The ISBGroup Blog

In my work as a scientist, one of the possible tasks I can take upon
myself is to be a reviewer. This is important, but almost completely
non-valued; there is no pay, and no funding agency cares very much whether you do it or not. However, the review process is the most important place where scientific
discussions are taking place, because currently – and unfortunately –
this is virtually the only criteria for what constitutes a scientific finding
today: has it been published in a peer-reviewed journal? I therefore do
spend some time doing this. And I fight a lot of the time with papers
and authors who want to publish mathematical modelling works without any
comparison with experimental data. I strongly believe that such works
are not science, and that they should not be published. Today, I just
submitted such a response, and since the reply is written in a
completely non-specific manner to the paper in question – it could have
been written to any paper with the same problem – I also post it here.
My principle for the the next decade of my life, which I just entered,
is “going public, going deep”, and this publishing of this here on the blog, is a
part of me following that new principle.

Here is the review reply that I wrote:

“Thank you for your comments.

I do recognize the fact that you and others have published similar
papers in the past, where models have been developed and presented with
no comparison with data. There is nothing I can do about that. However,
that fact does not transform such works into science. Modern science is,
in my very firm opinion, the truth-seeking tool that was established by
Galileo, many hundreds of years ago: it builds on i) the mathematical
formalization of mechanistic hypotheses of the system that you study,
and then ii) usage of *data* to distinguish between those hypotheses.
The hypothesis that has the best ability at describing data – in the
first round estimation data, and in the second round independent
validation data, based on predictions and *then* experiments – is the
superior hypothesis. It is this formula for truth-seeking that
distinguished the science that started with Galileo, and the
church-driven epistemology that ruled science before him (note that the
prior Aristotelian science worldview also involved mathematics, and
data, but not in the same hypothesis-testing manner). If Galileo’s
formula is not followed, it is not science. That a paper has been
published in a scientific journal does not make that work science. It
does mean, however, that that work *should not* have been published. The
only exception to that principle exists within the field of
mathematics, which has other criteria for its judgements of a paper:
e.g. that what is presented should be i) previously non-proven and ii)
should hold true for a large family of equations/examples. Another type
of paper in mathematics can be that of a new method, e.g. for
optimization, that is proven to be superior to existing methods.

Unfortunately, this conception of what science is was lost in the field
of modelling of biological systems during a large part of the 20th
century. During this time, it was called mathematical biology,
complexity theory, etc. This was, to a large extent, rectified, during
the beginning of the 21st century, with the conception of systems
biology. However, unfortunately, much old-school data-free modelling is
still done. This has to stop! It is giving, and has been giving, the
field of modelling in biology a bad reputation, with the impression that
it has nothing to do with reality or biology – and rightly so, such
modelling has nothing to do with reality! At least not in any way that
has been demonstrated by science.

Two further clarifications and responses to your reply are in order:

i) You say that your model is based on data. That is true. The model structure is based on data. Your manuscript does
therefore function as a review of existing biology. But that is
something different than publishing an original research paper, with
novel results. That is something that is fundamentally different than
the kind of comparison between simulations of the *entire model
structure* and data that I am referring to above. It is a bit like
saying that the Ptolemaic worldview (with the sun in the middle) is
based on data because it includes the sun, the planets, and the earth;
which are observed in experiments. The question is not if they are
present. The question is which way of connecting them in relationship to
each other that is the correct one. To go beyond what can be said with
biology alone – i.e. to do mathematical modelling – requires that one
puts the structure together using competing hypotheses (e.g. one with
the sun in the middle, and one with the earth in the middle), and then
sees which of the two corresponding models that produces simulations
that best agrees with data (existing data and future data). That is how
science has functioned since Galileo, and that is how it should still
function today.

ii) You say that a model component
in your model – that has not been validated in any fashion whatsoever –
produces a prediction that a specific component is important; you then
also point to some papers that claim the same thing. That could, on
the surface, seem like a comparison with data. However, with the model
structure that you have put together – with the most well-known and most
often considered main players in the beta cell ethiology – you could
identify any component in your model as the most important one, and then
find many papers that claim that that component is the most important
one. That is, unfortunately, how biology is allowed to work today, with
many co-existing hypotheses, that are allowed to continue to co-exist,
where each lab focusing on one of the components is allowed to point to
limited results as to why their particular component is the most
important one, and without forcing anyone to challenge these claims with
respect to each other; without finding out what the big picture looks
like. That is where systems biology can and should come in and make a
difference: by putting up alternative hypotheses regarding what is the
most important component(s), and then letting data, systems-level data,
judge which of the hypotheses that is the most compelling one. This is
how systems biology has worked in many/most of the papers that are cited
in the review paper that I gave you. That way requires a model that
produces simulations (time-curves, typically), that agrees with
estimation data, and with validation data.

