Subscribe
Share
Share
Embed
A guest interview with Dr. Ida Grundberg, Chief Scientific Officer of Olink Proteomics, about what it was like in the earliest days of Olink Proteomics, and the journey from the first Nature Communications paper in 2014 until today, 1,000 publications later.
Proteomics in Proximity discusses the intersection of proteomics with genomics for drug target discovery, the application of proteomics to reveal disease biomarkers, and current trends in using proteomics to unlock biological mechanisms. Co-hosted by Olink's Dale Yuzuki, Cindy Lawley and Sarantis Chlamydas.
Welcome to the
Proteomics in Proximity podcast, where
your co hosts Dale Yuzuki, Cindy
Lawley and Sarantis Chlamydis from OLink
Proteomics talk about the intersection of
Proteomics with genomics for drug target
discovery to the application of
proteomics to reveal disease
biomarkers and current trends in using
proteomics to unlock
biological mechanisms. Here we have
your hosts, Dale, Cindy, and
Saranits hello everybody.
Uh, I'm happy to welcome you in another
episode from Proteomics in Proximity,
together with um, my cohost Dale and Cindy.
We are really happy
today to have with us
Ida Grunberg, the Chief Scientific Officer
of Olink. Really happy to discuss with
her about her career and uh,
ongoing projects and
ongoing outcomes in uh,
Olink. Welcome Ida.
Uh, thank you very much. Thanks for
having me. Great to meet you can join
us.
Thank you very much Ida, for joining.
Actually, I will start the question by
I would like to know a little bit more
about your background and
what was your status and
how do you see your transition from
academic to the industry? Can
you share a little bit more of your
experience? It would be really nice to hear.
Of course, I don't know how much time we
have, but I try to keep it
short. So I'm a
molecular biologist by training.
And then I joined
professor of
Landergren's group to start my
PhD. And at the time I
joined, he had actually just founded
Olink. So they were just the floor
above us. Uh,
so I was very
early introduced to OLink,
but uh, it was a great
group to spend a PhD
in. Very uh, inspiring and
creative. We were developing
all different molecular tools
targeting DNA,
RNA and protein, depending on the
research question. So I think that thanks to
my time there, I got
a great foundation,
um, and a very open
mind to the importance
of all the
omics. Ida,
I'm curious what years
roughly, I mean, not to date yourself,
but was it the early 2000s,
late 1990s? It was the
I joined Ulf's (Ulf Landegren's)
Lab in January 2006,
and they had founded Olink
in late
on board, I think in
was there between 2006 and
And then I focused actually more,
sorry to say, not on proteins, but on the
transcriptomic side. So I was working
more with Professor Mats Nilsson's group,
developing uh, technology for
transcript detection, more for point
mutation and genotyping
applications for colorectal
cancer and lung cancer
applications. So it was
actually I'm
sorry, was this using sequencing? No,
it was an in-situ-based application. Uh,
so we were using padlock probe
that had been invented in
Ulf's and Mat's group. So there
was a microscope,
... readout. So really
like padlock probes
similar to molecular
inversion. Exactly. Halo Genomics,
yeah, they were all in the family. So
again, we were almost
fantastic group to be very
privileged to have done my PhD
there. So it was during my
last year where actually the
technology I have been working on
developing that we licensed that together
with Olink. So that was my door
into the company.
So uh, just after I finished my
PhD, I continued at R&D,
at Olink and first focused on
continue commercializing that
technology that was later
sold off to another
company and I continued that
R&D, working more on the
technologies that...
we are focusing on now, more on the
proximity extension assay.
That was also around
the time when we first
launched the first panel.
So that a lot of
things were happening
and still keep in mind
that that was the early days. So I don't
know, maybe we're 25 people,
more than half of the company were at
R&D. But then of
course we had
commercialization going.
so then I
was recruited to the small
commercial team at the
time, um,
first
I had been in academia, so
for me it was like, wow, uh,
transition, should I go into
sales? But at that time
we had zero customers,
one product coming
out. So it was really, from the
beginning, a very scientific sell
working with our
collaborators,
close
friends, in
Uppsala. So I think we were three
people at the commercial team at the
time. So I didn't see
it as a scary
transition actually. But I
always think... Help me with. The date,
help me with the dates. Now is this
So still
very early days.
