The project began when I was sitting in a therapy session, and I was looking at this
very generic, very ugly painting on the wall, and as I looked closer at the print, I noticed
that there was a crack in this glass, and as I looked closer at the crack, I noticed
that in it was lodged a single hair.
When I left, I couldn't stop thinking about it.
As I walked down the street, as I came home, I just kept thinking about that hair and thinking
about whose hair it could be, you know, what they might look like, what they might do.
I just kept seeing hairs everywhere, and I kept thinking about all these forensic shows
that we watch on TV, and the fascination that we have with that science, with the science
of trying to figure out from evidence who was there.
The current project that I'm working on is called Stranger Visions, and it is essentially
an exploration of what we can learn about a person from a hair.
And it also explores this kind of possibility for genetic surveillance.
So things like hair, skin, saliva, we're constantly shedding these traces all over the place
in public, and not even thinking about it.
So I'm going to cut just a tiny bit of the root end of the hair.
In particular, we're looking for 1.5 to 1 centimeter piece.
The process begins when I discover a hair, it might be in a bathroom, it might be in
the subway.
Try to get them at the bottom, and I'll bring it to gen space.
Gen space is a fully functional molecular biology lab where people can come from very
diverse backgrounds and experience DNA-based technology in a very visceral hands-on way.
Once you kind of know what you need to do, you start going in, you do your internet research,
you start looking through the articles and the papers, and then you come across the protocols.
And so the protocols will list step-by-step, this is exactly what you need to do, like
a recipe.
So now we're going to add some proscienese K, which helps break down the cells.
Originally I was just thinking, you know, I'll see if I can extract DNA from the hair,
and then I'll see what that can turn into, maybe it'll be some kind of portrait.
The first thought I had was to do a two-dimensional portrait, and then that idea eventually grew
into the idea of doing a full physical sculpture to print an actual 3D model.
In a way, this project kind of came out of these popular concepts of science.
So when you watch a TV show, you know, you watch a show like Bones, for example, where
you know there's some tiny fragment of something is found and suddenly it gets turned into
this entire person's face.
And there's always an element of truth there.
So as I started looking into forensics, I started wondering what I could know.
Assuming that from a hair I get a perfect full genome, you know, if I can have my pick
of that genome, how much can I possibly know about what this person looks like?
In my initial research, I discovered that there are a set of SNPs, and SNPs are single
nucleotide polymorphisms.
SNPs are just parts of the genome that vary by one base pair, person to person.
And that particular base pair is linked to traits or disease risks.
So that single SNP is a clue, you know, that you probably have brown eyes.
You probably have blue eyes.
This is the SNP spreadsheet, so what you can see here is that I've organized this list
of locations on the genome that are linked to traits that I'm interested in.
There have been large studies done with people where all of the SNPs in their entire genome
have been looked at, and then that data has been correlated with things like their susceptibility
to certain diseases and appearance, maybe hair or eye color.
A lot of the traits that are common traits that we think of associated with appearance,
like hair color and eye color, are the result of multiple genes.
I heard an estimate that your height was something like cooperativity between 10 different genes.
So a gene could be composed of different SNPs.
It could be, you know, just one, it could be 20, it could be 100.
And so, you know, the correlation is not always a simple one.
In order to do this project, what's required is essentially sequencing 40 different locations
on the genome.
So this is the DNA at the bottom of the tube, and what remains in the filter are the cells
from the hair.
So from that DNA, then, I can check for certain traits.
Okay, so I'm going to run two PCR reactions, one as the control, and one as the mystery.
So there's different ways of extracting DNA.
There's also different ways of analyzing the DNA.
The kind of lowest tech, oldest method, which is the one that I've been using so far, is
to run PCR, which is polymerase chain reaction.
So what PCR is, is it's a process by which you can amplify sections of the genome.
So by taking DNA and adding what's called a primer to it, I can say, basically, here's
an in and out point on the genome, right?
I want you to take everything between this and this and amplify it and make a bunch of
bunch of copies of it.
I'm going to add this primer to these little containers here, these tiny little guys.
What Heather's doing is she's using primers that have been designed to target a region
that tells us something about a person's ancestry.
And then moving into areas that code for propensity to be overweight or not, eye color
and eye color is actually one of the first traits that has been rolled out to the police
in law enforcement around the world to use in a kit.
So there is now a kit that is available.
It's called Iris Plex.
And this kit allows a detective at a crime scene to take a DNA sample and very quickly
ascertain the eye color of the suspect.
And then I'm also, although I'm not including it in the portrait set, I'm also looking at
hair, so looking at hair color, hair curl, areas that code for lighter versus darker
skin, freckles and then gender.
So this involves making new primers to look for those particular sites within the genome.
What I've been doing so far is actually using kits.
So the next step for me actually is really figuring out how to get into that primer design
or to find someone who can help me with that.
That's around 500, 500 bases, okay, that's perfect.
Oh, that's great, what good news.
And once you have that amplified section of the genome, you can take it and send it to
a company for sequencing.
And so what I get back from them are these files, which you can open up in any text editor,
and you can see very simply the result of that sequencing, right?
So you can see the code here, just a written noun.
I feed this text file into a custom program that I wrote.
So the program does two things, first it'll spit out a list of traits.
Now I can run this DNA, get traits, I can add this in.
And the result of the analysis?
Okay, now this says, Northern European female, heavier, heavier, skinnier, skinnier, skinnier,
heavier, so it sounds like there's a 50% chance here of being overweight or not.
White skin, European, 56% chance of brown eyes, 37% chance of green eyes, straight hair,
and freckles.
And then it will call another program that I wrote, which generates a face.
And so the face generating program was based off of some facial recognition research that
a team was working on in Basel, Switzerland.
And so what I did was I took that framework that they were working with, and I kind of
expanded it and turned it around so that it could become a facial generating program.
We're not at the point yet where we can take DNA and know exactly what somebody looks like
or where they came from, but in a lot of cases we can get a pretty good general idea.
One of the mysteries is why these SNPs are associated with particular traits.
Sometimes it's in a very obvious gene, like hemoglobin and a red blood cell disease.
However, most of these correlations that have been found aren't as easily explainable.
That kind of probabilistic nature is something that I wanted to probe with this piece.
I'm really interested in that uncertainty, and I'm interested in exploring what we're
uncertain about now, but maybe certain of in the future.
And I'm actually working up at RPI doing my PhD.
And what I'd really like to do is also to explore age.
And the research for age is still new, and one of my goals for the semester is at least
to ascertain whether it's possible for me to figure out the age from a sample.
And this new study that came out recently about facial structure, so they found some
genes in Europeans which code for specific arrangements of features on the face, locations
being like the cheekbones, the centers of the eyeballs, tip of the nose, top of the nose,
bottom center of the nose, and then the edges of the nostril.
Heather's work is really interesting because it's a very accessible way for the public
to engage with this new technology.
It really brings to light how powerful it is, the idea that hair from your head can
fall on the street, and a perfect stranger can pick it up and know something about you.
And with DNA sequencing becoming faster and cheaper, this is the world we're all going
to be living in.
So the biggest influence this project has had on me is that now I'm very aware of everything
that I leave behind.
So when I sit on the subway and run my hand through my hair, I notice if there's a hair
on my hand and it falls to the ground in the subway, I pay attention to all this stuff
now.
If I go to a bar and have a beer and leave a little saliva there in the bottom of the
cup, I think about it now.
