The ISBGroup Blog

On Thursday and Friday, the 6th Swedish Meeting on Mathematics in Biology takes place in Linköping.

We will be there and do a small presentation of our work. Hope to see you there!

Karin Lundengård just did her half-time, and it’s on youtube!

“My research project is on the activity of the human brain. The activity can be measured by functional Magnetic Resonance Imaging, which measures the Blood Oxygen Level Dependent (BOLD) response. As the name states, it is the oxygen levels being measured, but they are connected to the electrical signaling of the neurons via the neurovascular coupling. There are several hypothesis about the underlying mechanisms of the neurovascular coupling and here we have used mechanistic dynamic models to investigate two of them. So go have a look at the video to see how it went!”

Hopefully the article will be submitted for press before Christmas.

Karin Lundengård
PhD student at IMH and CMIV

Can anything be nicer than combining work with a trip to
Norway? – Probably, the pizza was as expensive as the myths foretold.

Joking aside, during the field trip to the conference
Virtual Physiological Human in Trondheim, a summit to discuss multi-level
models representing human physiology, the ISB group once again got a chance to
display its work. With the help of several presentations, both oral and
posters, new contacts were made and there was much rejoicing!

As a student visiting such an event for the first time, one
cannot help but feel a bit lost among the field leaders and other experts. I
can also admit that I was pretty nervous about my own poster presentation, but I
started to relax some time into the poster session.

To sum up the trip: Making new connections and getting out
our research is enjoyable, eating pizza in Norway, not so much.

Right before christmas, Linnea Bergenholm presented her master thesis Modeling as a Tool to Support Self-Management of Type 1 Diabetes. This thesis work was done in our group, collaborating with the companies Linkura and Wolfram Mathcore, aiming to develop and investigate dosing tools for diabetics.

You can download the thesis here

Abstract:

Type
1 diabetes (T1D) is an auto-immune disease characterized by
insulin-deficiency. Insulin is a metabolic hormone that is involved in
lowering blood glucose (BG) levels in order to control BG level to a
tight range. In T1D this glycemic control is lost, causing chronic
hyperglycemia (excess glucose in blood stream). Chronic hyperglycemia
damages vital tissues. Therefore, glycemic control must be restored.

A
common therapy for restoring glycemic control is intensive insulin
therapy, where the missing insulin is replaced with regular insulin
injections. When dosing this compensatory insulin many factors that
affect glucose metabolism must be considered. Linkura is a company that
has developed tools for monitoring the most important factors, which are
meals and exercise. In the Linkura meal and exercise tools, the
nutrition content in meals and the calorie consumption during exercise
are estimated. Another tool designed to aid control of BG is the bolus
calculator. Bolus calculators use input of BG level, carbohydrate
intake, and insulin history to estimate insulin need. The accuracy of
these insulin bolus calculations suffer from two problems. First, errors
occur when users inaccurately estimate the carbohydrate content in
meals. Second, exercise is not included in bolus calculations. To reduce
these problems, it was suggested that the Linkura web tools could be
utilized in combination with a bolus calculator.

For this purpose,
a bolus calculator was developed. The bolus calculator was based on
existing models that utilize clinical parameters to relate changes in BG
levels to meals, insulin, and exercise stimulations. The bolus
calculator was evaluated using data collected from Linkura’s web tools.
The collected data showed some inconsistencies which cannot be explained
by any model. The performance of the bolus calculator in predicting BG
levels using general equations to derive the clinical parameters was
inadequate. Performance was increased by adopting an update-algorithm
where the clinical parameters were updated daily using previous data.
Still, better model performance is prefered for use in a bolus
calculator.

The results show potential in developing bolus
calculator tools combined with the Linkura tools. For such bolus
calculator, further evaluation on modeling long-term exercise and
additional safety features minimizing risk of hypoglycemia are required.

Today Zaheer Ali, who has been working in our group since spring 2012, presented and defended his work.

The title was Mathematical Modeling of Electrophysiological Data of Facilitation in Layer 6 Pyramidal Cells (abstract below).