So the first
now we call it Target 96 oncology was
launched in
years uh, covering the
Nordic and European and then
2015 we
had decided that the biggest market is in
the US. We need to have someone
being based there. So I
was asked if I could go over
and start off our US
market. So that's when I did the move
to Boston.
That must have
been such an interesting transition
for you to
start something in the US.
Where... no
recognition of the technology, no friends,
no colleagues, maybe a few colleagues.
Yeah. How was that?
How was that?
No, you're
right, it was extremely
exciting, but also extremely scary,
but a...
very valuable experience.
But as you said,
I had zero colleagues
in the US. I moved to Boston by myself.
I had been there once before
for like a day, so I
knew basically no one.
And since we didn't
have anything in place, really,
we had to start from
scratch. But took day by day
and first settling in at an
incubator in Cambridge and
started connecting with the amazing
few customers we had at the time
and went from there.
What was one of the
most influential customers at
that time? Is there
someone you can point to or a couple of them
that really opened your eyes
to how the technology could
change things? Yeah. No, but for sure, I
mean, we were very lucky to, for
example, working with The Broad from really
the beginning. So the
incubator, that I was working
from was basically in the same block as
The Broad. So definitely there I spent some
time and trying to get in and set up some
seminars and so on.
...We had some there and
also at the MGH (Massachusetts General Hospital), we had some
early adopters there as well. So I tried to
stick very close to them.
so that was really the beginning. And
also why we decided to
set up in Boston,
Cambridge. That is the Mecca of
life science, but also where we
have the early
adopters. Yeah, not a
bad neighborhood to hang out
in. Right next door to The Broad,
just down the street from MGH, actually. I
remember like, walking down Broadway and I
felt my IQ was just raising.
That was a
good spot. Would you
remember any of the first projects that you
have been involved?
Let's say the first favorite
one. You mean
from the US market? Yes,
from the US market. One of the first that
you had close to your heart, actually.
For sure. Since we
basically no one knew about us, uh, we
had to have some strategies on
how are we going to actually
generate some evidence and
content. Because already back then, we
believed in
it's. Not enough just us showing what was the
power, the power of the technology, but
generating real data. So we
hooked up with,
some researchers at
Stanford then in San Diego. So with
Stanford, ah, we had
more of a wellness study where we
were comparing different
diets that,
the different group of
peoples were on. So we could
track the effect of inflammation
if you're on a high
fat or low carb diet. So that
was one of these
case studies that we early
did. Another one was done in San Diego
with Professor Doug Galasko where
he had a study on the
effect of antioxidants on
Alzheimer's disease.
So there we also did a study together with
him so we could generate great, data
that we could go out and have
joint road shows with.
I mean, from your work
in the in situ transcript-
omics now to
Alzheimer's and wellness. Right. What
a shift. A huge shift, right? What early
days to be looking at wellness, right? This
is such a hot topic. Right?
Isn't that amazing? Yeah, exactly. I don't
know if it was just luck, but those
early studies have still been
shown to be very modern and
been very
active in the spaces as well
now. So luck
maybe Doug and
Doug is. Still leading the way. Or the
Alzheimer's...
arena, for sure. Yeah, exactly. So
remarkable. But we have also, I think
all of you have heard about us. That could
be a nice story as well. So again,
since coming to the US. And
basically no one knew about
Olink and this little
small Swedish company.
but,
... for being such a
small country, there's a lot of
great researchers and
scientists in the States. So
we try to also work very closely with
them. We formed what we called the Swedish
Mafia so we could
all yeah.
...we didn't say it... To be
clear, this is not the "Swedish House
Mafia." That's a music
thing.
No, but seriously, so it was
great. So we have that at
Stanford, then in
Rockefeller, and teamed up with them. And
then they were also happy to
support and really put Sweden on the map
there. And... we have for sure, a
good reputation,
especially in protein history.
So we try to also leverage on that and
not just saying with this
unknown Swedish company, but we're
proud and we're here for a
reason, because we know this.
And as far as in the early
days, people were sending samples.
Right. You were getting customers sending
samples to Uppsala. But
then shortly after right. Didn't
you start actually building a laboratory
here? No, exactly. So I moved in 2015. Then,
just a few months
after, we started to have these
studies coming in. And some of them were
critical to have them run in the States.
