music
Herald Angel: Well, a lot of hackers in
this room, but I don't know if you knew
that, but but every being on this planet
has a source code. Even your cats, your
girlfriend, your boyfriend and your dog.
And the next speaker actually know how to
read the source code of human beings - the
human genome. Anna Müllner is a German
medical biotechnologist and she completed
her PhD in cancer research, but she is
also a science blogger, podcaster and
science slammer, known under the name
Adora Belle. She says something that's
quite interesting for me as a privacy
activist, she says genome sequencing
provides us with opportunities for medical
and biological science, but with
challenges in ethics and privacy. Please
give a round of applause to the next
speaker, Adora Belle!
applause
Adora Belle: Yeah, hello and welcome to my
talk about genetic codes and what they
tell us and everyone else. You might
realize it's a little different code than
most of you are used to, so at first I'd
like to introduce myself, to say who I am,
and how did I get here. Katasha already
told you some of that, so I can go quite
quickly.
I'm a biologist and I did my PhD in cancer
research, so I'm always interested about
cancer. In this talk will have a slight
influence by that. I'm blogging, I'm
podcasting, I'm talking about science and
slamming about science as well, but what
people might actually wonder is how did I
actually get here?
And you may remember a couple of years ago,
the Chaos Communication Club, Chaos
Computer Club, sorry, they stole the
fingerprint of Wolfgang Schäuble, because
he wanted to put the fingerprint of each
of us on our ID cards. And as you might
have realized the last time you renewed
your ID card, that actually happened. And
at the time I was in Scotland and I was
doing a course in forensics. And it's
actually me at a mock crime scene, so no
people were harmed but I was quite sweaty
under the suit so... and I realized that if
you take the glass of someone they drinks
from and you take a fingerprint you can
also take the genetic fingerprint and
actually
there's a whole genome of someone on
there. So to quickly summarize where you
can find genetic information: You can find
it just about anywhere. You can find it on
shed skin cells, in the saliva, and blood
and hair, in urine, and feces, and sperm, and
vaginal fluid, and we actually, we spread
it all the time, and for example, you have
it on your toothbrush, on your hairbrush,
on your keyboard, other personal items
like your cell phone, it's even on used
condoms and remember, there's probably also
the DNA of someone else on there, and also
you have it on other people. So if you
scratch someone you have their DNA. If you
lose hair and skin cells, if you touch
people you will transfer some of your
genetic information and also, like I said
with Wolfgang Schäuble, on glasses, on
letters that you lick to close them, on
cigarettes that you smoke and also some
part of your DNA, and this is quite
important, is in your relatives. Because
you share the genetic information to some
extent. So to quickly introduce the terms.
"Genetic fingerprints". Genetic
fingerprints actually provide no personal
information as such, besides the biological
gender, that's what you can actually check
for, but the other information is nonsense
information, that will not tell you
anything about the person. But it gives
you a positi... possibility to find
relatives since you share these
information, and it gives you the
possibility to reidentify people and it's
a unique sequence. And if you compare this
to a whole genome, which is becoming more
common now in research, you can find the
biological gender, of course, you can find
the ethnicity of a person, you can look at
genetic diseases, you can find out
something about the looks of the person,
and you can find out things about their
relatives as well. And as we do more
research, we will find that there will be
even more information to come and similar
to the fingerprint, you will be able to
reidentify people with this unique
sequence.
And in between the fingerprint
and the genome there will be a lot of
genetic profiles that have some kind of
reach depending on how far you want to
look, how deep you want to look.
