Team:NUDT CHINA/Modeling

From 2014.igem.org

(Difference between revisions)
 
(8 intermediate revisions not shown)
Line 23: Line 23:
<li><a href="https://2014.igem.org/Team:NUDT_CHINA/Modeling">Modeling</a> </li>
<li><a href="https://2014.igem.org/Team:NUDT_CHINA/Modeling">Modeling</a> </li>
<li><a href="https://2014.igem.org/Team:NUDT_CHINA/Notebook">Notebook</a> </li>
<li><a href="https://2014.igem.org/Team:NUDT_CHINA/Notebook">Notebook</a> </li>
-
<li><a href="https://2014.igem.org/Team:NUDT_CHINA/Safety">Safety</a> </li>
+
<li><a href="https://2014.igem.org/Team:NUDT_CHINA/Safety">Safety Policy & Practices</a> </li>
<li><a href="https://2014.igem.org/Team:NUDT_CHINA/Attributions">Attributions</a> </li>
<li><a href="https://2014.igem.org/Team:NUDT_CHINA/Attributions">Attributions</a> </li>
Line 31: Line 31:
<p>
<p>
<center>
<center>
-
<img src="" />
+
<img src="https://static.igem.org/mediawiki/2014/a/a2/NUDT_CHINA_modeling_figure_1.png" width="300" height="170" /><br>
-
Fig. 1 Cascade Regulatory Framework
+
Fig. 1 Cascade Regulatory Framework <br>
-
<img src="" />
+
<img src="https://static.igem.org/mediawiki/2014/b/bb/NUDT_CHINA_modeling_figure_2.png" width="500" height="180" /><br>
-
Fig. 2 Cascade Regulatory Pathway in DNA
+
Fig. 2 Cascade Regulatory Pathway in DNA <br>
</center>
</center>
</p>
</p>
<p>
<p>
-
According the cascade regulatory framework (Fig. 1) to solve the shortest path problem, we can build the cascade regulatory path in the plasmid of E.coli (Fig. 2). Now, we divide the whole cascade regulatory pathway into five units, which share same structures and similar properties (Fig. 3). Every unit can perform three common behaviours, i.e. promotion, transcription and translation. Usually, we can combine the process of promotion and translation when building and calculating the mathematic model of cascade regulatory. After combination, it is reasonable to assume that the transcriptional rate is in direct proportion to the extent of promotion. So we now get five easier units which can achieve two separated functions: promotion & transcription and translation.  
+
According the cascade regulatory framework (Fig. 1) to solve the shortest path problem, we can build the cascade regulatory path in the plasmid of <i>E. coli</i> (Fig. 2). Now, we divide the whole cascade regulatory pathway into five units, which share same structures and similar properties (Fig. 3). Every unit can perform three common behaviours, i.e. promotion, transcription and translation. Usually, we can combine the process of promotion and translation when building and calculating the mathematic model of cascade regulatory. After combination, it is reasonable to assume that the transcriptional rate is in direct proportion to the extent of promotion. So we now get five easier units which can achieve two separated functions: promotion & transcription and translation.  
</p><p>
</p><p>
The logic of the cascade regulation is:</p>
The logic of the cascade regulation is:</p>
Line 54: Line 54:
Here, we donate:</p><p>
Here, we donate:</p><p>
<ul>
<ul>
-
<li>the concentration of promoter binding sites in unit <i>i</i> is <math>D_i</math>;</li>
+
<li>the concentration of promoter binding sites in unit <i>i</i> is <img src="https://static.igem.org/mediawiki/2014/d/d8/NUDT_CHINA_modeling_equation_di.png" />;</li>
-
<li>the concentration of upstream promoter protein in unit <i>i</i> is <math>P_i(t)</math>;</li>
+
<li>the concentration of upstream promoter protein in unit <i>i</i> is <img src="https://static.igem.org/mediawiki/2014/6/6f/NUDT_CHINA_modeling_equation_pit.png" />;</li>
-
<li>the concentration of mrna in unit <i>i</i> is <math>R_i(t)</math>;</li>
+
<li>the concentration of mrna in unit <i>i</i> is <img src="https://static.igem.org/mediawiki/2014/f/ff/NUDT_CHINA_modeling_equation_rit.png" />;</li>
-
<li>the concentration of product protein in unit <i>i</i> is <math>X_i(t)</math>;</li>
+
<li>the concentration of product protein in unit <i>i</i> is <img src="https://static.igem.org/mediawiki/2014/0/06/NUDT_CHINA_modeling_equation_xit.png" />;</li>
-
<li>the hill coefficient of promotion in unit <i>i</i> is <math>n_i</math>;</li>
+
<li>the hill coefficient of promotion in unit <i>i</i> is <img src="https://static.igem.org/mediawiki/2014/f/f5/NUDT_CHINA_modeling_equation_ni.png" />;</li>
-
<li>the dissociation constant in unit <i>i</i> is  .</li>
+
<li>the dissociation constant in unit <i>i</i> is  <img src="https://static.igem.org/mediawiki/2014/5/55/NUDT_CHINA_modeling_equation_kpdni.png" />.</li>
</ul>
</ul>
</p>
</p>
 +
<p>where <img src="https://static.igem.org/mediawiki/2014/a/a8/NUDT_CHINA_modeling_equation_i12345.png" /> (See Fig. 3)</p>
-
</td>
+
<p><center><img src="https://static.igem.org/mediawiki/2014/b/b1/NUDT_CHINA_modeling_figure_3.png" /><br>Fig. 3 Five Units of the Cascade Regulatory Pathway and the Statement of Symbols</center></p>
-
</tr>
+
-
<tr><td bgColor="#e7e7e7" height="1px"> </tr>
+
-
<tr><td>
+
-
<p>
+
-
<h3> Content</h3>
+
-
</p>
+
-
<p>Here is the detailed explain for our project.</p>
+
-
<h4>Overall project summary</h4>
+
<p><h5>III. Mathematic Model</h5></p>
-
<p>Recent years, there were some researches aiming to solve specific mathematical problems using synthetic biology, one of which even shed lights on the possibility of solving Hamilton Problem in engineered bacteria. Inspired by them, we are trying to design and construct gene circuits to deal with other problem in graph theory. What we want to prove is that the organism have potential for computation as long as they are designed appropriately.
+
 
