Is Beating a DNA Test Possible?

  • Josiah Zayner



Humans contain about 25,0000 protein coding genes and much more non-protein coding DNA, all of which uniquely identifies us. Because of this, DNA tests have become the standard is criminal forensics for identification of individuals at the scene of a crime. When done properly these tests can identify individuals with a theoretical probability of 1 in 10,000,000,000,000,000,000(1018). These statistics come from using the Combined DNA Index System or CODIS and focuses on identity through only 13 genes (alleles). Well, it is not actually 13 genes, it is only small parts of 13 genes. To me this seems like a very breakable and hackable system so let’s talk think about that for a minute or seven.

The problem with DNA privacy is that we leave DNA everywhere and even a faithful skin scrubber, hairnet wearer, ala GATTACA, will still eventually leave an eyelash behind. Maybe instead of preventing DNA from being found, people could modify their DNA temporarily in a few hours time to fool the match. Is it possible?

I made a post on craigslist as a sort of speculative exploration of what the world may be like if people wanted to temporarily engineer their DNA to prevent genetic identification. An ad was placed looking for someone to help me engineer my genome using a new genetic based system called CRISPR-Cas. The post went “viral” as I made the author of the ad seem Tony Stark-esque (totally what I was going for) because he is a friendly trope.


The response was interesting, I received a number of emails of people offering their services. Real people (I could find them on Linkedin) with actual experience editing DNA using CRISPR-Cas. They seemed serious.

A few months ago, and about a week after my craigslist post, a group in China used the CRISPR-Cas system to engineer human embryos. Though they saw a number of off-target effects in their engineering attempts, they were successful and the research has brought these ideas to the forefront of Science. The simplicity of CRISPR-Cas makes it so that once it is optimized it will be trivial to use.

I am a strong proponent of using DNA evidence to help identify individuals who commit a crime but how often is it manipulated? As early as the 1990s we have evidence of people fooling DNA testing by manipulating the sample given for matching. Generally these people are criminals but that is bound to change if it hasn’t already. The information in your DNA could be priceless or just interesting for a genetic voyeur. Are there ways to protect our DNA privacy? Ways to prevent our DNA from being properly traced and identified?

Artist and DNA privacy researcher Heather Dewey Hagborg has suggested that people use a form of DNA “spoofing” by using synthesized DNA to obfuscate their own DNA. Though I imagine most forensic analyses do not differentiate synthesized DNA from actual host human DNA a company called Nucleix, based in Israel, have touted how they can separate out the two. They propose the use of DNA methylation as a marker because synthesized DNA is not methylated. It is unknown the exact methods Nucleix use (Whole Genome amplification?) but genes cloned and expressed in bacteria or even human cell culture could be purified with methylation patterns similar to what would occur normally. Using an HGH like strategy one could simply collect samples of others DNA, extract that DNA and use it at crime scenes and it would contain methylation patterns and all. The more the better, one of the biggest deterrents to identification through DNA is contamination of multiple DNA samples. Purifying DNA from chewing gum and cigarette butts can all be done at HOME and relatively inexpensively.

What about next level protection? It should be possible to take the 13 genes that are used in CODIS and clone them into bacteria that colonize our skin, providing constant protection and obfuscation or our genome. Thanks to things like the 1000 genomes project you can even incorporate the DNA of real individuals.

A new genome everyday?

Because CODIS is a database of already collected samples it is difficult to update or change collection parameters and procedures. Next year the FBI could change CODIS to include 23 genes for all new samples but it wouldn’t change the fact that the old samples still only have 13 genes sequenced. This would be more useful for matching individuals at the scene but not increase the helpfulness in identifying individuals through their DNA till the database was sufficiently populated again.

How long will it be before old DNA evidence is no longer admissible in court?

How long before forensic analysis starts to take into account the ability of individuals to temporarily modify or obfuscate their own DNA?

Criminals or people looking for privacy evolve with the times and identification methods. People started to wear gloves to prevent identification through fingerprints. We can hide our tracks online by blocking cookies, accessing the web through proxies and using encryption. It seems that soon we will see attempts by investigators to find new ways to track and identify people using Bio-Forensics and by others to fool those methods.



This article and its reviews are distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited.