This is amazing. Sequencing an entire human genome is now going to be cheap and fast. We predict this will be a game-changing technology for the use of DNA technology.
Right now, DNA evidence is looked at much like fingerprint evidence. With fingerprints, law enforcement doesn’t compare every single ridge and whorl to see if there’s an exact match. Instead, particular locations are compared, to see if those locations are the same. And there’s a lot of subjective interpretation that is needed to make that call. DNA evidence is no different. The entire DNA sequence is not compared. Instead, a handful of locations are compared, to see if the DNA at those locations is the same. And there’s a lot of subjective interpretation that is then needed to make that judgment call. This can call DNA evidence into question, particularly when there are mixtures, degraded samples or equipment glitches that create room for errors of judgment.
If one were to compare the entire genome, however — all 3.3 billion base pairs of it — there would be much less room for interpretation and error.
The problem is that sequencing an entire genome has, to date, been prohibitively expensive. The first genome was a massive undertaking. As of 2009, only 7 people’s genomes had ever been sequenced. The time and expense needed to compare the genomes from a bit of evidence and from a single suspect would take forever and cost a huge amount of money. It’s just not practical.
But if the new technology announced in this WSJ video is for real, all that could change very soon.
If the technology is available to analyze and compare all 3.3 billion base pairs rapidly and cheaply, the current system of comparing 9 or 10 or 13 loci will be woefully inadequate. The current system relies on interpretations that can be incorrect, and statistics that can very easily be misinterpreted. A full-genome comparison knocks out most of the interpretation and guesswork, and eliminates the need for statistics.
Statistics won’t be necessary any more, because (with the exception of identical twins) no two people share the same genome. So leaving aside identical twins, the odds of someone else sharing your genome are nil.
Statistics are only used now because we only look at a few loci, not the whole genome, and all kinds of people share the same alleles at each of the various loci. Presuming the independence of the stats at each locus, the odds of a match at each locus get multiplied to reach truly astronomical odds against there being a random match. (Remember, though, that the stats don’t say anything about the odds against this being the match. And also remember that a test with even a 99.9% success rate will have false positives 0.1% of the time — and given the low odds of a correct positive match in the first place, the chances of a false positive actually occurring are amazingly high, in the realm of a coin toss.)
This can be a boon to innocents who otherwise might be wrongly charged based on bad DNA interpretations. It can also give society at large greater confidence that people convicted based on a full-genome match are rightly convicted.
Sure, we defense attorneys would lose some arguments for reasonable doubt in the wiggle room of current DNA analysis. But we’d gain arguments against those who would continue to use the present-day analysis when there is a more exact method available.
It hasn’t happened yet. But it’s coming soon. We can’t wait to see how this shakes out.