Team:USTC-Software/Project/Method

From 2013.igem.org

(Difference between revisions)
 
(92 intermediate revisions not shown)
Line 1: Line 1:
{{USTC-Software/hidden}}
{{USTC-Software/hidden}}
-
{{USTC-Software/header}}
+
{{USTC-Software/header2}}
 +
 
 +
<!--{{USTC-Software/project}}-->
<html>
<html>
Line 8: Line 10:
<title>Methodologies</title>
<title>Methodologies</title>
-
 
+
<script type="text/javascript" src="http://igem.stlover.org/js/meth.js"></script>
<link rel="stylesheet" href="https://2013.igem.org/Team:USTC-Software/Project/Method/css?action=raw&ctype=text/css" type="text/css"/>
<link rel="stylesheet" href="https://2013.igem.org/Team:USTC-Software/Project/Method/css?action=raw&ctype=text/css" type="text/css"/>
 +
 +
<link rel="stylesheet" href="https://2013.igem.org/Team:USTC-Software/Project/css?action=raw&ctype=text/css" type="text/css"/>
 +
<script src="https://2013.igem.org/Team:USTC-Software/Project/Overall/js/slide1?action=raw&ctype=text/javascript" type="text/javascript" charset="utf-8"></script>
<script src="https://2013.igem.org/Team:USTC-Software/Project/Overall/js/slide1?action=raw&ctype=text/javascript" type="text/javascript" charset="utf-8"></script>
Line 20: Line 25:
<script src="https://2013.igem.org/Team:USTC-Software/Project/Overall/js/slide8?action=raw&ctype=text/javascript" type="text/javascript" charset="utf-8"></script>
<script src="https://2013.igem.org/Team:USTC-Software/Project/Overall/js/slide8?action=raw&ctype=text/javascript" type="text/javascript" charset="utf-8"></script>
-
 
+
<style> #methodbutton { color:#aaed51 !important;}</style>
<body>
<body>
Line 26: Line 31:
<div id="direction">
<div id="direction">
         <ul>
         <ul>
-
           <li><a href="#abstract" class="button">Abstract</a>
+
                             
-
           <li><a href="#Fetch_Database" class="button">Fetch Database</a>
+
           <!--li><a href="#sequence" class="button">2.1 Sequence</a></br>
-
           <li><a href="#Alignment_Analyze" class="button">Alignment Analyze</a>
+
                <a href="#nwa" class="button" id="subbutton">2.1.1 Needleman-Wunsch Algorithm</a></br>
-
           <li><a href="#New_Network_Construction" class="button">New Network Construction</a>
+
                <a href="#asg" class="button" id="subbutton">2.1.2 A Supplementary Game</a>
-
          <li><a href="#Network_Model" class="button">Network Model</a>
+
          </li>
-
          <li><a href="#Predict" class="button">Predict</a>           
+
 
 +
           <li><a href="#filtering" class="button">2.2 Filtering</a></br>
 +
                <a href="#rn" class="button" id="subbutton">2.2.1 Random Noise</a></br>
 +
                <a href="#filter" class="button" id="subbutton">2.2.2 Filter</a>
 +
        </li>
 +
 
 +
           <li><a href="#rc" class="button">2.3 Regulation Calculation</a></li>
 +
 
 +
           <li><a href="#main" class="button">Top</a></li-->
 +
 
 +
 
 +
        <li><a href="#Fetch_Database" class="button">Database</a></li>
 +
        <li><a href="#Alignment_Analyze" class="button">Operon Theory and Regulatory Model</a></li>
 +
        <li><a href="#fa" class="button">Forward Analysis</a></li>
 +
         <li><a href="#ra" class="button">Reverse Analysis</a></li>
 +
        <li><a href="#reference" class="button">Reference</a></li>
 +
        <li><a href="#main" class="button">Top</a></li>
 +
 
