Suppose in the future a CSI tech is trying to ID a perp…and only has, say, a hair or a drop of blood…and has no matches in the system.
Is it at all possible for some DNA analyzer to read the gene sequences and draw a picture of what the un-sub looks like?
In the future…
Theoretically…
Certainly.
Accuracy, in black and white, would be slightly better than a police sketch artist. Still hairstyle, shaved condition and glasses go a long way in appearance.
As pan1 has said, theoretically yes. However there are other factors which may alter the man/woman’s appearance, such as bodybuilding, cosmetic surgery and others like facial tattoos (who knows?!), hairstyle, glasses and tanning (to a certain extent).
I don’t know why you’d say black and white: Hair and skin color would probably be the easiest things to pick out. While admittedly those can be changed by dyes or a tan, they’d still be at least somewhat useful.
With hair specifically, there are too many environmental factors and environmental related aging factors for DNA alone to be indicitive.
Skin tone is especially variant with sun exposure. And not just how much time you spent in a tanning booth this last year - but how much you spent playing in the yard as a child would factor in.
A black and white image allows the observer to not force colors into the picture, but to focus on the shapes. Add a witness and color can be added with enough certainty to make it useful.
The level of accuracy is not at all clear. I suspect even with perfect knowledge of genetics, accuracy would be fairly low, though some particularly distinguishing features might be identified (“By God, he’s got the Hapsburg lip!”).
I mean, even if a semi-reliable way to estimate age from DNA was found, I can’t imagine it being much more accurate than “Before puberty/around puberty/18-35/25-50/40-80/50-100”
This article discusses recent advances that allow DNA analysis to be used to identify ethnicity, eye color, hair color, and other characteristics based on SNP (single nucleotide polymorphism) analysis.
I don’t see being able to take DNA from an unknown perpetrator and coming up with a sketch of the person happening any time soon. As has been noted, so many of the characteristics often used for identification can be easily changed. Also, many of the things that make faces unique are not likely to be genetic–moles, minor asymmetry, etc. are more likely to be due to non-genetic factors.
keeping in mind this is “In the Future” and “Theoretical”
There would certainly have to be age progression images. Although its concievable to determine age based on other factors available within a few cells.
Theoretically, in the future…
I’m not saying this is gonna happen any time soon. The future is pretty big…
Environmental conditions can dramatically alter appearance without changing DNA. Fetal Alcohol Syndrome is unmistakable in someone with a severe case. Height is affected by diet and nutrition. The condition of teeth is influenced by diet, hygiene and procedures like braces; even when your mouth is closed, your teeth can affect the shape of your lips. Weight gains can’t change some facial features, but a DNA-sketch that doesn’t know whether the perp was 150 or 250 lbs isn’t going to be real useful.
I doubt that DNA will ever be used alone to generate a sketch. Rather, I think we’ll come up with lists of traits that we can be reasonably confident about.
Well, in principle it should be about as good as the resemblance between “identical” twins separated at birth, except for the age issue. But, again in principle, you could have a picture for every few years.
I, one of my maternal uncles, and my maternal grandfather all have the same mole, on the same side of the nose. So somewhere in our DNA, there apparently is a gene for “small smooth mole on the left side of the nose”.
The scientist in me just started screaming.
What, at my arriving at that conclusion from only three data points? I would think that, given that I’ve never seen that particular mole on anyone else, and given that the only other two people I have seen it on are closely related to me and to each other, that it’s a pretty reasonable conclusion.
Yes.
Probably, yeah. I’ve been having it drilled into me pretty intensely how difficult it is to say you actually know something in science, and how much evidence you have to have backing you up before you can say that something is a fact. The problem is mine, not yours, and my previous post was meant largely tongue-in-cheek.
And it’s been drilled into my head, for similar reasons, to always qualify statements of knowledge. Like the “apparently” weasel-word in my previous post.
Twins share not only genetics but a uterine environment. Have twinned embryos ever been implanted in different mothers and raised to childhood? It would be an interesting experiment to see what effect, if any, the developmental environment would have on appearance. If the effect is large, then knowing someone’s genetics wouldn’t tell you much about their appearance.
Good point. I make liberal use of those myself. Like I said, the problem was mine. 