And of course people said, this is quite
important information,
so they said the human genome will
actually be a final frontier in biology,
because it actually, it is our source code,
and this is what makes us us. And so they
started the Human Genome Project and said
if we sequence all this DNA, then we will
be able to reach kind of the Holy Grail
and they expected 100,000 genes and Bill
Gates then said, this would be the
language in which God created life, so
this was kind of a, yeah, a major goal
that they wanted to do. And then they
found out that there are only 19,000 to
20,000 genes, which is about the same
number as nematodes, so - that's a kind of
little worm - and four times more than the
bacteria in your gut, and so the leader of
the project, Craig Venter, whose own
genome was sequenced, he then said "we
don't know a shit", he said, we have no
idea what it means. And there they said
then, well, to find out more we actually
need to sequence more genomes. And this
does make sense, because, if you want to
find answers in the genome, then you have
to compare these genomes, and this then
lead to different projects, which are still
going on, like the 1000 Genomes project, the
10,000 autism genome project, the
100,000 genomes project in the UK,
and the 1,000,000 genomes project in the
US, and there are other genome projects,
like for example the Cancer Genome Project
and these all aim at a kind of
personalized medicine, so to compare your
DNA and then adjust treatment to your
genome. But as I will tell you soon, these
will not answer all the questions, because
DNA is much more about the regulation. Our
DNA is regulated, in a way that it's more
mobile and agile to respond, or, actually
the DNA is quite static, but the
regulation of the DNA actually made... makes
it extremely adaptive. And then there is
another thing, because
as I said, DNA is a code. It's similar to
your code that you do for a program, but
then how the user uses this program can
vary extremely widely, so you will find
that, even though the similar... the
information is similar in people, it might
look different in the person itself. So,
just to give you a quick impression about
how these interactions look like. These
are the products of the genome, or some of
the project... products, they're called
proteins and these interact with ... with
each other, and we will find that they have
all these interactions, all these
crossroads. One interacts with the next
and this then inhibits something else and
this is quite complex. But still, DNA
research does have its uses, so it will
provide us with valuable information, but
what you need to keep in mind is that is
it valuable for whom? So it could be used,
the whole genome sequencing could soon be
used instead of specific tests, because
it's becoming more and more cheaper all
the time and it would give us the
possibility to study specific genes in a
population, a genetic disease, inherited
cancer and genetic risks. And so, since I'm
a cancer researcher, I'm doing a quick X
course here. You can study single gene
diseases, which are usually, then, if you
have a mutation, and you find this in the
genome, the person will have the disease
with a very very high likelihood. There're
very rare cases, which this does not
happen but these are very rare, these
single gene diseases and also there are
some cancer genes. These are genes, that
we all have but if they are mutated they
will very likely lead to cancer in a very,
at a very early point in life. And you
might remember Angelina Jolie, who found a
cancer gene or gene in her, that was ...
that would lead to her having breast
cancer very early in life and so she had
her breasts removed and this also exists
for colon cancer and there are also
special syndromes, which lead to having
more or being more susceptible to cancer,
so these are also rare, actually. But it
is very likely, that in the Western world
we will die from two major causes,
one is cardiovascular disease and the
other is cancer. And when you think about
how to not get cancer I always say: It is
to be boring, so you need to have a
healthy lifestyle, so no smoking, less
drinking, staying fit and not eating too
much and avoiding radioactivity also plays
a big part and not go into the Sun without
sunscreen and to accept your screening
appointments with your doctor, but even
then I always say that cancer is mostly
bad luck, which is also the opinion of
many cancer researchers. And if you don't
get it, it just means that you have not
died of something else earlier. Because it
is a disease that comes with age and it
gets more likely to have this disease as
you become older and genetic risk factors
then play very little role, actually. So
then let's talk about the private genetic
sequencing companies that are sprouting
up everywhere. You might have heard of
23andme, which is a mail in genetic test,
which tests you for diseases and ethnicity
and they sequence over 500,000 gene
locations. There's a similar company
called ancestry, which just check your
ancestry, so it's a kind of ethnicity. And
there's the ingenia.com "surname project",
which compares your DNA to a male lineage
in the male lineage with the last name.
