[Applause]
most of us when we think about decisions
we have to make will go by averages
right so I think you know when we when
we hear the sound of hoofbeats what
would you think of you would think of
horses right but today we’re going to
talk about situations where you think
about zebras and not horses which is a
much rarer phenomenon but you know it
seems counterintuitive right it’s almost
irrational to expect a miracle that
you’ll hear hoofbeats and then suddenly
a zebra shows up and not not horses
right but let’s see why this is
important
the reason is we’re going to talk about
cancer right now cancer is a of course a
challenge today to humanity and many of
us here in fact how many people in the
audience actually have a family member
or a friend or a friend’s parent or who
have actually battled cancer have you so
you see it’s I would say roughly about a
third of the audience has has had that
experience so it’s becoming really
something that really affects everyone
and now with cancer there are two
aspects of technology that are that are
actually starting to help us address
cancer as an individual’s disease not
necessarily only something that you use
evidence-based or statistical decisions
to treat right so so this is sometimes
called the N of one phenomenon in
medicine it’s also being called
precision medicine and for cancer this
is actually not that far away and so
this is the idea
want to converted it now what is
actually causing this and I’m in an
institute of technology and what I would
like to point out is that the there are
you are at a very interesting time right
now because we have two technologies
that have hit exponential scale right
one is of course Moore’s law which which
we know is you know an exponential law
that has made electronics and computing
and communications happen and the it’s
shown here on a log linear scale so it’s
actually an exponential law but it is
what has driven in some ways the fourth
Industrial Revolution right and we’re
all affected by it we carry around a
smartphone today which has pretty much a
very powerful computer in it and that’s
all possible because of Moore’s law well
it turns out that there’s another law
it’s that is also accelerating and that
is called flat Lee’s law and flat leaves
law is about being able to read DNA
sequences in great detail and at
reasonable cost right and if you see the
curve there actually flattens law is
drawing propping faster than Moore’s Law
in the last maybe ten years and this is
called next-generation sequencing and
it’s actually moving roughly at the
square of Moore’s law you see the
timeline during which information
technology has taken off is roughly the
timeline during which the field of
genomics has also developed so around
the time that the first computers were
being conceived you also had that the
double helix structure of DNA solved by
Watson and Crick and then you know
industrialization in information
technology have
a little bit earlier but by about the
mid-80s instruments for being able to
sequence and read off genome data
started becoming available if you look
at the progress in in electronics in you
know 1958 I guess the the first company
that that made semiconductors was set up
similarly for genomics it started in
1985 or so so it’s not a genomic started
off a little bit later but by about 2006
you know we had reached what we call the
second half of the chessboard you know
the story about the chessboard right put
one grain of rice on the first square
double it on the second doublet again on
the third and so forth so when you get
halfway through the chessboard you reach
about four billion now that means that
the technology has improved four billion
times right and that’s very substantial
so once you get past that into that
second half of the chessboard you see
really miraculous use of technology and
that’s what we’re experiencing today
whether it’s autonomous vehicles or
artificial intelligence systems and all
kinds of impact of you know the the
second half of the Moore’s Law impact in
in technology now what what has happened
in genomics is that it went along with
Moore’s law for a while but then it
accelerated even faster
so it so happens that this year in 2017
we have entered the second half of
genomics as well so what does this
actually mean this means that it’s
getting cheaper faster and better to be
able to read the DNA of of an individual
and be able to make certain decisions
based on that and that’s what
this whole thing with precision medicine
and cancer is about right so ah there we
go so what is this DNA this DNA is
essentially a book that we have
trillions of copies of right and it’s in
the DNA of that you’re born with and
that’s actually a book which is written
in an alphabet of just four letters ACGT
and it it’s a fairly long book in the
sense that it’s three billion letters
long right and we all have this it’s you
know I think most of you have had some
introductory biology so I won’t belabor
this now what did the human genome
project achieve it gave us the first
reference genome of of a human being now
so if the black band there is the
reference genome now when I take an
individual patients DNA and start
sequencing it I’ll sequence it in small
bits right and those are called reads
now I can do a string matching and put
that read against the reference genome
and I know where it where it actually
fits right but if you just do one copy
of the DNA you’re likely to make errors
so what we do is actually put many many
copies of DNA cut it up into B’s pieces
sequence those pieces and then assemble
them so that’s a big computational
challenge but it can be done right and
so for cancer patients for example you
take about a thousand copies of the DNA
from the tumor and you will actually
sequence those and put those thousand
copies in fragments in reads as as you
see in the picture there now you can
actually look down a specific position
in there
friends genome and see if the same
pattern is in all the reads and you see
here there is there is a I think it’s an
a which is replaced by a T right that’s
called a single nucleotide polymorphism
so it’s actually something so the the
the easier way to call it is a snip it’s
a snip right and these the snip actually
is such that it is from in some sense
from the the both copies of the of the
DNA which means that it came from both
parents right you can also have snips
which are heterozygous that is it comes
from only one parent right if it comes