 +
<p>In the model of the cascade regulatory pathway, we can gets equations according to the mRNA & protein metabolism of each unit and the protein transmit among different units.</p>
 +
 
 +
<p><h6>i. mRNA Metabolism of Each Unit</h6></p>
 +
<p>According to and the Hill equation, we suppose that mRNA degraded without enzyme. So the degradation rate is proportional to the concentration of the mRNA itself. Now, we can describe the rate of promotion & transcription by time:</p>
 +
 
 +
<p><center><img src="https://static.igem.org/mediawiki/2014/1/1c/NUDT_CHINA_modeling_equation_3_1.png" />&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(3.1)</center></p>
 +
<p>Where <img src="https://static.igem.org/mediawiki/2014/3/3f/NUDT_CHINA_modeling_equation_vi.png" /> is the maximum formation rate of mRNA in unit <i>i</i> and <img src="https://static.igem.org/mediawiki/2014/f/f8/NUDT_CHINA_modeling_equation_ri.png" /> is the degradation rate constant of mRNA in unit <i>i</i>.(<img src="https://static.igem.org/mediawiki/2014/a/a8/NUDT_CHINA_modeling_equation_i12345.png" />)</p>
 +
 
 +
<p><h6>ii. Protein Metabolism of Each Unit</h6></p>
 +
<p>In the same way, we suppose that protein degraded without enzyme. So the degradation rate is proportional to the concentration of the mRNA itself. Now, we can descript the rate of translation  by time:</p>
 +
<p><center><img src="https://static.igem.org/mediawiki/2014/c/c8/NUDT_CHINA_modeling_equation_3_2.png" />&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(3.2)</center></p>
 +
<p>Where <img src="https://static.igem.org/mediawiki/2014/c/c8/NUDT_CHINA_modeling_equation_ui.png" />is the maximum formation rate of protein in unit <i>i</i> and <img src="https://static.igem.org/mediawiki/2014/4/4d/NUDT_CHINA_modeling_equation_xi.png" /> is the degradation rate constant of protein in unit <i>i</i>. (<img src="https://static.igem.org/mediawiki/2014/a/a8/NUDT_CHINA_modeling_equation_i12345.png" />)</p>
 +
 
 +
<p><h6>iii. Protein Transmit Among Different Units</h6></p>
 +
<p>From the topological relationship of the cascade regulatory pathway (Fig. 3), we can get the equations by describing the protein transmitting among different units:</p>
 +
<p><center><img src="https://static.igem.org/mediawiki/2014/2/2c/NUDT_CHINA_modeling_equation_3_3.png" />&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(3.2)</center></p>
 +
<p>Theoretically, we can get the function of <img src="https://static.igem.org/mediawiki/2014/9/9b/NUDT_CHINA_modeling_equation_x5.png" /> , which tells the temporal concentration of green fluorescence protein (GFP), by simultaneous equations (3.1)(3.2)(3.3).</p>
 +
 
 +
<p><h5>IV. Calculation and Results of the Mathematic Model</h5></p>
 +
<p><h6>i. Calculation</h6></p>
 +
<p>Limited by the calculation scale and experimental data, we assume that parameters (except the input and output) of each units are identical.</p>
 +
<p><center><img src="https://static.igem.org/mediawiki/2014/3/3b/NUDT_CHINA_modeling_equation_4_1.png" />&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(4.1)<br>
 +
<img src="https://static.igem.org/mediawiki/2014/d/d9/NUDT_CHINA_modeling_equation_4_2.png" />&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(4.2)</center></p>
 +
<p>In addition, we set the parameters above according more to calculation results than to experimental data so that the parameters have no units. In this way, we can still get useful semiquantitative result.</p>
 +
<p>Accoring to the equations (3.1)(3.2)(3.3)(4.1)(4.2), we wrote calculation program in Matlab to conduct numerical simulation and get the curve of <img src="https://static.igem.org/mediawiki/2014/4/4a/NUDT_CHINA_modeling_equation_x5tt.png" /> under the cascade regulation pathway “111”.</p>
 +
<p>Similarly, by modifying the equation (3.1) we can get the the curve of <img src="https://static.igem.org/mediawiki/2014/4/4a/NUDT_CHINA_modeling_equation_x5tt.png" /> under the cascade regulation pathway“110”, “001”, “101” and “011”, respectively.</p>
 +
<p>The modification rule is show in the chart below.</p>
 +
 
 +
<p><table border=1 width=100%><tr>
 +
<td width="25%"><strong>110</strong></td>
 +
<td width="25%"><strong>001</strong></td>
 +
<td width="25%"><strong>101</strong></td>
 +
<td width="25%"><strong>011</strong></td></tr><tr>
 +
<td width="25%"><img src="https://static.igem.org/mediawiki/2014/3/38/NUDT_CHINA_modeling_equation_r3t0.png" /></td>
 +
<td width="25%"><img src="https://static.igem.org/mediawiki/2014/0/0c/NUDT_CHINA_modeling_equation_r2tr4t.png" /></td>
 +
<td width="25%"><img src="https://static.igem.org/mediawiki/2014/4/4f/NUDT_CHINA_modeling_equation_r4t0.png" /></td>
 +
<td width="25%"><img src="https://static.igem.org/mediawiki/2014/c/c0/NUDT_CHINA_modeling_equation_r2t0.png" /></td></tr></table>
</p>
</p>
-
<h4>Project Details</h4>
+
<p>where the rest equations are the same as “111”.</p>
-
<p>SPP-for-short problem asks the shortest pathway between two given points in a directed graph. We came up with an idea to translate this shortest of space into quickest of time with the help of transcriptional cascades in <i>Escherichia coli</i>. That is to say that we build gene circuits to encode a directed graph where each nodes and edges within are represented by certain promoters and TFs respectively. Then a proof-of-concept experiment in vivo was carried to confirm the validation of this design.
+
 
-
</p>
+
<p>Now, we put the curve <img src="https://static.igem.org/mediawiki/2014/4/4a/NUDT_CHINA_modeling_equation_x5tt.png" />  of “111”, “110”, “001”, “101” and “011”in one figure and label them in different colors. (Fig. 4) So we can find difference of the GFP formation rate among these cascade regulatory pathways.</p>
-
<h4>Materials and Methods</h4>
+
 
-
<p>We used several parts from iGEM registry and constructed five new parts and devices. The experiments involved are basic ones in gene engineering including gene amplification and expression; polymerase chain reaction; electrophoresis etc.
+
<p><center><img src="https://static.igem.org/mediawiki/2014/d/de/NUDT_CHINA_modeling_figure_4.png" /></center><br>Fig. 4 Curve of <img src="https://static.igem.org/mediawiki/2014/4/4a/NUDT_CHINA_modeling_equation_x5tt.png" />. where the “001”, “101”, “011” nearly coincide with each other.</p>
-
</p>
+
 
-
<h4>The Experiments</h4>
+
<p><h6>ii. Result</h6></p>
 +
<p>Before giving the result, we list the cascade regulatory pathways of “111”, “110”, “001”, “101” and “011” by figures.</p>
<p>
<p>
 +
<table border=1 width=90%>
 +
<tr>
 +
<td width="30%"><strong>111</strong></td>
 +
<td width="30%"><strong>110</strong></td>
 +
<td width="30%"><strong>001</strong></td></tr><tr>
 +
<td width="30%"><img src="https://static.igem.org/mediawiki/2014/1/19/NUDT_CHINA_modeling_figure_5_111.png" /></td>
 +
<td width="30%"><img src="https://static.igem.org/mediawiki/2014/9/92/NUDT_CHINA_modeling_figure_5_110.png" /></td>
 +
<td width="30%"><img src="https://static.igem.org/mediawiki/2014/2/2d/NUDT_CHINA_modeling_figure_5_001.png" /></td>
 +
</tr>
 +
<tr>
 +
<td width="30%"><strong>101</strong></td>
 +
<td width="30%"><strong>011</strong></td>
 +
<td width="30%"><strong></strong></td></tr><tr>
 +
<td width="30%"><img src="https://static.igem.org/mediawiki/2014/0/09/NUDT_CHINA_modeling_figure_5_101.png" /></td>
 +
<td width="30%"><img src="https://static.igem.org/mediawiki/2014/5/56/NUDT_CHINA_modeling_figure_5_011.png" /></td>
 +
<td width="30%"></td>
 +
</tr>
 +
</table>
</p>
</p>
-
<h4>Results</h4>
+
 
-
<p>We designed a series of genetic circuits in Escherichia coli forming a regulatory network in order to translate the form of mathematical problem from geometrical graph into biological structure. The nodes and arrows are marked by well-assigned promoters and transcription factors (TFs) respectively. The promoter of destination node in SPP is followed by a green fluorescent protein (GFP) as a reporter. In this way, the connections among nodes in a directed graph are simulated by transcriptional regulation network. The temporal ordering of the fluorescent protein expression in bacteria reflects distance differences among varied paths, in which an individual bacteria clone containing the shortest pathway encoded gene would show fluorescent earlier than the others. In the following one bit knock-out process, we verified the phenotype which showed fluorescent earlier resulted from the genotype that represented the shortest pathway solution, shedding lights on that our encoding system functioned as expected with linear computation complexity and this approach is theoretically proved feasible.
+
<p>Obviously, the GFP formation is later in “110” than others because the absence of pathway ③. The result is a powerful support to our iGEM design to solve the shortest path problem.</p>
-
</p>
+
 
-
<h4>Data analysis</h4>
+
<p><h5>Problems</h5></p>
-
<p>
+
<p>There are still some problems that need more experiment and time to deal with during building and calculating the model.</p>
-
</p>
+
<p><h6>i. During Building the Model</h6></p>
-
<h4>Conclusions</h4>
+
<p>Including the neglected details of Hill model itself, our model do not take the enzymatic degradation of mRNA and protein which, in fact, is common in cells.</p>
-
<p>The bio-computing method we have come up with is a brand-new way to solve the SPP problem effectively. Based on the standardization principals and abstraction strategies of synthetic biology, the results also validated synthetic biology as a valuable approach to biological engineering. In the discussion concerning its augmentability, the potential value and superiority of solving NPC problems can be observed when combining metabolism pathway modification engineering.
+
<p><h6>ii. During Calculation</h6></p>
-
</p>
+
<p>We suppose the parameters of the five is identical and set the parameters based more on calculation results than on experimental data. But in fact, the difference between different regulatory units can be obvious.</p>
<div align="right">
<div align="right">
-
<a href="https://2014.igem.org/wiki/index.php?title=Team:NUDT_CHINA/Project&action=edit"> Click here  to edit this page!</a>
+
<a href="https://2014.igem.org/wiki/index.php?title=Team:NUDT_CHINA/Modeling&action=edit"> Click here  to edit this page!</a>
<a href="https://2014.igem.org/Special:Upload">Click here to upload! </a>
<a href="https://2014.igem.org/Special:Upload">Click here to upload! </a>
</div>
</div>

Latest revision as of 03:02, 18 October 2014


The Mathematic Model and Main Results

I. Analyses of Cascade Regulatory Pathway
According the cascade regulatory framework (Fig. 1) to solve the shortest path problem, we can build the cascade regulatory path in the plasmid of E.coli (Fig. 2). Now, we divide the whole cascade regulatory pathway into five units, which share same structures and similar properties (Fig. 3). Every unit can perform three common behaviours, i.e. promotion, transcription and translation. Usually, we can combine the process of promotion and translation when building and calculating the mathematic model of cascade regulatory. After combination, it is reasonable to assume that the transcriptional rate is in direct proportion to the extent of promotion. So we now get five easier units which can achieve two separated functions: promotion & transcription and translation.