And also now we have whole genome
sequencing companies like the "Full
Genomes Cooperation", "Guardium", "Gene by
Gene" and even more. And you have to
realize that these companies will have
quite some genetic information stored. And
this is all nice and safe since the
government in America - where most of
these companies are - has repeatedly shown
to respect privacy. And of course that
won't change under the new president, I
think. so... already in 2010 Kashmir Hill
an author at Forbes, she wrote an article
called "Genome Hackers" where she showed
a lot of foresight. And - I'm just going
to quote here - "As gene tests become
common, possibilities for abuse will
intensify. Banks might not offer you a
mortgage if you were likely
to die before it was paid off. A pregnant
woman might secretly get DNA from her
lovers, so she knows who the father is.
Someone might check out a potential mate
for genetic flaws. Politicians might dig
up dirt on their rivals. Another question:
How far should law enforcement be allowed
to go? Should prosecutors be allowed to
subpoena a company's DNA database of
thousands of people if they suspect it
contains a match to a crime suspect? And
then a year later, Robert Langreth, he
referred to this article then said: "I
think this issue is just starting to
emerge. It will be a classic conflict
between scientists' desire for more data
and Americans' desire to keep sensitive
personal information private. If your DNA
is an easily accessible database,what are
the limits of what bureaucrats can do with
it?"
And what I find quite interesting here is
that he just makes this between scientists
and privacy oriented Americans, when we
think about it today it might be even more
like companies and people who don't really
think about privacy. And then we come a
little year, a couple of years later and
actually this is what then happens. So
23andme and ancestry were repeatedly asked
by law enforcement to hand over DNA
databases.
And they actually disclosed that they have
five DNA samples that they gave to the
cops and so one case for example, which
was not 23andme, but that's a filmmaker.
And there was a cold case and there was
DNA on the murder victim and they compared
it to a voluntary Y chromosome database or
so. A male lineage searched. And they
found out that the DNA on the murder
victim belonged to someone who was related
to someone in this database and that
someone actually had a son so they said,
well, then we test him. And so they found
filmmaker Usry and they tested him.
However the complete DNA then did not
match and he was cleared of the charge and
a statement by the privacy officer of
23andme kind of shows what we are getting
into because she said: "In the event we
are required by law to make a disclosure
we will notify the affected customer
through the contact information provided
to us, unless doing so would violate the
law or a court order."
So if your DNA becomes interesting in a
crime case they might tell you but they
might also not tell you and of course then
there's this "I've got nothing to hide and
the suspect was cleared so it will all be
in order after all" - just always remember
that there can be planted evidence.
Because, as I said, you shed your DNA
everywhere. And it's quite easy to obtain
your genetic information or place it at a
crime scene which might at least lead to
confusion. And there could be
circumstantial evidence so that a crime
happens somewhere where you're working or
where you are often. And always remember
the Heilbronn Phantom case, where they
found the DNA of a woman at completely
unconnected crime cases and this DNA was
later found in the swabs that they used to
test the evidence. Because the lady who
had produced these swabs had contaminated
the swabs and as you might remember this
led to a lot of confusion in the research
or in the criminal investigation. And if
you think that data is the new oil, it
truly is. For companies like 23andme who
have reportedly sold genetic data to
private companies and this was - of course
- for research. And they did it with 1.2
million genetic profiles that they have in
their database for parkinson research. And
there seemed to be more deals planned and
Anne Wojcicki of 23andMe, she said that
she wants the whole world's healthcare
data accessible to everyone. And of course
they do have a consent form and this is
signed by about 80% of customers, which
probably think "Well, if if I can help
with my DNA to do some research that's
fine." So, but do they really know what
they're getting into? And this is why I
would like to come to Genetic Sequencing
Privacy. And if we talk about privacy we
have to think about for whom is a genome
interesting. Since the 23andme test for
example is a mail-in tests so you could
send in the DNA of someone else and then
you could test someone else on their
genes. And this could be a prospective
partner - if they have
good DNA to have offspring with, maybe, or
might die early, or might die late. Family
members - if you want to know if your son
is really your son, your daughter is
really your daughter, test for paternity
and maybe if you want to know if you were
adopted. Insurance companies might be
quite interested in this data. Employers
could be interested. And prospective
parents could be interested, because you
can test - from just one cell - the genome
of an embryo for example. And who knows
who might else be interested, once more
information becomes accessible. And the
price is quite cheap actually to do. The
genetic test with 23andme is 200 dollars,
the price of a whole genome is now below
1,000 US dollars and the price will
decrease further. So we could be, should be
quite weary about what's going on. Because
for example 23andme blogs completely
openly about what they do. And they
connected a man to his biological father.
