The Future of Medicine Computational Chemistry
[Music]
[Applause]
[Music]
hi
today i’d like to talk to you about the
future of medicine computational
chemistry
um so just for a quick introduction hi
my name is sarah i’m a sophomore at los
altos high school
and i’m also a high school researcher at
aspiring scholars directed research
program
which is basically a program that allows
high schoolers to conduct their own
original research
and so there i’m a synthetic organic
chemist and what i work on is in the
drug discovery process
and that’s how i can tell you that the
drug discovery process takes a really
really long time
you start off with something called
analog design where you figure out what
you even want to make
and oftentimes people will want to make
libraries of compounds even up to like a
hundred
and then you go on to chemical synthesis
is and that’s like where you actually go
into lab
and you make whatever you said you
wanted to and this process takes a
really long time because
you have to start off with setting up
your reaction
and then after you set up your reaction
you have to purify it and finally you
have to characterize it
and these three steps can take up to
weeks
but then you combine the fact that
perhaps he said that you wanted to make
100 compounds in the beginning
you have to do this process over and
over again 100 times
and oftentimes this process takes up to
years to even complete your compound
library
and that’s barring any complications
and finally after you’ve spent perhaps
two to three years making your library
of compounds
you can finally test it and imagine how
disappointing it is if you spent two to
three years
working hard to make all your compounds
and then you find out that none of them
work
obviously this process is highly
inefficient because we see this happen
all the time and that’s where
computational chemistry can step in
computational chemistry can pre-screen
our compounds before we actually go into
the lab
and synthesize them and spend inordinate
amount of times
trying to make them and finally test
them
and so to do this we use computer models
for drug discovery
it starts off with de novo molecular
design which is almost analogous to the
analog design of the traditional drug
discovery process and it’s basically
where you
figure out what you want to screen then
you go on to something called density
functional theory and this is something
you can just download onto your computer
it’s completely open source
and you just plug in your molecule
through there and it comes out and it’s
completely optimized it’s like it would
be in the real environment
then after that you can use something
called molecular docking where you see
how your compound interacts with its
target
so for example if your compound’s
supposed to interact with the protein
this molecular docking can first of all
show you the orientation of it
but second of all it can tell you how
well this
compound would bind to the protein and
that’s really powerful
because if your compound binds really
really well to that protein
then you can basically say hey this is
actually worth synthesizing
but if it doesn’t bind very well to the
protein in the first place you don’t
want to be spending those years
making it and then finally you can do
something called a molecular dynamic
simulation
and this just shows you a real-time
simulation of your protein against your
compound
and this is really cool because you can
see perhaps your protein
folding or something and all of this
just goes to show that there are so many
computational techniques to help you
refine the drug discovery process
and so at our lab we have this 26 core
server which we use to screen 400
compounds to find a treatment for
covid and we basically just used those
400 compounds and we plugged them into
the computational chemistry process
and so here are a couple articles from
the stars cup 2 outbreak a year ago and
i think it’s really noteworthy to
mention that we actually started our
research
on kovid right at the outbreak of the
pandemic
and so even though no one was allowed to
be anywhere in person we were able to
continue our research
because all of our research was now
stored on our computers
and so if you want to figure out how to
cure covid you first have to understand
how covid works in your body
and you can kind of think of it like
this it’s like a long chain of events
and at the end of it covet proliferates
inside your body making you sicker and
sicker
but if we cut off one of the links to
those chains
covid won’t be able to grow through this
process and it’s not going to be able to
proliferate
and so that’s exactly what we did we cut
off this process right here
where there’s an enzyme in your body
called the main protease and it binds to
the covid
protein and since we can stop that
process
we can stop covid from being able to
proliferate inside your body and so to
do that
this is an example of our inhibitor and
the bottom thing is basically the
protein that we were trying to inhibit
and you can kind of think of this as a
door stopper our inhibitor is a door
stopper
and it’s permanently stuck there so that
covid can no longer enter it can no
longer bind with the protein
because our inhibitors already stuck
there
and so to do this we actually use
computer models for drug discovery
to test whether or not our inhibitors
would actually work before
we went into lab and synthesized them so
first we start off with some biological
inspiration for
our protein so this is the native
compound that binds
to our enzyme and we just militarize
that so this is basically the de novo
molecular design
portion of the computational chemistry
process
and then after that we plugged it into
the rest of the steps so the density
functional theory and the molecular
docking
and once we did that we identified our
hit compounds
so you can see here these are our two
best compounds and you can judge that
because they have the two best binding
affinities
to our covid enzyme and because of this
we have cut down our library of 400
compounds
to simply two and synthesizing going
through the traditional drug discovery
process
with 400 compounds might have taken us
even 10 years
but two compounds could take us a matter
of months and not only that it took us
only two months to even um screen them
on the computer
and so i just really really want to
emphasize how cool computational
chemistry is
not only that it’s improving the drug
discovery process
by making it so much faster and so much
more efficient but the fact that
it is so accessible the fact that
teenagers like me and my friends are
able to use this program
to find treatments for covid really mean
something and i hope
that computational chemistry will
improve the future of medicine
by not only making it more efficient but
by making it more accessible
and creating a safer world for everyone
thank you