Abstract:

Facilitation is a distinctive activation pattern shown by pyramidal
cells in the layer 6 of the cerebral cortex upon rapid and repeated stimulations.
This behavior leads to the generation of tonic-clonic seizures which is a
defining feature of epilepsy. Patch-clamp recording techniques performed on
these cells show facilitation in different experiments. A mathematical modeling
approach is used in order to find the possible mechanism of facilitation
depending on different calcium domains in the presynaptic neurons. Four
different versions of the model have been constructed using ODEs. First version
of the model is a simple model of three compartments in the presynaptic neuron
with an influx of calcium at the triggering calcium compartment, the buffer
compartment and the stable internal domain of calcium can describe the data of
facilitation. Second version of the model has the calcium influx through a
micro-domain compartment near the cell membrane because of physical distance
between ion channel and the vesicle, in addition to the three compartments. The
third version of the model has a limited calcium buffer without micro-domain
calcium compartment which provides saturation to the buffer in this model and
the final version of the model is similar to third version of the model with
the addition of a micro-domain calcium influx. All models can describe the
experimental data sets of facilitation separately fitted to the model. In
addition, the third version of the model can describe all the experimental data
at the same time qualitatively.

We are now in the middle of preparing for (hopefully) four workshops to take place here in Linköping in the period of June 17-23.

One of the reasons for these workshops is that we want to make use of our newly recruited guest professors Hiroaki Kitano and Jens Timmer, by having some of the most important international players come here and discuss some vital issues with us. Kitano is one of the world-leading players in model construction (e.g. by being the founder of CellDesigner, the most widely used software in the field), and he has also recently moved into hierarchical modelling, which is the topic of one of the workshops. Similarly, Jens Timmer heads one of the leading groups of sound data-analysis approaches to systems biology modelling, including the handling of uncertainty in data; this is the topic of another workshop. The third workshop comes from a network that we are forming on multi-level hierarchical modelling of type 2 diabetes, and the final workshop is meant for us in the Linköping Centre for Systems Biology.

More information on all these workshops can be found here

So, welcome to Linköping in June! 🙂

Time is passing by, and several talks are on their way. Actually, today the first of them happened, via Anita Öst’s Ph.D. thesis defence. She defended her thesis “Lipid metabolism and insulin Signalling in Adipocytes – enhanced autophagy in type 2 diabetes”, with Anna Krok from Karolinska Institutet as opponent.

Next week, Gunnar Cedersund will present our work in general, and the ideas concerning unique estimation in unidentifiable models in particular, at ISY’s seminar-series on diagnosis. The details for this is as follows:
When: October 13, 10.15-12
Where: Systemet, ISY, Linköping University, Sweden
What: Unique identification in unidentifiable models, its development for systems biology, and its possible applications to diagnosis
For whom: All that are interested, it is centered around discussions, and very informal

Also on Friday in the next week, October 16, Gunnar will give a presentation of our work, again with a focus on the theoretical methods, at a conference in Hamburg. A link to this conference is found here. We recently found out that Hamburg has already established links with Linköping, in terms of possible exchanges between students, so it will be exciting to see if we can conribute to making good use of these possibilities.

Tomorrow (Wednesday, October 7th) Peter Nyberg presents his master thesis work “Evaluation of two Methods for Identifiability Testing”. A joint supervised project of our group, Division of Automatic Control at the Department of Electrical Engineering, and Mathcore. The presentation will take place at campus Valla, Glashuset (Building B, entrance 25) at 10:15.

Abstract:
This thesis concerns the identifiability issue; which, if any, parameters can be
deduced from the input and output behavior of the model? The two types of
identifiability concepts, a priori and practical, will be addressed and explained.
Two methods for identifiability testing are evaluated and the result shows that
the two methods work well if they are combined. The first method is for a priori
identifiability analysis and it can determine the a priori identifiability of a system
in polynomial time. The result from the method is probabilistic with a high
probability of correct answer. The other method takes the simulation approach to
determine whether the model is practically identifiable. Non-identifiable parameters
manifest itself as a functional relationship between the parameters and the
method uses transformations of the parameter estimates to conclude if the parameters
are linked. The two methods are verified on models with known identifiability
and then tested on some examples from system biology. Although the output from
one of the methods is cumbersome to interpret, the results show that the number
of parameters that can be determined in practice (practical identifiability) are
far fewer then the ones that can be determined in theory (a priori identifiability).
The reason for this is the lack of quality, noise and lack of excitation, of the
measurements.