In summary, for me to
judge this or any paper as publishable, you need to produce (at least)
these two things: i) at least one curve, e.g. a simulation of a variable
as it progresses in time, that agrees with corresponding data; ii) a
prediction that is validated by another dataset, not used for estimating
the parameters in the model. In fact, apart from that you should also
demonstrate that your model is superior to other models, i.e. that it
can describe all data that one of the currently most important and
realistic models can, and then more data apart from that.

In
other words, there are many papers that are published for beta-cells,
including for their ethiology. These models can describe a lot of data,
in the above manner. Why not take one of those models, find a feature or
dataset that they cannot explain (there are many), and then go ahead
and improve the model to make it able to explain those data (while still
retaining the ability to explain all old data). If you then also show
that this is not due to overfitting w.r.t. all data, i.e. if you then
show that your new model also can describe some validation data, not
used for model fitting, then you will have contributed with an
improvement that follows the tradition of science. Then, and only then, I
will judge your – or any scientist’s – paper as publishable.

We have gotten money – and lot’s of it!

While on a 6h train ride between Malmö and Stockholm, I at last have time to share some of the good news that have come to us, one after the other, but that I haven’t had the time to share with you here until now. The good news is that we have had a series of successful grant applications, which mean that we now have reached a whole new level, in terms of money flows, and that our group will significantly grow in the next 2-3 months.

This new money
flow actually started about a year ago, when I got money for a new 2
year postdoc position from AstraZeneca, i.e. approx 3 MSEK (300 000
USD). In this postdoc, we will create models for a new type of
animal-free experiments, called organ-on-a-chip. In this technique, you
build a system of realistic 3D organs made up of human cells, which are
interconnected by an artificial blood flow. This is a supercool
possibility, which is ideal for both modelling, for replacing animal
experiments, and for understanding diseases.

This grant was then
followed by a new EU project, called PRECISE4Q, which was approved in
January this year. In this project, we will make use of our multi-level
mathematical models for diabetes and cardiovascular disease, to create a
new clinically useful tool. The basis of this tool will be quite
general, and applicable to a wide range of diseases, but the focus will
be on helping patients who have or who are at risk of suffering a
stroke. The total budget of this project is 60 MSEK (approx 6 million
USD), and of that approx 4 MSEK goes to my group (400 000 USD).

After that, during the spring, I used these already approved projects to
write applications about other projects, which could build upon my
already approved projects. And this too has now started to bear fruit.

First off was an application to SSF, the Swedish Foundation for
Strategic Research. They have awarded me and a researcher from
Karolinska Institutet in Stockholm (Roland Nilsson) a grant of 7.5 MSEK
in total, out of which I will receive 3.5 MSEK (~350 000 USD). The topic
of this project is to use my modelling to extract quantitative fluxes
for all of the metabolic reactions in a cell; this is possible because
of Roland’s unique experimental skills, where he tags metabolites
outside of the cell with C13 carbon molecules, and then uses mass
spectrometry to measure where these tagged carbon molecules ends up in
all in metabolites inside the cell. We will use these money to both
transform this tool from a big potential to something that is useful in
practice, and then also to apply the technique to understanding both
metabolic diseases and cancer on a new level – useful for both research,
drug development, and diagnosis.

