Difference between revisions of "Team:ETH Zurich/Model/In Vivo"

Line 22: Line 22:
 
<section>
 
<section>
 
     <!--<h1>Why</h1>-->
 
     <!--<h1>Why</h1>-->
−
     <p>The COMSOL model helped us to extend our MATLAB model by including Diffusion physics for AHL and Azurin.
+
     <p>The COMSOL model helped us to extend our MATLAB model in the following ways:
 
     <ul>  
 
     <ul>  
−
         <li>Diffusion physics added to extend the ODEs used <span style="color:red;">earlier</span> into PDEs<br>
+
         <li>Diffusion physics of AHL was included to extend the ODEs into PDEs<br>          
−
          <ul>
+
         <p class="description">Our <span style="color:red;">MATLAB model</span> does not take into account diffusion of AHL, since in the viable lab experiments (in test-tubes), the only sink for AHL is its degradation. However, in a tumor as explained in <span style="color:red;">reasoning</span>, diffusion has much more significant contribution than degradation. Thus our model helped us gauge and verify the behavior of our tumor-sensing circuit in more real-life conditions pertaining to the intended application context. Using the results obtained from our simulations, we could check the behavior of the AND gate functioning in different conditions of d<sub>cell</sub> and lactate.</p>
−
              <li>AHL diffusion as it is produced<br>
+
−
              <p class="description"></p>                             
+
−
              </li>
+
−
              <li>Temperature controlled trigger of Azurin diffusion to simulate effect of lysis<br>
+
−
              <p class="description"></p>
+
−
              </li>
+
−
          </ul>
+
−
         <p class="description">The ODEs as formulated in <span style="color:red;">linkToPage</span> were extended to include diffusion in case of AHL and Azurin. Diffusion coefficients and other parameters are given in <span style="color:red;">linkToParameterSpace</span>. To simulate effect of lysis, Azurin production is stopped and Azurin diffusion is started using a trigger controlled by a Temperature step function which simulates the increase in the local temperature due to FUS. Protein E dynamics have not been used for the lysis, as they were not necessary.</p>
+
 
         </li>  
 
         </li>  
−
         <li>Different bacterial colonization patterns <br>
+
         <li>Diffusion physics of Azurin was included to simulate the effect of lysis<br>          
−
          <ul>
+
         <p class="description">To simulate the effect of lysis, our COMSOL model stops the production of Azurin and starts its diffusion when temperature reaches 42&deg;C. This simulates the effect of increase in temperature with FUS to cause cell lysis. Using data obtained from such a simulation, we could also find the temporal-maximum concentrations of Azurin at each point in the tumor, effectively helping us to estimate the killing area. </p>
−
              <li>Homogeneous distribution in a Single spherical shell layer in tumour <br>
+
−
              <p class="description"></p>                             
+
−
              </li>
+
−
              <li>Heterogeneous distribution in a Single spherical shell layer in tumour<br>
+
−
              <p class="description"></p>
+
−
              </li>
+
−
              <li>Heterogeneous distribution in Double spherical shell layer in tumour<br>
+
−
              <p class="description"></p>
+
−
              </li>
+
−
              <li>Homogeneous distribution throughout Healthy tissue<br>
+
−
              <p class="description"></p>
+
−
              </li>
+
−
          </ul>
+
−
         <p class="description">As explained in <span style="color:red;">linkToModelPage</span> bacteria are shown to colonize in a spherical shell layer. We also tried a few other colonization patterns to show our system works as expected in a tumour while stays dormant in healthy tissue.</p>                            
+
 
         </li>
 
         </li>
 
     </ul>
 
     </ul>
 +
 +
      <p class="description">Using our model, we also tried a few other colonization patterns to show our system works as expected inside a tumor while stays dormant in healthy tissue. We simulated the following patterns:</p>     
 +
      <ul>
 +
        <li>Homogeneous distribution in a Single spherical shell layer in tumour <br>
 +
        <p class="description"></p>                             
 +
        </li>
 +
        <li>Heterogeneous distribution in a Single spherical shell layer in tumour<br>
 +
        <p class="description"></p>
 +
        </li>
 +
        <li>Heterogeneous distribution in Double spherical shell layer in tumour<br>
 +
        <p class="description"></p>
 +
        </li>
 +
        <li>Homogeneous distribution throughout Healthy tissue<br>
 +
        <p class="description"></p>
 +
        </li>
 +
      </ul>                       
 
</section>
 
</section>
  

Revision as of 21:26, 30 October 2017

Behavior of CATE inside Tumor

FIXME

We developed a model to gauge the behavior of our sensing circuit in the real life conditions of solid tumor colonization. Since it was not practically feasible to conduct experiments of bacterial colonization inside tumors, we used results from reference to simulate colonization in a thin spherical layer inside a solid tumor.

The COMSOL model helped us to extend our MATLAB model in the following ways:

  • Diffusion physics of AHL was included to extend the ODEs into PDEs

    Our MATLAB model does not take into account diffusion of AHL, since in the viable lab experiments (in test-tubes), the only sink for AHL is its degradation. However, in a tumor as explained in reasoning, diffusion has much more significant contribution than degradation. Thus our model helped us gauge and verify the behavior of our tumor-sensing circuit in more real-life conditions pertaining to the intended application context. Using the results obtained from our simulations, we could check the behavior of the AND gate functioning in different conditions of dcell and lactate.

  • Diffusion physics of Azurin was included to simulate the effect of lysis

    To simulate the effect of lysis, our COMSOL model stops the production of Azurin and starts its diffusion when temperature reaches 42°C. This simulates the effect of increase in temperature with FUS to cause cell lysis. Using data obtained from such a simulation, we could also find the temporal-maximum concentrations of Azurin at each point in the tumor, effectively helping us to estimate the killing area.

Using our model, we also tried a few other colonization patterns to show our system works as expected inside a tumor while stays dormant in healthy tissue. We simulated the following patterns:

  • Homogeneous distribution in a Single spherical shell layer in tumour

  • Heterogeneous distribution in a Single spherical shell layer in tumour

  • Heterogeneous distribution in Double spherical shell layer in tumour

  • Homogeneous distribution throughout Healthy tissue

Model Overview

MAGIC!

  • Parameters

  • Equations
    • Growth phase

    • Sensing (AHL and LuxI)

    • Azurin

  • Assumptions

Results

Beautiful GIFs coming soon .... MUST REALLY SOON

  • CASE: Tumour colonization - High dcell AND High Lactate

  • CASE: Healthy tissue colonization: High dcell AND Low Lactate

    Test_gif Test_gif Test_gif
    HighDLowL_norm
    Figure 1. Normalized concentration of AHL and Azurin, Cell density (as a ratio of its steady state value), PluxlldR and Temperature (as a ratio of its steady state value 42°C).
  • CASE: Tumour not colonized: Low dcell AND High Lactate

  • CASE: Healthy tissue not colonized: Low dcell AND Low Lactate

Bacterial colonization patterns

Beautiful GIFs coming soon .... MUST REALLY SOON

  • Homogeneous distribution in a Single spherical shell layer in tumour

  • Homogeneous distribution throughout Healthy tissue

  • Heterogeneous (Partitioned) distribution in a Single spherical shell layer in tumour

  • Heterogeneous (Partitioned) distribution in Double spherical shell layer in tumour

Tools used