So then, of course, we wanted an
office in Boston, but
we realized we also need a lab. So
that's when we started to look for
different places. So my role was very
broad, going around, looking at
different spaces. And
then we found this beautiful old
garage in Watertown,
literally a
garage. Show you a
picture.
It was going to
be rebuilt into
a small lab space.
So that's where we started. And... the
rest is history. And we always thought this
garage story was great and used to say that,
well, Apple started in the garage
too.
Wow.
This particular garage. Then you
installed a BioMark. Right? And
I understand an employee,
Dan Frederick of ours, who's an
application scientist, he told me a
little bit about that first
installation. What can you tell me about
it? I don't know any
details. It was two
or one. Uh,
we did, but the space was so
tight, so
it was funny. We grew
out of that mobile garage
in no time, and
especially when we were going to
have two BioMarKs. And I can't
remember I think we even had three there.
And these are big
pieces. Very big. In the end, we
were having meetings in the
cars outside
because we couldn't fit in there.
The cars
"...meet me in my
car!"
We're doing a podcast.
Right.
That's an interesting.
Dan told
the story of how it grew
so quickly. And then, of course,
when the opportunity came for him
to work for Olink, it was a very
easy decision for him to make,
just witnessing
firsthand that kind of growth.
And, these customers
then you had to hire, I guess you then
hired people to
run the laboratory assays,
that kind of thing you were central to that. Yeah,
for sure. So one of the first employees was
Jen, who is still working with us.
She's great. So she's been there from the
beginning and many others as well. So
many times we get
nostalgic and remembers the
garage.
Yeah. And I think also, together with
all that growth, the
need of people to use proteomics, right.
You see this change of people of thinking
from one publication to thousand
publications, right? How do you see
this? How do you see this, actually? How was
your feeling for the first publication when
it came out? Can you describe? It
was that
huge. We used to have like a
bell, so we're walking in the corridors
like, wow,
it was big. It was
big. And especially, I
mean, looking back, the first paper
that was in Nature Communications,
our close collaborator, Professor Ulf
Gyllensten, it was a
great paper. I mean,
looking back, it was
only one panel of 92
proteins, but a big study.
So fascinating outcome
that really paved the way
to where we are today.
So, yeah.
For those who may
not be familiar, right, this paper
did over a thousand
Swedish individuals. So they had
genotyping data and then they looked at
to protein. Then they had health
outcome data and then they had all the
clinical data. So they can talk about
the influence of genetics upon
the biomarkers,
genetics upon the outcomes,
and then the other
lifestyle and clinical factors. I
mean, reading it now, I mean, it was
a 2014 paper and it was like,
wow. Exactly
the
fascinating chord of it. So it's
a population study from the very
far north of Sweden,
Karesuando, where it's like
freezing year round.
And they had been following these
thousand individuals for a long time
and as you said, had all the information
on their lifestyle and
diets, but they wanted to add
proteomics to it. And then,
I mean, still less than hundred proteins,
but uh, they could gain so
much data and insight from,
uh, these results
quickly. Their conclusion, or
the take home from the study was that they
could really see the effect of
non-disease factors. So this was
a healthy population, but
still see that. I think more than
varying. All the variance
were actually coming from these
genetic and lifestyle factors.
So like age and blood pressure
and weight and smoking
habits, all of that had an enormous
effect... on proteins.
And they mapped all of
that. It was just
demonstrating that value of
capturing real time biology in
the context of the genetics.
It's a phenomenal paper. Yeah,
for sure. And
also it was a very
broad paper that they could also follow
up. They called it like a publication
generator, the data that they get out
because then they could also follow up and
see. So what should we do
with all this information? Is there any way
that we can associate these
results to more actionable,
diagnostical, clinical,
biomarkers? In the
follow up paper, they proposed... a
model to adjust for
these... variables
and that
algorithm that they proposed back then
is very relevant
today. That they
find really select
the most robust biomarkers that
are not
varying because of these non-disease
factors. So a great start
of the thousand papers.