The last major grant we got,
and just a few days ago, this Thursday, was from VR, the Swedish
Research Council. This high-prestige grant builds upon my previous
collaboration with AstraZeneca, and will allow us to spend 4.2 MSEK (420
000 USD) to further strengthen our collaboration with AstraZeneca. More
specifically, we will make use of the new and more complete models
developed within the PRECISE4Q EU network (mentioned above) to
understand how a brand new diabetes drug, dapagliflozin, works, and if,
how, and when it can be used to also treat cardiovascular diseases like
stroke, heart attack, and heart failure. This project will also allow us
to understand more about which patients that should have which
treatments, and more about the different mechanisms at play in the newly
sub-divided grouping of diabetes into 5 sub-types.

One very good
aspect of this new situation is that we are really well-prepared for it.
This is due, in part, to some very useful grants we have gotten from the Swedish Foundation
for Research without Animal Experiments. Using this money, we have been
able to train talented undergraduate students in real research projects,
which they have done in parallel to their M.Sc. studies, by awarding
them scholarships (several of the recent blog posts have been devoted to presenting new such students). These new, and previous old, such quite unique students are now ready and eager
to start as Ph.D. students, and they are much more well-trained than
normal applicants would be, which we could find in normal open
announcements. For this reason, we already now know that we will be able to
fill all of the new positions with really great people, and are therefore
looking much forward to working with for, at least, the next 4-5 years. However, all of the new positions will be announced in open competition, so if you are a great candidate, who wants to join our group, don’t hesitate to apply, or to contact me for discussions on joint collaborations or positions.

All in all, it is also a very great feeling to suddenly have such
much money now at our disposal. And also quite humbling. Now we need to
really demonstrate to our funders that we can convert these great money into the equally great
research we have described in our visionary applications. Into research
that will be useful for both other researchers and for those who want to
develop and use new and improved treatments – both treatments such as drugs, and
treatments such as yoga and meditation. It will be great fun to enter this
next step in our group’s development!

This
is a picture of our group during a recent group meeting. Not all were
there, but most of them were. Some of the people in the picture are
excellent M.Sc. students in the end of their studies, who we now can
offer 2 Ph.D. positions. Apart from that, we will
hire two new people, one new Ph.D. student, and one postdoc, but also those will probably be recruits building on previous collaborations and projects. To help as manage all of these new people, we are also very fortunate that Gunnar’s second-in-command in the group, Associate Professor Elin Nyman, now is due to come back after an almost 2.5 year long leave in Harvard,
Portland, and an almost 1 year maternity leave. For all of these
reasons, made possible thanks to the new money described in the post
above, our group will now take a step up to functioning on a new level –
one with more senior people, more people financed fully by the group itself, and with
long-term funding secured. Now we will be able to fully focus on all of
the great research projects we are working on, without worrying if the money will ebb out in the middle of the project!

There is a workshop upcoming about modelling of biological systems at Linköping university. It will take place Thursday-Friday this week, May 11-12, 2017, and is open to everybody who wants to attend: students, scientists, and the general public. We will contribute with one presentation, at Thursday 16-16.20. This presentation will consist of an overview of our modelling philosphy and of our various modelling projects. Apart from us there will be some 20-25 other groups who will present as well. Almost all of the presenters either still work, or have worked, at LiU, i.e. Linköping university.

More information about the workshop can be found here.

Last Friday, one of our Ph.D. students – Mikael Forsgren – nailed up his Ph.D. thesis on our local thesis tree. He thus followed in the footsteps of many many others, ranging back possibly even beyond Martin Luther, by thus publically announcing that he has something that he would like to publically discuss: the thoughts and results in his thesis.

The formal defense will be held at 13.15 in Eken, at HU, Linköping, Sweden. The opponent will be Stephen Sourbron, from Leeds, who is an expert on perfusion modelling in different organs. Mikael’s thesis is about modelling of not only perfusion, but also of intracellular uptake and release, concerning the contrast agent Primovist. He shows how the estimation of such uptake properties using mechanistic modelling can be used to obtain new patient-specific biomarkers, which can be obtained from MRI examinations, and which thus can move to eventually become a non-invasive replacement for today’s invasive liver biopsy examinations. The goal is that this should lead to a reduction in healthcare cost and individual suffering, and to an increase in accuracy and diagnostic power.

Welcome!

