find say GGA for proline. You can then use GGT and GGG in that position to keep everything working (both triplets will code for proline), but at the same time encode either a 0 or a 1. By comparing the known human reference with the clone sequence, you can extract a binary string:

human: TAA GCT GCT CAG CGT 
clone: TAG GCA GCG CAC CGA ...
code :  0   1   0   0   1

(This form of 'meta-coding' might have biological significance. In Frameshift by Robert J. Sawyer, it "unlocks" genetic sequences of DNA that trigger evolution - it's a sort of super-code hidden inside DNA).

However, care must be taken to avoid sequences where the synonymization actually allows coding two different and slightly offset DNA messages in the same sequence.

$\endgroup$
  • $\begingroup$ The issue with replacing active sections is that there's a lot more going on there than just 3 bases = 1 amino acid. For instance when replacing parts of "TAA GCT GCT CAG" you also have to make sure the amino acids for the offset versions "T AAG CTG CTC AG" and "TA AGC TGC TCA G" are also the same. Plus you also need to make sure the changes don't affect folding or any transcription factors; and make sure what you're editing only codes a protein rather than a ribozyme or performing some other function. $\endgroup$ – AJMansfield Mar 5 '18 at 16:26
  • $\begingroup$ @AJMansfield very good point. Amending answer. $\endgroup$ – LSerni Mar 5 '18 at 16:35
  • $\begingroup$ @AJMansfield why would you have to ensure that the offset versions code the same? $\endgroup$ – frodoskywalker Mar 5 '18 at 22:38
  • $\begingroup$ @frodoskywalker because in some instances, both versions actually code something. In those instances you're forced to leave the code alone, there's not enough leeway to encode anything. But it matters little; there's a lot of space that's both available and risk-free. By the time we're at cloning humans, I expect that all risk-free zones will have been thoroughly mapped and confirmed. $\endgroup$ – LSerni Mar 5 '18 at 22:51
  • $\begingroup$ @LSerni thanks! I'm familiar with this offset encoding but had only ever heard of it being used by viruses. Was very surprised to read that it happens (a lot, apparently) in humans. $\endgroup$ – frodoskywalker Mar 6 '18 at 7:24
3
$\begingroup$

Disclaimer: I'm not a biologist.

Think of DNA as source code for programming. It is read out and executed, thus very important. So if you just add some ID on it, if in your story DNA is not completely understand, it could be very easy to destroy something, just like when you randomly insert a line of code in a software project with million lines of code. Nothing could be changed, or everything could go to hell.

(EDIT: Junk DNA was mentioned in the comments. It would be a reasonable place for IDs, given you can place it there. As I understand from the Wikipedia article, biologists are still guessing why it exists in the first place, it may as well be important for growth before birth. Money quote: "Several lines of evidence indicate that some "junk DNA" sequences are likely to have unidentified functional activity")

Also DNA changes over time. Due to different factors like radiation, it is altered and doesn't reproduce like earlier, but with errors. Most obvious example of this is aging, but tumors are also one. So it could as well be that by "reading out" the ID, a wrong ID is read out cause it was altered.

Alternative method

Here is an idea how you could do it instead: Write the ID inside their skulls. Take out a part of the skull bone, a random unique part, carve the number on the inside in, and sew everything together again.

  1. Clearly permanent.
  2. Difficult to temper with: A single bone can normally replaced by a similar one without the body feeling noticing. But a brain operation is delicate. And since the part of the skull bone is selected random, to temper with it you would have to replicate that exact part (or people could see there are several traces on the skull bone, for which they would only have to open the head, but not the skull). For that, you first need to take it out of the skull to take measurements. For the time being the clone would lie around with exposed brain. A lot of difficulties and problems. The drawback is that it is difficult to carve the number in the first place, but still a lot easier than changing it.
  3. As long as the operation is done by expert in a sterile environment, there would be no consequences but a scar on the head, usually covered by hair.

Ok, right before posting I notice it would also be very difficult to check the ID (as long as the clone needs to stay alive), but you can circumvent this by encarving the ID on both sides of the skull bone part. Easy check and hard check possible. Another variant would be to only write it on the outside, much less dangerous because no brain exposed (because operations always have a risk), but then again easier to temper with, the random unique traces of the removed skull bone part would be missing.

EDIT: Another way of checking the encarvings would be a MRT. Giving technology for clones is available, it is reasonable that scanning the skull bone is mobile possible too. Today MRTs are done in clinics and can take e.g. half an hour, even that would be reasonable. And if the encarving is filled with some metal e.g. iron and a protective silicon layer so the metal doesn't interact with the rest of the head, this would make checking even easier.

$\endgroup$
  • 3
    $\begingroup$ Considering that at least some of the non-coding DNA is likely just Junk DNA, which has no biological function - you can always use some such region to retain the ID number. $\endgroup$ – G0BLiN Mar 4 '18 at 19:11
  • $\begingroup$ @G0BLiN Interesting, didn't know about that! Still, one would have to know where these parts are and the other problems I described still hold. $\endgroup$ – SK19 Mar 4 '18 at 19:42
  • $\begingroup$ @G0BLiN Please post this. Could you explore issues such as the use of PCR? Issues such as radition and copy errors? Addressing my latter point, would multiple copies of the ID in diffferent areas of junk DNA reduce error issues? $\endgroup$ – Englishman Bob Mar 4 '18 at 19:58
  • 3
    $\begingroup$ SK19 - While I'm not downvoting this as biology isn't my field, I suspect you greatly overestimate the results of adding a few scores of non-coding nucleotides to a Junk DNA region - if the ID sequence is repeated "enough" (say, a hundred times), even random mutations will still leave it readable - and these mutations are extremely unlikely to repeat across multiple cells - so just take saliva+blood+epidermis samples to get multiple cells from different sources, correlate (maybe using the fictional fast and accurate PCR OP mentioned :) ) - and you've got a very reliable DNA-based ID. $\endgroup$ – G0BLiN Mar 4 '18 at 20:20
  • $\begingroup$ Another question - in the partial-skull-replacement method - how will the ID be scanned? how will tampering be detected without opening the head of a suspected clone? $\endgroup$ – G0BLiN Mar 4 '18 at 20:22
3
$\begingroup$

The current state of the art version of this is based on CRISPR, which is a set of antiviral genes found in bacteria. They are used today as a powerful tool for gene splicing, where you can identify a location you wish to cut, snip the DNA there, and add content. It's pretty fascinating.

Of course, the big issue with DNA biometrics like this is that if you have the tools to add a serial number, you likely have the tools needed to edit one out. You might develop a GATTACA situation.

Then again, what story involving clones doesn't involve the numbered clones going "haywire" and rebelling against being numbered?

$\endgroup$
1
$\begingroup$

Why not simply use an iris or retinal scan to match it to an ID? The patterns in the iris and capillary system in the retina are unique: not even genetically-identical twins have the same patterns, and an individual doesn't even have the same pattern in both eyes, which by definition comes from the same DNA and developed in the same conditions. Fingerprints would work as well, but eye scans can be read at a distance, are harder to fake (since you can beat a fingerprint scanner by duplicating someone else's print) and are subject to fewer issues such as scarring that can affect fingerprints. Hell, use both iris and retinal scans to reduce the possibility of error and potential duplication.

The downside is that you need to maintain a database linking the scans to clone IDs, but that should be fairly trivial thing to deal with if you're at the point of mass-producing clones, and such would be necessary for the option of coding in the DNA anyway. But as an upside, you don't need to do anything to generate the unique identifier: nature does that for the clone already.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.