Rik's Treehouse
>
Academic Aberrations
>
UBC ISCI 422: Models in Science
> Past Projects
Past Projects
Home
Past Projects
2007-08
Adrian Verster
2006-07
Jesse Popov
2005-06
Mary Ko
Vivian Pan
2004-05
Jen Batterink
Matt Ingham
Software
Grades (XLS)
Grades (web)
Administration
Related Links
Forum
Integrated Sciences
Course Calendar
Here are some select
Model Construction
projects from past years, kindly shared by former students.
2007-08
Adrian Verster
An Agent Based Model of the Action Potential
Abstract
The Hodgkin Huxley equations were the first to characterize the action potential. It was done by describing a sodium, a potassium and a leak current; they do not explicitly describe the discrete ion channels causing these currents. The number of ion channels can be directly regulated by the neuron, but since the Hodgkin Huxley model does not explicitly deal with the ion channels, it is unable to predict how this would affect the characteristics of the action potential. We created an agent based model of the action potential which simulated each ion channel individually to try and understand how changing the number of ion channels could affect the characteristics of the action potential. Here we show that the amplitude of the action potential increases, and the time until the maximum and minimum voltages decreases as the number of sodium and potassium ion channels increase. Since generating the action potential is the rate limiting step of current conduction in the axon, our data implies that increasing the number of ion channels should increase the speed at which the action potential is propagated down the nerve fibre, and that regulating the number of ion channels is critical in neurons involved in reflex arcs that must be optimized for speed.
Files
AdrianVerster-Report.pdf
175 KB
Report: An Agent Based Model of the Action Potential
by Adrian Verster
AdrianVerster-Model.nlogo
16.7 KB
Model: NetLogo Computer Simulation
by Adrian Verster
2006-07
Jesse Popov
Developing an Innovative Nanopharmaceutical for Cancer Treatment
Abstract
I have developed a numerical computer model to help develop a drug technology (nanopharmaceutical) that is intended for use in cancer patients. The formulation that we have developed in our laboratory includes the drug trastuzumab, which is used to treat breast cancer, and we have found that our nanopharmaceutical is much more efficacious than trastuzumab alone. In an effort to explain these effects, we have turned to the lipid raft concept, a revolutionary new theory on the way that cellular membranes function. My model encompasses several aspects of the way that our nanopharmaceutical would be expected to interact with lipid rafts in the cell membrane, and has provided results that are readily and directly testable in our laboratory. I intend to experimentally verify these results to strengthen the assertions of the model.
Files
JessePopov-Report.pdf
350 KB
Report: Developing an Innovative Nanopharmaceutical for Cancer Treatment
by Jesse Popov
JessePopov-Model.nlogo
56.3 KB
Model: NetLogo Computer Simulation
by Jesse Popov
2005-06
Mary Ko
Relating Calcium Influx Induced Dendritic Expansion to the Increased Conductance in Long-Term Potentiation Maintenance
Abstract
Long-term potentiation has been implicated in memory formation and learning in the hippocampus. While the presynaptic terminal is associated with the induction of LTP, the postsynaptic terminal is associated with the maintenance of it. This paper examines the possible role of calcium-induced postsynaptic restructuring in LTP by constructing a numerical model based on the characterized kinetic properties of the individual components. The model simulates the proposed pathway from calcium influx to AMPA receptor insertion and the subsequent increase in membrane conductance. As predicted by the hypothesis, the magnitude of conductance increase was found to be dependent on the peak calcium concentration. However, because the increases were only minimal, it was proposed other postsynaptic mechanisms also contribute to the increased conductance in addition to AMPA insertion.
Files
MaryKo-Report.pdf
83 KB
Report: Relating Calcium Influx Induced Dendritic Expansion to the Increased Conductance in Long-Term Potentiation Maintenance
by Mary Ko
MaryKo-Model.nlogo
22.1 KB
Model: NetLogo Computer Simulation
by Mary Ko
Vivian Pan
Making Sense of the Unknown: A Model for Increased Cognitive Load for Recognizing Ambiguous Words
Abstract
People can distinguish and categorize a wide range of phonological variations in speech into distinctive categories of words. However, as the speech sound becomes more degraded (with noise or with differential attention and speech production), it becomes harder to hear and takes longer for us to recognize (Andruski, Blumstein, & Burton, 1994, Hoff, 2001, Aydelotte & Bates, 2004, & Blumstein, 2004). This paper will propose a mathematical model of processes that contribute to the increased processing time for ambiguous sounds. Using Microsoft Excel and the construction of an artificial language, the model will use a probabilistic approach to describe sounds. This study will explore the frequency effect and the contextual (lexical) effects of sounds on word recognition. The number of steps taken to achieve word recognition is compared to quality of sound input. Results and predictions made by the model are found to be comparable to observations of reaction times over acoustic continua constructed with varying voice onset time. Suggestions of future improvements involving the Bayesian Probability Theory and Neural Networks are also suggested.