But this was not because his father had
entered his DNA in the database it was
because his cousin had. So someone put...
got tested by 23andme, 23andme said well
here is your cousin and then he found out
that this... that there could be his
father that he had been looking for. And
quite interestingly could be faith so
there's an African American woman who was
always interested in the Jewish faith and
then found out via 23andme that she is
related to the Ashkenazi Jewish tribe. And
today this information is well just
information, it doesn't really matter to
us.
But just remember if this information had
been available 70, 80 years earlier. And a
similar example could be Indian castes. So,
the caste system in India is outlawed. But
if you're still a traditionalist there,
you could test people to which caste they
belong to and discriminate against them.
And then there's another... another case
that just happened this year, where there
were plans of testing... gene testing at
the Kuwaiti border and they say, of
course, "This is anti-terrorism.", which
does not really make sense. Because you
need some DNA to compare and to find
terrorism and terrorists. There is no
"terror gene". And what could be the real
reason could be to keep out non-Kuwaitis,
because they have nomads, like beduines
that they don't really like. And that they
could also test family members and then
put them under pressure if they might have
an illegitimate child or if their wife has
been unfaithful just to put on some... yeah...
some kind of bad information about them.
And when I was doing research for the
talk I found quite interesting that a
blogger had his whole genome sequence and
he got a hard drive from Illumina and this
hard drive was actually encrypted and
wasn't encrypted by TrueCrypt. So this was
2 years ago and we now know that this
might have not been completely safe. And...
but we also have to take a... take up that
genetic data can be useful but we have to
have this compromise because it can be
misused.
And since it does have relevance in
research they are... there's a large
amount of genome stored for research
purposes at many institutes. And David
Goldstein said at the Institute of genome
medicine at Columbia University that there
is an irreversible drive toward obtaining
more and more complete genetic
information. And we are all going to be
sequenced the question is just who does it
and what is done with it. The challenge
will be to do good things with the data.
And if you want to do good things you have
to share the data and the genomes need to
be compared. And their data size is a
problem, because genomes can be extremely
large and depending on the coverage of the
data and of the genome and there's about
200 terabytes stored in Amazon Cloud for
the 1000 Genomes Project and there's also
now Google Genomics which wants to help
you with a big data of genomes.
And is that worth it? Well, maybe genome
research can be worth it for specific
purposes during research and to adjust
treatment of diseases, which works to a
point. But also in forensics. But then we
have to make up which limits. And for the
individual person genetic tests are
probably not necessary unless your doctor
advises you to. And you have to wonder if
your ancestry really matters that much to
you. And always keep in mind that this is
not just your information. It's also the
information of your relatives. And do you
really want to know what the test tells
you? Does it... If it comes up with a
genetic disease that cannot be treated -
do you want to know? And also if the DNA
gets out there, if your genetic
information is disclosed and you're
connected to it, you cannot change your
DNA. It will always be the same and you
can always be recognized by it.
So I'd like to thank you for all for your
attention and I hope you have some
questions for me.
Applause
Herald: Thank you so much for this talk. We
have six microphones here on the ground
floor. So if you want to... If you have a
question, you can line up there and we
still have some time left. There was one
question, at number 1.
Mic 2: Yeah, thank you for your talk and
for the information.
AH: No, number 1.