Late last year, we were featued in a news article in the local news paper: Östgöta correspondenten (often referred to simply as “Corren”). The news article can be seen in the picture below, and it deals with research developed to replace animal experiments, and argues that Linköping university probably is the leading university on this topic in Sweden: we e.g. have a clear 3R-policy, have strong researchers in the area, and both of the two most prestigeous prizes from the Stockholm-based “Swedish Fund for Research without Animal Testing” called “Nytänkaren” has gone to researchers who (at the time) were located in Linköping. The article then goes on to feature an interview with me, and on our type of research: how mathematical modelling can and is replacing animal experiments.

This is not the first time we are featured in media lately. In fact, there have probably been written some 10 news articles in different media on this aspect of our research following our award in late 2015, and they have all been very similar. Most often, the reactions have been very positive, and congratulory, showing appreciation and a new hope over these new possibilities. Similarly, many researchers who have heard my lectures have also been very enthusiastic, and a common reaction nowadays is something like “I had no idea about these recent breakthroughs; I now feel that I lag much behind in my methodology, since I don’t use modelling; what can I do about that?”. This time, however, was the first time these things were featured in Corren, and perhaps for this reason, many of my colleagues read it, and some of them (mostly researchers who themselves do a lot of animal testing) reacted negatively to what it says. I don’t know all that was said about it, but some of the reactions I heard indirectly said that “one shouldn’t write that replacements of animal experiments are possible, since this will lead to the public mandating that this happens generally”. One colleague wrote an email to me directly, stating some specific and somewhat similar concerns about the artice. To this colleague, I wrote a detailed reply, which then led to us writing some rather long exchanges back and forth. All in all, this has now resulted in about 20 pages of text back and forth, dealing with different aspects on this topic.

Since much of these concerns are concerns and thoughts that I have heard many times before, and which most often are based on mis-understandings of what (sound) modelling is about, or in an unawareness of what the most recent developments actually entail, our exchanges have resulted in a growing resolve on my side to do something about this unawareness. To give a little bit of perspective on that: in all previous appearances we have made in the public world, the initiative has always come from some other source; we got the initial award “Nytänkaren” without even applying for it, and in all following media appearances, we have just responded to invitations, to give presentations, to do interviews, etc. Now, however, because of these exchanges, I have realized that I do feel quite strongly about these things, and that I really am concerned about all the mis-conceptions that are floating around (perhaps especially in the biological research communities) concerning what modelling can and cannot do, not only but also in relationship to animal experiments, I am hereby opening a new category in our blog, called “3R and animal experiments”. In this category, I will feature my thoughts and news concerning this exciting topic, and it is intended for both colleagues and for the interested public. Then all exchanges will be public for all to see immediately.

In general, I also want to state that I think that these kind of debates is exactly what I think that science should be about: on topic, honest, and sincere discussions, where ideas can be compared, exchanged, improved, refined, and viewed for all to see.

To start this off, I am attaching the 20 pages of emails below (also including a powerpoint presentation with some figures). That text, however, is pretty long-winding, and in Swedish. Therefore, many of these points will be summarized and explained more simply/clearly in shorter following blog-posts and youtube-videos, which also most often (but not always) will be in English. So stay tuned for that!

Part 1, text from colleagues in italics

Part 2, my colleague’s response to my reply in italics (written then, as a continuous text, without specific replies to my points), upon which I replied in the same way as the first time, by breaking down the statements in small bits, and answering to them one-by-one.

Part 3, final part of the interchange (so far), same structure as above.

Part 4, powerpoint with the figures alluded to in the initial answer (Part 1, above).

A couple of days ago, Gunnar Cedersund held a presentation at the local event “Linköping vego”, which is a little festival on vegetarian-related topics, featuring around 300-500 attendees, 10-15 lecturers, and 10-15 exhibitioners. Gunnar’s lecture was in the “Blanche Lindgren” seminar series, which featured 4×40 min lectures on topics related to research without animal testing. The 4 lectures complemented each other: the first, by Karin Gabrielson Morton, featured an overview of the field as a whole, some historical examples of breakthroughs, and also an overview of the organizing body, the Swedish Fund for Research without Animal Experiments; the second presentation was the one by Gunnar (picture below), who talked about how we and others use mathematical models to reduce and replace animal testing; the third presentation was held by Anna Herland, who is the newest recipient of the same prize that we got the first edition of, “Nytänkaren”, and she works with organ-on-a-chip developments for the human brain; the final lecture was held by Kathrin Zeller who is a member of one of the flag-ship groups for the research fund, which involves an animal-free test for whether a new skin product is likely to lead to allergic reactions. Apart from spanning 4 interesting directions of biomedical research, there were also natural overlaps of interests between us: Anna’s organ-on-a-chip might allow us to test some of our hypotheses regarding how the brain’s fMRI signal is created on a cellular basis, and Kathrin’s omics-based tests for the immune system might complement our own exciting plans and projects on the same topic together with AstraZeneca.

