path: root/transcripts/open-science-summit-2010/raymond-mccauley.mdwn

DIYgenomics

I have the hard task of coming to wrap up. I want to show something that is
not as fantastical, it's very down to earth. I have a simple way where we've
tried to apply citizen science to some things that we're doing. So, this is
who I am, and a couple of places I've worked. Right now I do some work for a
biotech company where we work on DNA sequencers. When I was doing this, I got
really interested in the ideas of what yuo can do with all of the sequencing
datat hat is being generated. Something kind of obvious that hit me, was that
the tools are getting so much cheaper and much more accessibble, but the
information that we were getting uot of these things was fairly scarce, we
were getting a lot of data but not a lot of meaning. What would it look like
to start relating genomic data to something that mattered to people on a day-
to-day basis? That was my big question. We tried to come up with a couple
things, and some folks I met through the Biocurious Labspace in the DIYbio
meetups, we were just sitting adn talking after these meetings, a good simple
example as these things became cheaper and more accessible, so we came up with
a very little question, which is do vitamins work for me?

And so there's a gene that produces a protein called MTHFR that has received
some notority lately because it's a protein being tested on the On The Same
Page program, and the incoming freshman are looking at. What it is is it's a-
taking an inactive form of vitamin B9 and converting it to the active version,
which is a nice job for a protein to have in the metabolic system. If you have
one or two mutations, it doesn't happen, several different things, but you can
tell because homocyssteine levels rise. One of the proteins at the end of the
cascade converts homosystein to something else. Looking at 23andme results, we
found out about the mutations in this, does this really matter? Does this
matter for my vitamins and the foods I eat? Does it get converted into
anything? We sit around and think it's exciting to think about whether your
vitamins are working. Can we do some little experiment with thiis? We weent
aahead and took our dataaa from a 3andme genootype and tried... wwhat happens
iif we doo nothiing, iff we ttake a regular off-the-shelf vitamin, and a
special form of this vitamin,w hich is the direct active form of B9, and it's
more expensive than the regular vitamin. Do these things make a difference,
can you see? Can we test this by looking at plasma levels of homocysteine? We
wanted to do this with off the shelf tools.

So, when we started to talk with people about it, yuo ould would not think you
have a groundswell of support for this where people had to give blood and pay
for it. We found some people to do this. Before we started to enroll people
and doing a real study, we went through and said let's do a little pilot, so
we're in the middle of this right now, I am Citizen Scientist #2, and I have
the most serious mutation (a duoble mutation). The data that we found so far,
actually shows that highpoint 1 is the washout where nobody is taking any
vitamin, the other one is a regular vitamin, and #3 is the special vitamin.
This is preliminary and not statistically significant. My homocysteine levels
wentt up wheen I took the vitamin. My body thought that B9 was good enough and
shutdown the other pathways that would process and make B9 work. When I do
nothing, I am better off. Everyboddy else had aa differentt resultting.
Heere's aa littlle questiiion. Pulling some money together and doing different
things. This is random preliminary data.. higher is worse, and the horizontal
axis is the three different interventions, the wash-out, the regular vitamin
B, and the special form of B9 being the third one. I wanted to show this to
show you about these results, it's really not great science- it's really
simple. People have done thingsl ike this before, but the trick is that people
have done these things ina cademic labs or from commercial sponsorship to
determine what's going on with a patentable drug. what we found from this is..

What we found from this is that it's really dead easy from the standpoint of
technology and access to tools. Blood tests over the internet, you show up,
get a paper, and in 24 hours they email you the results, and getting access to
23andme results, maybe a little expensive for recreational genomics, it's
pretty cheap pretty quick. We didn't have much trouble with the self-funding
for our project, we wanted to do a prototype-based approach where instead of
thinking and talking about, we would just jump into it and see what mistakes
we would make. But the hard part is really been anything where we're
connecting with traditional scientific institutions, specifically funding and
so on, when we ask them if this is interesting for funding or whatever, or if
the results are interesting, would you help us publish it? And they say yes,
if you have IRB approval. But that's a good thing, but it's a long and slow
institutional process. We have a very participative approach. We have friends
who have got together in the living room or on the internet, let's talk about
the risks, let's talk about what we're going to produce, so it has been a
little different. So just to wrap up, I want to say that the idea that a
scientist who has a PhD or somebody who works at a ceratin place like a school
or sponsored lab, is not at all the case. A scientist is someone who has
questions and wants to find answers. In genomics, it's so cheap to get all
this data, we don't have enough scientists to figure out what all this
information means to individuals. As this becomes more democraticzed, I think
that we all will become scientists and answering our own questions, and I hope
this is a small model for helping them. So, a scientist is someone who has
questions, we would all like to hear your questions. Thank you.