from both friends you call it homozygous
if it’s from only one parent equal it
heterozygous right and then you can
start collecting and looking at all of
these snips right now for each human
being there are you know lots and lots
of snips that that arise in the genome
but only one of them or maybe a few of
them together are causing problems
causing a disease or some so you have to
be able to sort it all out so we won’t
get into the details of how you sort it
all that but but let’s let’s look about
look at how this is starting to help
people right and so this is the famous
story of a Ling young boy called
Nicholas Volker who was in some sense
the poster you know child of genomic
medicine because you know he was having
lots of problems with essentially with
digestion and I had to have many many
surgeries and at some point the doctors
suspected that there was some problem
with his immune system but to confirm it
they sequenced his DNA and they actually
found the culprit gene
and they found that mutations which had
caused the immunodeficiency at which
point they were able to get him a bone
marrow transplant and and this was about
maybe eight years ago so so this boy is
now healthy he is going to school then
and so on so so they you know the the
story of how this boy was you know
diagnosed and then recovered is was
written up as a book and it won the
Pulitzer Prize
the next story of course many of you are
familiar with because this happened just
a couple of years ago Angelina Jolie
actually discovered that she has
inherited some terrible genetic factors
for breast and ovarian cancer from her
mother and and she had two family
members a mother and her aunt affected
and and therefore took some bold
decisions on elective surgeries to
improve her chances of not getting
cancer went in later age right so it’s
an example of again you know where
individualized medicine or you know I
said the zebra right comes into play
because in her case she had ovarian
cancer which then later on there was a
relapse in breast cancer and and then we
analyzed her DNA a tumor DNA and we know
not a tumor DNA but her the germline DNA
the DNA she was born with and found that
there was an inherited factor which
which then led to treatment all of this
type of diagnosis today is possible in
India it’s done by this company called
strand which which a bunch of us started
and we have a platform called strand
Ramanujan that actually helps with with
this diagnosis if you want to read about
more zebras stories there’s a book
called genomic quirks that remain Sharan
my co-founder has written which is
available on Amazon and which can will
also shortly be issued by the Indian
Institute of Science Press so
essentially we are entering an era of
precision medicine based on I think the
first impact we are seeing is in cancer
or as in the case of Nicolas Volker in
rare disorders
you know immunodeficiency ease and
lysosomal storage disorders and so on
where you are able to to actually treat
the patient on the specifics of what the
patient’s inherited DNA is and what is
causing the disease and and so I think
in some sense this is transforming the
way in which medicine is being practiced
and that’s the idea that I wanted to
convey today thank you
[Applause]
you
most of us when we think about decisions
we have to make will go by averages
right so I think you know when we when
we hear the sound of hoofbeats what
would you think of you would think of
horses right but today we’re going to
talk about situations where you think
about zebras and not horses which is a
much rarer phenomenon but you know it
seems counterintuitive right it’s almost
irrational to expect a miracle that
you’ll hear hoofbeats and then suddenly
a zebra shows up and not not horses
right but let’s see why this is
important
the reason is we’re going to talk about
cancer right now cancer is a of course a
challenge today to humanity and many of
us here in fact how many people in the
audience actually have a family member
or a friend or a friend’s parent or who
have actually battled cancer have you so
you see it’s I would say roughly about a
third of the audience has has had that
experience so it’s becoming really
something that really affects everyone
and now with cancer there are two
aspects of technology that are that are
actually starting to help us address
cancer as an individual’s disease not
necessarily only something that you use
evidence-based or statistical decisions
to treat right so so this is sometimes
called the N of one phenomenon in
medicine it’s also being called
precision medicine and for cancer this
is actually not that far away and so
this is the idea
want to converted it now what is
actually causing this and I’m in an
institute of technology and what I would
like to point out is that the there are
you are at a very interesting time right
now because we have two technologies
that have hit exponential scale right
one is of course Moore’s law which which
we know is you know an exponential law
that has made electronics and computing
and communications happen and the it’s
shown here on a log linear scale so it’s
actually an exponential law but it is
what has driven in some ways the fourth
Industrial Revolution right and we’re
all affected by it we carry around a
smartphone today which has pretty much a
very powerful computer in it and that’s
all possible because of Moore’s law well
it turns out that there’s another law
it’s that is also accelerating and that
is called flat Lee’s law and flat leaves
law is about being able to read DNA
sequences in great detail and at
reasonable cost right and if you see the
curve there actually flattens law is
drawing propping faster than Moore’s Law
in the last maybe ten years and this is
called next-generation sequencing and
it’s actually moving roughly at the
square of Moore’s law you see the
timeline during which information
technology has taken off is roughly the
timeline during which the field of
genomics has also developed so around
the time that the first computers were
being conceived you also had that the
double helix structure of DNA solved by
Watson and Crick and then you know
industrialization in information
technology have
a little bit earlier but by about the
mid-80s instruments for being able to
sequence and read off genome data
started becoming available if you look
at the progress in in electronics in you
know 1958 I guess the the first company
that that made semiconductors was set up
similarly for genomics it started in