Picture of me in the perhaps most central slide in the whole presentation. The slide illustrates with a simple thought example that a model does not have to be perfect to provide a valid replacement of test animals – it just has to be comparably good/bad as the corresponding animal test. Then, because computer simulations are so much faster and cheaper (among other things) than animal testing, the switch will happen automatically, especially in the industrial sector, which typically switches immediately, if there is money to save. There are already examples where this has happened.

At the event, we also made some initial plans for how to feature ourselves at this year’s edition of Almedalen, to which we this year might go jointly as a group. More on that to come! 🙂

The travel period is now just completed, and I am looking forward to writing a report and summary of some of the most important discoveries and news discovered at these meetings in a later blog. Our field – modelling of biological systems – is in a very intensive phase of expansion right now, with a wide variety of events and simultaneous developments going on. Apart from the 4 meetings I/we have attended in the last 3-4 weeks, there are several other important events that we will miss, that are happening now (FOSBE, Magdeburg, Oct 9-12, on Foundations of Systems biology in Engineering) and within the next few weeks (e.g. the first conf of the European Association of Systems Medicine, Berlin, Oct 26-28, Berlin, link).

Figure 1: Me giving a lecture, before recieving the award Nytänkaren from the Swedish Fund for Research without Animal Experiments.

For now, however, I just want to say that one of the developments that is especially interesting to me right now is that concerning modelling in drug and device certifications, and how modelling there can be used in a 3R context, i.e. to replace, reduce and refine usage of animal experiments. I have previously won the first edition of a newly instated prize on this topic – Nytänkaren, Fig 1 – and several of the workshops and conferences I have just attended have had important news on the topic. For instance, I learned that the Food and Drug Administration in the US just have finalized a new report on how to use modelling in device certifications (Fig 2). I also learned interesting details of a second report, due to be published in 2017, which together with the first report will consitute a first complete guide to how modelling should be used in biomedical device certifications. In practice, the same guide will be used as a proxy also for certifications of new drugs, since the principles behind a sound usage of modelling are quite general. In fact, these guides have drawn quite a lot from similar guides already developed and established by NASA, where simulations already are considered mainstream in the development and certification of new space products. In short, I am quite impressed by these guidelines, and think that they have identified more sound principles than are being used in much of today’s research. In other words, these are really important developments!Figure 2: The new report just issued by FDA. Note the date, it was published a few weeks ago!

As a follow-up to these developments, and to me recieving the 3R award Nytänkaren, I have been invited to give a keynote lecture at a workshop at Karolinska Institutet, which takes place today. This workshop is arranged by Swetox and has the overall title “Replace animal use and increase scientific impact”. My talk is entitled “When laying the puzzle instead of just generating new pieces, animal experiments become increasingly irrelevant”. In this talk, I will give a more detailed report on some of the developments above, of my own research on the topic, and about some of the most important showcases that already are existing in the field. If you cannot attend the meeting yourself, you can still check out much of the material: some slides on the FDA developments are available here and a recent overview of our own research including an already FDA-approved glucose simulation model is found here. Finally, a two-sentence summary of the main statement given in the title of my talk is as follows: “if you want to test hypotheses regarding human mechanisms on the systems level, and create a systems-level understanding for the human system, you need data from a single system: humans. Thus, when science switches its focus to this more important endeavor, instead of just generating new hypotheses and pieces of knowledge that never are forged together and tested in a systems-level context, then animal experiments will become increasingly irrelevant”.

Figure 3: First page of my keynote presentation today. It is one of the first times, perhaps even the first time, that I am giving a lecture denoted as keynote, so I am looking much forward to it!