M2: Oh, sorry.
laughter
Mic 1: Hi! So you were talking about this
problem with people wanting to share
their... their genetic sequence for
science, but on the other side you have
the problem that the scope of that is not
obvious. Could you solve that by like
putting everything into public domain.
AB: Public domain of all genomes?
M1: Yeah. I don't know it's just a
thought.
AB: That would be kind of the post privacy
approach that you're all... hold... or
that the genomes of the world are all in
public domain like this Columbia professor
said, maybe. Well, that's not really
solving, that's just saying "Okay, if we
have the information of everyone
available, then no one can be
discriminated against because there's dirt
on anyone... maybe?" But I don't know if
that's the correct way, because we have to
make a decision for 7, 8 billion people on
the world, so...
Herald: Thank you for this question. We have
also some question from the internet.
Signal Angel: As you don't want to give
genetic information to corperations and
government is it possible to... to do the
test at home and how much would it cost?
AB: So... the testing of the DNA is done
with 23andme but you can do the... you can
disagree to share the information, so...
and then you'd have to hope that they do
it is... like such. Or as such. But that
doesn't really kno... But then I don't
know if it might still come up for police
investigation, still. So doing it at home
would be quite difficult, because the
sequencing machines are very cost... or
very costly and very difficult to use. But
there was talk about doing it with a
smartphone. To have a just a tiny device
who does this for you. But I have not
heard that this is now accessible yet.
Herald: Another question from number 3.
Mic 3: Hi, from your expert point of view,
have you been thinking of or are you in a
database for a bone marrow donations and
what do you think about that?
AB: Yeah I actually am and I did this when
I was 16 and wasn't really thinking about
it. I think now that this is... they will
probably not take your whole genome, but
they do have some information on me stored
and they might even have the probe still
stored. Like the blood that I gave at that
point. So they could still be doing - if
they were criminals - just test my genome
for that. So yes, but I am in the database
and also a blood donor. So my blood is
somewhere out there, all the time, and...
M3: And have you been thinking of revoking
it? Like... Maybe you can revoke your
database entry?
AB: I think I could... But for this... As
long as I don't know that they actually
take my genome out of it, as long as they
just store the information on my... yeah,
my major histocompatibility complex so
that's what they what they look at. But
they look at it genetically. I just hope
to do some good, but, yeah you're right.
They pro... they probably don't have my
whole genome as such, as information. But
they do have some genetic information and
they do have my probe stored, so...
Herald: Thank you for this question. There is
another question from the internet.
Signal angel: Do you think these kind of studies are
already carried out secretly from our
samples we give to health care orgs just
like blood giving?
AB: Well, well,... If they do it secretly,
then I don't... probably don't know about
it. But... So that's quite difficult to
answer. But it could be possible,
especially in regimes where there's no
democracy, for example. And but... I'm not
sure if this happens, because I don't have
that kind of information.
Herald: So, another question from number 2.
M2: Um, hello! I think there was a project
that, instead of working with a lot of
different genome sequence, try to work with
a single sequence and branching for
basically branching the little differences
for... for everybody. Would that solve the
privary... the privacy problems a bit?
AB: To just look at the differences to
other genomes?
M3: Yeah or would that there... we know
the research and stuff...
AB: Well this... So there is a format that
just checks for differences in the genome
which is... gives you a much smaller data
size, so you have your common genome and
then you have the data si... uhm... just
what... the delta of it. And... but this
will actually just give everything that is
not... well, not normal, so as a
probability term... than your... than your
DNA. So it's actually more condensed
information of what makes you... your
genetic code your genetic code. So that's
not really helping with the privacy.
M3: So it's a still... you can still
identify the single person,
AB: Yeah. Yeah.
M3: Okay thank you.
Herald: Yeah, that was, unfortunately, the
last question, because we are running out
of time. The next talk is waiting. Please
give again a warm applause to Adora Belle.
applause
music
subtitles created by c3subtitles.de
in the year 2017. Join, and help us!