Fig. 1 Cascade Regulatory Framework

Fig. 2 Cascade Regulatory Pathway in DNA

According the cascade regulatory framework (Fig. 1) to solve the shortest path problem, we can build the cascade regulatory path in the plasmid of E. coli (Fig. 2). Now, we divide the whole cascade regulatory pathway into five units, which share same structures and similar properties (Fig. 3). Every unit can perform three common behaviours, i.e. promotion, transcription and translation. Usually, we can combine the process of promotion and translation when building and calculating the mathematic model of cascade regulatory. After combination, it is reasonable to assume that the transcriptional rate is in direct proportion to the extent of promotion. So we now get five easier units which can achieve two separated functions: promotion & transcription and translation.

The logic of the cascade regulation is:

  • the translation of this cascade is regulated by the product (protein) of the upstream cascade;
  • and identically, the product (protein) of this cascade regulates the translation of the downstream.

In addition, we need to take the temporal degradation of mRNA and protein into account.

I. Analyses of Cascade Regulatory Pathway
Here, we donate:

  • the concentration of promoter binding sites in unit i is ;
  • the concentration of upstream promoter protein in unit i is ;
  • the concentration of mrna in unit i is ;
  • the concentration of product protein in unit i is ;
  • the hill coefficient of promotion in unit i is ;
  • the dissociation constant in unit i is .

where (See Fig. 3)


Fig. 3 Five Units of the Cascade Regulatory Pathway and the Statement of Symbols

III. Mathematic Model

In the model of the cascade regulatory pathway, we can gets equations according to the mRNA & protein metabolism of each unit and the protein transmit among different units.

i. mRNA Metabolism of Each Unit

According to and the Hill equation, we suppose that mRNA degraded without enzyme. So the degradation rate is proportional to the concentration of the mRNA itself. Now, we can describe the rate of promotion & transcription by time:

      (3.1)

Where is the maximum formation rate of mRNA in unit i and is the degradation rate constant of mRNA in unit i.()

ii. Protein Metabolism of Each Unit

In the same way, we suppose that protein degraded without enzyme. So the degradation rate is proportional to the concentration of the mRNA itself. Now, we can descript the rate of translation by time:

      (3.2)

Where is the maximum formation rate of protein in unit i and is the degradation rate constant of protein in unit i. ()

iii. Protein Transmit Among Different Units

From the topological relationship of the cascade regulatory pathway (Fig. 3), we can get the equations by describing the protein transmitting among different units:

      (3.2)

Theoretically, we can get the function of , which tells the temporal concentration of green fluorescence protein (GFP), by simultaneous equations (3.1)(3.2)(3.3).

IV. Calculation and Results of the Mathematic Model

i. Calculation

Limited by the calculation scale and experimental data, we assume that parameters (except the input and output) of each units are identical.

      (4.1)
      (4.2)

In addition, we set the parameters above according more to calculation results than to experimental data so that the parameters have no units. In this way, we can still get useful semiquantitative result.

Accoring to the equations (3.1)(3.2)(3.3)(4.1)(4.2), we wrote calculation program in Matlab to conduct numerical simulation and get the curve of under the cascade regulation pathway “111”.

Similarly, by modifying the equation (3.1) we can get the the curve of under the cascade regulation pathway“110”, “001”, “101” and “011”, respectively.

The modification rule is show in the chart below.

110 001 101 011

where the rest equations are the same as “111”.

Now, we put the curve of “111”, “110”, “001”, “101” and “011”in one figure and label them in different colors. (Fig. 4) So we can find difference of the GFP formation rate among these cascade regulatory pathways.


Fig. 4 Curve of . where the “001”, “101”, “011” nearly coincide with each other.

ii. Result

Before giving the result, we list the cascade regulatory pathways of “111”, “110”, “001”, “101” and “011” by figures.

111 110 001
101 011

Obviously, the GFP formation is later in “110” than others because the absence of pathway ③. The result is a powerful support to our iGEM design to solve the shortest path problem.

Problems

There are still some problems that need more experiment and time to deal with during building and calculating the model.

i. During Building the Model

Including the neglected details of Hill model itself, our model do not take the enzymatic degradation of mRNA and protein which, in fact, is common in cells.

ii. During Calculation

We suppose the parameters of the five is identical and set the parameters based more on calculation results than on experimental data. But in fact, the difference between different regulatory units can be obvious.