It sort of reminds me a little bit
of I'm sorry to make the
parallel to the sequencing space but that's
sort of sequence once query
often, right? You have such a broad
look at proteins that you can deep dive
into those data for
multiple different purposes across that
population. Just at that time that was very
high-plex, we call that midplex now.
But
that's a beautiful
... point... about
the data set.
And what was also really interesting
for me was this idea of clinical cut off.
Exactly. In that you have sort
of individuals, right, with
certain levels of certain
biomarkers and if you have a certain
one or a handful that
spike up or drop
off, that means something is
happening, right, health wise for that
individual. So here we're talking about
personalized medicine based
upon the individual's
protein profile at a given time
and their genetic background and of
course their lifestyle and the different
things that they do. I was
thinking "this is personalized medicine, this
is the future", right? It's in
reference to the
population that you're able to compare it
to. And today we're
seeing such diverse populations
being characterized with proteomics.
So it's an exciting time
there as well, right? But we
started somewhere and that ability to
be personalized and translate to
the impact genetics is
having on something that we might actually
take to the clinic.
Maybe not Olink, but
those customers is
exciting.
Mhm, this particular
population in the far north of Sweden is
still being followed then today? I
don't think they have additional
samples taken
as far as I know. But they
continued studying that and then
also since that was the first
more epidemiology study we did, they
also were part of a
European
network with other concerned populations
that followed. So that was really the
start uh, of it. And
also with all these lifestyle factors
they had many more papers
coming where they were looking into
the
specific factors affecting like the
Swedish tobacco with the snuff
for example. That they saw
association to some biomarkers
and they got some media
attention for also using that to
look at aging and biological versus
chronological aging and
how diets
like fish and coffee can
affect or reduce your
aging and so on. So...
it was fascinating.
that's great
news for us. Coffee drinkers for
sure. Yeah, exactly.
Well I think on the reference to snuff for
those not familiar with it. It's
really popular in Sweden, I guess most
of Europe, but I think it hasn't hit
the US yet. Is it popular in
Greece?
It's kind of a thing you put in your
gum. And is it nicotine? Yes,
for sure.
No, it's actually
not even all of Nordic, but
some of the Nordic countries, but yes,
that's why also this type of research
was supported. I would say it's very
popular in some areas of the US as well.
For sure. Okay,
so you've seen it here.
Absolutely. This cohort is like
it was really well conserved. An isolated
cohort. Right. They've seen the real
effects and the real association with the
biomarkers without any so much external air
pollution or other factor they may
influence with that. That was a beautiful
place, actually. It was a beautiful, amazing
yes, exactly.
It's hard to capture those
environmental variables right. But that
the protein levels could
exceed the challenges with
capturing some environmental
variables. I think that's a
message that I find we have to
convey to a lot of geneticists who are
used to the signals being weaker and harder
to see,
... and having the power to detect
some of these polygenic
signals... to disease
can be more challenging in genetic
studies that are solely
genetics and disease. But I think bringing
in proteins can allow for
magnification of that
ability to see that
association. Mhm yeah. So then
back to that. This was the first
publication, I don't think we could have
asked for a better start
because it really first, I mean,
demonstrated the robustness of the
technology that was very important being new
into the market, but also the
power, as Cindy said,
combining high quality
proteomics with genetics and this
epidemiological information.
And now 1000
publications later, I
understand the same group had a very
interesting recent Olink
publication as well. What can you
share about that? Yeah, I mean they
have since the beginning. So they started
more with population,
health studies and
then part of one of the
course they did, they saw some
interesting signatures for gynecological
cancers. So then they
continued
drilling down that path
and then saw early an
interesting signature,
for patient statification of
ovarian cancer. So
we worked with them around that
and together
with this group that we also
begin our journey with
our focus panel with
custom development. So
we
helped out with developing that protocol
with the input from FDA and that
whole story.
I was just going to say I also love
that Stefan Enroth, who was the
first author on that early paper,
is now the PI (principal investigator), the last
author in this most recent
paper.
We all evolve. Yeah.
No exactly. So they have continued I
mean, we developed
a focused signature
panel for ovarian cancer where they could
see that they were superior to what
is actually used today in
diagnostics, in the US.