1985 or so so it’s not a genomic started
off a little bit later but by about 2006
you know we had reached what we call the
second half of the chessboard you know
the story about the chessboard right put
one grain of rice on the first square
double it on the second doublet again on
the third and so forth so when you get
halfway through the chessboard you reach
about four billion now that means that
the technology has improved four billion
times right and that’s very substantial
so once you get past that into that
second half of the chessboard you see
really miraculous use of technology and
that’s what we’re experiencing today
whether it’s autonomous vehicles or
artificial intelligence systems and all
kinds of impact of you know the the
second half of the Moore’s Law impact in
in technology now what what has happened
in genomics is that it went along with
Moore’s law for a while but then it
accelerated even faster
so it so happens that this year in 2017
we have entered the second half of
genomics as well so what does this
actually mean this means that it’s
getting cheaper faster and better to be
able to read the DNA of of an individual
and be able to make certain decisions
based on that and that’s what
this whole thing with precision medicine
and cancer is about right so ah there we
go so what is this DNA this DNA is
essentially a book that we have
trillions of copies of right and it’s in
the DNA of that you’re born with and
that’s actually a book which is written
in an alphabet of just four letters ACGT
and it it’s a fairly long book in the
sense that it’s three billion letters
long right and we all have this it’s you
know I think most of you have had some
introductory biology so I won’t belabor
this now what did the human genome
project achieve it gave us the first
reference genome of of a human being now
so if the black band there is the
reference genome now when I take an
individual patients DNA and start
sequencing it I’ll sequence it in small
bits right and those are called reads
now I can do a string matching and put
that read against the reference genome
and I know where it where it actually
fits right but if you just do one copy
of the DNA you’re likely to make errors
so what we do is actually put many many
copies of DNA cut it up into B’s pieces
sequence those pieces and then assemble
them so that’s a big computational
challenge but it can be done right and
so for cancer patients for example you
take about a thousand copies of the DNA
from the tumor and you will actually
sequence those and put those thousand
copies in fragments in reads as as you
see in the picture there now you can
actually look down a specific position
in there
friends genome and see if the same
pattern is in all the reads and you see
here there is there is a I think it’s an
a which is replaced by a T right that’s
called a single nucleotide polymorphism
so it’s actually something so the the
the easier way to call it is a snip it’s
a snip right and these the snip actually
is such that it is from in some sense
from the the both copies of the of the
DNA which means that it came from both
parents right you can also have snips
which are heterozygous that is it comes
from only one parent right if it comes
from both friends you call it homozygous
if it’s from only one parent equal it
heterozygous right and then you can
start collecting and looking at all of
these snips right now for each human
being there are you know lots and lots
of snips that that arise in the genome
but only one of them or maybe a few of
them together are causing problems
causing a disease or some so you have to
be able to sort it all out so we won’t
get into the details of how you sort it
all that but but let’s let’s look about
look at how this is starting to help
people right and so this is the famous
story of a Ling young boy called
Nicholas Volker who was in some sense
the poster you know child of genomic
medicine because you know he was having
lots of problems with essentially with
digestion and I had to have many many
surgeries and at some point the doctors
suspected that there was some problem
with his immune system but to confirm it
they sequenced his DNA and they actually
found the culprit gene
and they found that mutations which had
caused the immunodeficiency at which
point they were able to get him a bone
marrow transplant and and this was about
maybe eight years ago so so this boy is
now healthy he is going to school then
and so on so so they you know the the
story of how this boy was you know
diagnosed and then recovered is was
written up as a book and it won the
Pulitzer Prize
the next story of course many of you are
familiar with because this happened just
a couple of years ago Angelina Jolie
actually discovered that she has
inherited some terrible genetic factors
for breast and ovarian cancer from her
mother and and she had two family
members a mother and her aunt affected
and and therefore took some bold
decisions on elective surgeries to
improve her chances of not getting
cancer went in later age right so it’s
an example of again you know where
individualized medicine or you know I
said the zebra right comes into play
because in her case she had ovarian
cancer which then later on there was a
relapse in breast cancer and and then we
analyzed her DNA a tumor DNA and we know
not a tumor DNA but her the germline DNA
the DNA she was born with and found that
there was an inherited factor which
which then led to treatment all of this
type of diagnosis today is possible in
India it’s done by this company called
strand which which a bunch of us started
and we have a platform called strand
Ramanujan that actually helps with with
this diagnosis if you want to read about
more zebras stories there’s a book
called genomic quirks that remain Sharan
my co-founder has written which is
available on Amazon and which can will
also shortly be issued by the Indian
Institute of Science Press so
essentially we are entering an era of
precision medicine based on I think the
first impact we are seeing is in cancer
or as in the case of Nicolas Volker in
rare disorders
you know immunodeficiency ease and
lysosomal storage disorders and so on
where you are able to to actually treat
the patient on the specifics of what the
patient’s inherited DNA is and what is
causing the disease and and so I think
in some sense this is transforming the
way in which medicine is being practiced
and that’s the idea that I wanted to
convey today thank you
[Applause]
you
Please follow and like us: