Team:Warwick/Modelling

From 2014.igem.org

(Difference between revisions)
Line 73: Line 73:
             <div id = "pageContent">
             <div id = "pageContent">
             <!-- THIS IS WHERE YOUR MAIN BODY GOES -->
             <!-- THIS IS WHERE YOUR MAIN BODY GOES -->
-
             <h1> MODELLING </h1> <br> <br>
+
             <h1> OVERVIEW </h1> <br> <br>
-
<p> Our modelling in this project has several aims: </p>
+
<p>Modelling is the most useful tool in a synthetic biologist’s toolbox. It allows us to experiment, without actually conducting experiments, something I’m sure every biologist wishes for. By use of computer programs and mathematics it is entirely possible to simulate a system in an artificial environment.</p>
 +
<p>But why would we want to model?</p>
 +
<p>Quite simply, it allows us to optimise our system under various circumstances and as a result could lead to potential cost savings in wetlab tests. For instance, what if you wanted to know how well GFP grows with respect to temperature. By modelling, we can save time and determine an answer rather quickly as opposed to waiting overnight for cultures to grow,etc.</p>
 +
<p>How does modelling fit in with our project?</p>
 +
<p>So we’ve discussed an overview, but I’m sure you all want to know how modelling fits into our project. We took three approaches with the modelling:</p>
<ul type="circle">
<ul type="circle">
-
<li>To find the amount of DPP-IV reduction reached when the system reaches equilibrium</li>
+
<li> We modelled individual experiments to directly relate results obtained from them and fit them to make our models more accurate enabling further predictions to be made.</li>
-
<li>To find a way to control the level of DPP-IV reduction</li>
+
<li> We modelled the whole system in its environment allowing us to explore the dynamics of replication.</li>
-
<li>To find the minimum number of RdRps, replicons, etc to be initially transfected into the cell, which are required to achieve a steady state for the system</li>
+
<li> Upon discovering that the leading method to treat Ebola is synthetic targeted-siRNA drugs, and the similarities that this bears with our system, we also took an epidemiological approach and examined the dynamics of widespread infections on a macro level as opposed to the cellular level previously explored.</li>
-
<li>To find out how long does it take for the system to reach equilibrium</li>
+
-
<li>To find out the level of reduction we need to treat diabetes</li>
+
-
<li>To find out how stable the system is (i.e. will the system only work in very specific situations, or in lots of different systems?)
+
</ul>
</ul>
-
<p> We are currently using Simbiology in Matlab and Copasi to model the system. We are currently adapting several different models, which come from research into HCV replicons, to our system. If our models can be made to fit our experiments well, we may extend our project to try and find a way to control the level of DPP-IV which is reduced. In addition modelling the system will allow it to be better optimised in the future, and optimum values for constants such as the strength of the ribosome binding sites, and the number of siRNAs produced by each degradation, so that the effect of our biobrick can be optimised. </p>
+
<p>Ultimately, the overall aim of our modelling efforts is to develop a model that enables us to take control of replication. This would allow anyone in the future who is exploring this method of information propagation to take our model, and with a few tweaks, be able to apply it to their system. This re-usability was at the core of our modelling ethos. We wanted to create something that wasn’t just specific to us but something that would help further iGEM teams and potentially academics.</p>
-
<p> We are currently using Simbiology in Matlab and Copasi to model the system. We are currently adapting several different models, which come from research into HCV replicons, to our system. If our models can be made to fit our experiments well, we may extend our project to try and find a way to control the level of DPP-IV which is reduced. In addition modelling the system will allow it to be better optimised in the future, and optimum values for constants such as the strength of the ribosome binding sites, and the number of siRNAs produced by each degradation, so that the effect of our biobrick can be optimised. </p>
+
<p>How did we go about modelling?</p>
-
 