But they have continued trying to
polish on that and hoping to
get it to something
clinically,useful. And also looked
into other matrices. So
this was first based on
plasma, but also looking
into alternative
matrices that can be, sampled
in a home environment with
filter papers. And
now, using our (Olink) Explore, the
broadest
.ibrary. So that was a
paper that came out, just was it last week
or two weeks ago, where they now
have continued to
really polish on this
signature. So great
work. And for background,
ovarian cancer,
third major fatality
rate, not
very good biomarkers at all. CA
ages, it's just not that
great. You mentioned
patient stratification. So this
particular Focus and what you're
talking about is a custom product of
they applied a handful
of markers. Was that to stratify
patients for treatment? So
the first question
was more for patient stratification
because many
women are having surgery
without needing it, basically. So that
was the first where they wanted to
stratify women
with benign cysts from ovarian
cancer, Like a
blood test to stratify those. So
that's where they started, but now
going into more diagnostic
applications for basically early
diagnostic of to
identify those early
stages. Yes.
So is this also for recurrence monitoring?
Yes.
So that's a huge application
right. Where women will
have a cyst
removed, like you mentioned, but they don't
know when they need
to go ahead. And there's really
no real good way to measure
when it's coming back. And so
it's like living underneath this constant
fear of
recurring, right, ovarian cancer,
which is a huge medical
burden, for sure. And especially if we can
come to the point where we could have for
more careful monitoring of
those in high risk.
Yeah,
I love that it's in women's health as
well, which we talk about being
an area that could
use some more funding. Right?
Yeah.
And
you mentioned then about Explorer and
perhaps alternative matrices, that kind of
thing. Right. I mean that has must
been... from
that start with the early
work with 92 Plex,
assays now to
time. I'm curious, you mentioned
alternative matrices. What were some
of the more unusual things you've
seen in terms of in all these
years ways people are measuring proteins?
Yeah, I think
we have covered the whole
human body, really.
So never
stop being surprised with like and why
do you want to run ground
teas? Okay, so that's one
example that was surprising to
hear, but teeth. Yes,
exactly. Ah, but of course,
tears, saliva, urine, those
are quite common CSF, of
course, but then
different types of biopsies
also extremely important, of course. fine-needle
biopsies,
interstitial fluids,
synovial fluids,
blister fluids from
... burn
victims. Wow.
Interesting.
Yeah, we have
covered, I think, everything.
Intestinal juice was
interesting when we got
there was some device that
collected intestinal juice. So
we ran it, and it worked.
How interesting. So, I think
a lot of these samples, right,
where you don't even bother to
quantitate how much protein is in the
sample because you consume a lot of it.
Right. And people
don't know beforehand if it'll
work. However, we
have assurance that we use
very small volumes. We were
able to be very sensitive to pick up
very low sort of
amounts of protein available.
Right? People just say,
well, we don't know, therefore, we just
need to go ahead and try it. And many times
it works. Right. I mean, we have had
applications where they said, we collected
these very precious samples years
ago, but we couldn't use it for anything.
And now we have the opportunity and we could
deliver extremely
valuable data. So all of these fine
needle biopsies or micro
vesicles, tape
strips, exosomes,
cells. Yeah, exactly.
So, huge
impact. Yeah. These antibody
hooks, right. They're hooking it out
of this solution in such a
low volume, it's, ah, pretty
exciting. And then
we haven't talked about the different
species that we have done as
well. That's the whole
'zoo' that we have done as well.
So tell
us.
Interesting.
Just to be
clear, our protein
targets are
optimized or developed in R & D
to, human with
all but our mouse mouse panel. Right?
We certainly have a mouse panel. And so when
someone comes in and wants to understand
what's going on with a cow, for example,
what do you tell them
and what are the most
extreme examples that you've seen?
Yeah, I mean, what we tell them, of course,
exactly what you said, that we
don't have any cow
specific panels,
but it's quite close
homology for many of these
species to humans, and they don't have
any other better options,
so we give
them more data than they could get
from anywhere else.
So, of course,
rodents, ...
nonhuman primates, dogs,
cows,
fish,
horse. Wow.
So actually, they're able to
measure proteins out of fish, even
though we know that our
antibodies are generated against
antigens from humans, but
still we're able to actually
conclusively identify
that the
proteins measured. In a fish sample,
I think the detectability
in fish was low.
So it's
not anything we recommend
normally. Yeah.
With some
underlying information that you can get from
for specific proteins, you can get
some interesting results out
of it. It reminds me,
again, bringing back into the genetics,
is we worked on a whale
project, and there are no
SNP chips for whales. Right.
So I think we use
the same HapMap
from Enroth et al. in 2014,
to see how many SNPs we could find
homology... across species.
And it was actually pretty
useful. So, yeah,
interesting.
It's coming back a
little bit populational like the generation
of big data. I
know now that there are some
projects running at
Olink in regards of
visualizing and integrating
data with Olink Insight. Could you
give us a little bit of
feedback on that? Would you give us some
comments on that just to get to know the
more details?
Do you refer to the big
publication?
Yes. Olink Insight is a very,
very exciting new initiative that
we have. So it's a
digital knowledge platform
for proteomics that we have built
for our customers and
community. And there is when
looking at in the digital
space, there's nothing of this kind on the
market today.
How we started that, we thought that
since we see us as the
leaders in proteomic assays, so we also
want to be pioneers in the digital
space and drive the development
for more modern innovative
tools to help.
The proteome is very
complex. The data analysis can be
extremely challenging. If we
can, with our insights and
experience, try to
support the community to
faster come to
conclusions and actionable
results. By
developing tools and
connect different
public databases
and different annotation tools,
we can help to support
them throughout basically the
journey. So I'm very, very
excited about Olink Insight. So
the difference we're going to have or we
have different apps to
support the analytics
side as well as open
data sets that you can use for
validation or confirmation of a
disease signature or
compare with a
healthy proteome, and
also share data
stories where it's like best
practice analytics,
ah, for different
applications. So it's a
fantastic platform.
Yeah. One of the best features within
that, that I've struggled with since I
joined Olink is that conversion
from gene name, where you
have many different gene
names for a single gene,
converting that to the
protein that is coded for by that
gene. Right. Because in the genetics
base, so many people work with genes, so
there can be a dash
in there or no dash in there, or there
can be something that looks like a date. So
you throw that into Excel and it's a real
pain to look up DEC1. Right?
But to be able to pull that out of
Insight, it's such a simple thing,
but it's got
a nice functionality,
that I think
is just a microcosm
of the bringing together of the genetics,
and the proteomics. Just a nice
example of how thought out it was in
the needs.
And for those interested in trying
it out, it's free to
use it's at
insight.olink.com,
One of the other interesting features
that relates to the 1000
publications is the ability to
search publications by
Biomarker, is that correct Ida?
Correct.
So this is basically,
why wouldn't I use
PubMed versus
this particular feature within Insight?
No, exactly. And I think that's one
of the features at Insight that you can go
to PubMed, but of course, if
you want to see the thousand publications in
different categories, that's
the place to go. But also in.
general, it has very smart, smart search
tools, but it's public information.
But we have tried to make
it in an even more user
friendly interface,
the same as with the
Pathway Explorer that we call it,
that is based on the amazing Reactome
database. But it also
has some additional layers to it
where you can
insert a list of your top
markers and really see the underlying
biological pathways and
mechanisms. So it's a,
fantastic tool to go there.
You can download
the images as well. Right. Once you got
that publication set
up. When I was playing with
it, I was shocked when
I clicked on a single
node and the amount of
information that I can find
about that node in
really well written prose with
references and everything I wanted to
know almost in like three
or four paragraphs. It
was remarkable work.
A lot of work, obviously, has gone
into it, and remarkable
resource.
Well, thank you, Ida, for
joining us today. I really
enjoyed this conversation. Are
there any last words you like to share with
our audience? No, I think this is just
the beginning. I think still 1000
is an amazing
milestone, but, I think
already this year it's been
I can be invited again when we reach
be there you go. Absolutely.
You are
always welcome. Always.
A big shout out to
all of our both the early adopters
that's been with us since the
beginning and supported us.
... we wouldn't be anywhere without
them. So those are the most important
players here. Absolutely.
For sure. All right, until
next time. Thank you for
joining us. Thank you. That's great.
Bye. All right.
Thank you for listening to the
Proteomics in Proximity podcast brought to
you by OLink Proteomics. To
contact the host or for further
information, simply email
info@olink.com.