Files
VivianPan-Report.pdf
92 KB
Report: Making Sense of the Unknown: A Model for Increased Cognitive Load for Recognizing Ambiguous Words
by Vivian Pan
VivianPan-Model.xls
1.52 MB
Model: Excel Computer Simulation
by Vivian Pan
2004-05
Jen Batterink
A Model of Adelie Penguin Colony Establishment
Abstract
Adelie penguins are flightless aquatic birds that breed on gravel beaches along the Antarctic coastline. The ecosystem in Antarctica has a delicate balance due to the low number of species. Changing climatic conditions have led to variations in the ice cover, thus breeding sites of Adelie penguins are more dynamic now. New colony establishment is becoming more common as penguins emigrate from crowded colonies to the new breeding sites. It is important to understand colony formation as it will affect overall penguin population dynamics. A model was developed to further understand population patterns of an establishing colony and the number of immigrant penguins required. In order for a new colony to be self-sustaining, the immigrating penguins must produce enough chicks that survive to maturity to replace the adult penguins who perish. The various mortality and survival factors of adults and their young were integrated into a model on the Simile computer program. The model was able to predict that 44 was the minimum number of immigrating adult penguins required to establish a self-sustaining colony. However this number was determined to be highly dependent on fluctuations in mortality. As well, a pattern for population growth during a colony establishment period was found which included a ‘lag time’ and a ‘recovery period.’
Files
JenBatterink-Report.pdf
74.2 KB
Report: A Model of Adelie Penguin Colony Establishment
by Jennifer Batterink
JenBatterink-PenguinTrial4.sml
186 KB
Model:
Simile
Computer Simulation
by Jennifer Batterink
Matt Ingham
Influenza A Reassortment
Abstract
Influenza A is an RNA virus, and found in a wide range of strains and subtypes around the world. The various subtypes of influenza A viruses are identified by which versions of the hemagglutinin (HA) and neuraminidase (NA) proteins they possess. New subtypes can result from reassortment when two virus subtypes infect a single host. Pandemics of influenza A can be caused if such a subtype is highly pathogenic. Using sequence similarity and knowledge of existing subtypes as a basis for defining the characteristics of various types of HA and NA proteins, a computer model was designed in which there is infection of a population of humans by avian influenza subtypes and thus new subtypes are created through reassortment. Based on the predicted pathogenicity, virulence, and complementary action of the HA and NA proteins in a new subtype, its ability to spread amongst and kill the model population is simulated. Using this model, two experiments were conducted. The first involved the infection of the population by many new subtypes, from which insights into how different subtypes behave under the same conditions were derived. The second involved the infection of only one avian subtype. Pathogenicity, virulence and reassortment rates were all varied for the new subtypes resulting from reassortment. The results showed pathogenicity to be the dominant factor in determining the number of deaths due to a new subtype. Such a model can be used to predict the likelihood that new subtypes will cause a deadly pandemic, and thus could prove useful in directing efforts to create vaccines and the distribution of anti-viral drugs to high risk individuals.
Files
MattIngham-Report.pdf
45.2 KB
Report: Influenza A Reassortment
by Matthew Ingham
MattIngham-experiment1.nlogo
17.6 KB
Model:
NetLogo
Computer Simulation
by Matthew Ingham
MattIngham-experiment2.nlogo
17.8 KB
Model:
NetLogo
Computer Simulation
by Matthew Ingham
MattIngham-fluupdate.ppt
147 KB
Ancillary Material
by Matthew Ingham
MattIngham-HAblast.txt
2.45 KB
Ancillary Material
by Matthew Ingham
[Rik's Office Hours]
[Contact Rik]
Last updated: Sun Nov 2 2008, 5:49pm