+
<p>In order to model, we made use of the mathematicians and physicist on our team. Together they compiled some rather complex mathematics and constructed several types of models. We took both deterministic and stochastic approaches but also introduced time-delay into our systems, something that we noticed iGEM teams failed to incorporate in the past. Time delays do complicate our model, but it adds a greater depth to its realism because it is an unreasonable assumption to make that everything occurs instantaneously; RdRp has to bind to the RNA and then travel along it, this doesn’t occur straight away. We also used multi-dimensional PDE systems when exploring the epidemiology of Ebola.</p>
-
<p> Initially we determined that our system should reach some equilibrium after a certain amount of time. This is because firstly, HCV is a successful virus, so the replicons should not completely degrade away as time goes to infinity. Secondly, since there are only a finite amount of resources within the cell, the number of replicons in the system cannot keep increasing forever. This means either the number of replicons must tend towards a certain constant (constant with respect to time), or the number of replicons should tend towards oscillations. </p>
+
<p>Primarily, we used MATLAB as our software of choice, but also used Mathematica and Origin when fitting experimental data.
-
 
+
As we decided to explore modelling avenues previously unexplored by iGEM teams, we decided to create our own ToolBox of solvers using . This would enable future iGEM teams to develop more realistic models of their systems without fear of struggling how to solve them or obtain results. Again we wanted to make sure that we left a legacy with regards to modelling, enabling teams to become better equipped in this field.</p>
-
<p>
+
-
        \begin{eqnarray}
+
-
\label{system1}
+
-
\frac{dm}{dt} &amp;=&amp; \alpha_m - \beta_m m - k_s ms
+
-
\\ \label{system2}
+
-
\frac{ds}{dt} &amp;=&amp; \alpha_s - \beta_s s - p_s k_s ms - k_r sr
+
-
\\ \label{system3}
+
-
\frac{dr}{dt} &amp;=&amp; \alpha_r - \beta_r r - p_r k_r sr
+
-
\end{eqnarray}
+
-
</p>
+
-
</script>
+
</div>
</div>
<!--CONTENT END-->
<!--CONTENT END-->

Revision as of 01:26, 17 October 2014

OVERVIEW



Modelling is the most useful tool in a synthetic biologist’s toolbox. It allows us to experiment, without actually conducting experiments, something I’m sure every biologist wishes for. By use of computer programs and mathematics it is entirely possible to simulate a system in an artificial environment.

But why would we want to model?

Quite simply, it allows us to optimise our system under various circumstances and as a result could lead to potential cost savings in wetlab tests. For instance, what if you wanted to know how well GFP grows with respect to temperature. By modelling, we can save time and determine an answer rather quickly as opposed to waiting overnight for cultures to grow,etc.

How does modelling fit in with our project?

So we’ve discussed an overview, but I’m sure you all want to know how modelling fits into our project. We took three approaches with the modelling:

  • We modelled individual experiments to directly relate results obtained from them and fit them to make our models more accurate enabling further predictions to be made.
  • We modelled the whole system in its environment allowing us to explore the dynamics of replication.
  • Upon discovering that the leading method to treat Ebola is synthetic targeted-siRNA drugs, and the similarities that this bears with our system, we also took an epidemiological approach and examined the dynamics of widespread infections on a macro level as opposed to the cellular level previously explored.

Ultimately, the overall aim of our modelling efforts is to develop a model that enables us to take control of replication. This would allow anyone in the future who is exploring this method of information propagation to take our model, and with a few tweaks, be able to apply it to their system. This re-usability was at the core of our modelling ethos. We wanted to create something that wasn’t just specific to us but something that would help further iGEM teams and potentially academics.

How did we go about modelling?

In order to model, we made use of the mathematicians and physicist on our team. Together they compiled some rather complex mathematics and constructed several types of models. We took both deterministic and stochastic approaches but also introduced time-delay into our systems, something that we noticed iGEM teams failed to incorporate in the past. Time delays do complicate our model, but it adds a greater depth to its realism because it is an unreasonable assumption to make that everything occurs instantaneously; RdRp has to bind to the RNA and then travel along it, this doesn’t occur straight away. We also used multi-dimensional PDE systems when exploring the epidemiology of Ebola.

Primarily, we used MATLAB as our software of choice, but also used Mathematica and Origin when fitting experimental data. As we decided to explore modelling avenues previously unexplored by iGEM teams, we decided to create our own ToolBox of solvers using . This would enable future iGEM teams to develop more realistic models of their systems without fear of struggling how to solve them or obtain results. Again we wanted to make sure that we left a legacy with regards to modelling, enabling teams to become better equipped in this field.

Retrieved from "http://2014.igem.org/Team:Warwick/Modelling"