         </ul>
         </ul>
</div>
</div>
Line 39: Line 61:
         var href = $(this).attr("href");
         var href = $(this).attr("href");
         var pos = $(href).offset().top - 100;
         var pos = $(href).offset().top - 100;
-
         $("html,body").animate({scrollTop: pos}, 1500);
+
         $("html,body").animate({scrollTop: pos}, 1500);//the smaller the quicker
         return false;
         return false;
     });
     });
Line 52: Line 74:
   <div id="abstract">
   <div id="abstract">
   <h1 align="justify">Methodologies</h1>
   <h1 align="justify">Methodologies</h1>
-
   <p align="justify">To simulate and analyze a genetic regulatory network (GRN), we need to build an objects’ array to store the complete information of each gene. It contains regulation relationships between genes, sequences of genes, sequences of promoters and so on. However, it’s hard to find an appropriate database online containing all information we need in a simple file. RegulonDB has downloadable files about the regulation between transcription factors (TF) and genes. Files about genetic information, transcription unit information and promoter information can also be downloaded from the RegulonDB. All those files have been put into file “source data” in the root directory of our software. They contain all information the simulation needs and we use fetching module to achieve data extraction and integration. There are four steps: fetch regulation relationships, fetch gene information, fetch promoter information and integrate information above.
+
   <p align="justify">In order to simulate the GRN's working and analyze the changing after exogenous gene imported, some advanced algorithms and classical methods are employed in the software. These algorithms and methods include Binary Tree method, Needle-Wunsch Algorithm, Decision Tree method, Hill Equation and PSO Algorithm.</br><br/>
 +
 
 +
<img src="https://static.igem.org/mediawiki/2013/6/6b/USTC_Software_FLOW.png" style="width:1000px;"/></br>
 +
 
 +
There are four parts of methodologies: Database, Operon Theory and Regulatory Model, Forward Analysis and Reverse Analysis.
   </p>
   </p>
        
        
Line 61: Line 87:
<div id="Fetch_Database">
<div id="Fetch_Database">
-
<h2>Fetch Database</h2>
+
<h2>Database</h2>
 +
<div id="jobs_container">
 +
        <div class="jobs_trigger" id="dbabstract"><strong>Abstract</strong></div>
 +
<div class="jobs_item" style="display: none;"><p class="bodytext"></p><p align="justify">To simulate and analyze a genetic regulatory network (GRN), we need to build an objects' array to store the complete information of each gene. It contains regulation relationships between genes, sequences of genes, sequences of promoters and so on. However, it's hard to find an appropriate database online containing all information we need in a simple file. RegulonDB has downloadable files about the regulation between transcription factors (TF) and genes. Files about genetic information, transcription unit information and promoter information can also be downloaded from the RegulonDB. All those files have been put into file “source data” in the root directory of our software. They contain all information the simulation needs and we use fetching module to achieve data extraction and integration. There are four steps: fetch regulation relationships, fetch gene information, fetch promoter information and integrate information above.
 +
</p>
 +
                </div>
   <div id="jobs_container">
   <div id="jobs_container">
-
        <div class="jobs_trigger"><strong>Fetching Regulation</strong></div>
+
        <div class="jobs_trigger" id="fetch"><strong>Fetch Regulation</strong></div>
-
<div class="jobs_item" style="display: none;"><p class="bodytext"></p><p align="justify">In GRN, there are two kinds of files (download here[]): TF to TF and TF to Gene. Since the database about the regulation between TFs and Genes contains only one-way interaction, the matrix of GRN is a rectangle.
+
<div class="jobs_item" style="display: none;"><p class="bodytext"></p><p align="justify">In GRN, there are two kinds of files: <a class="content" href="http://regulondb.ccg.unam.mx/menu/download/datasets/files/network_tf_tf.txt"> TF to TF</a> and <a class="content" href="http://regulondb.ccg.unam.mx/menu/download/datasets/files/network_tf_gene.txt">TF to Gene</a>. Since the database about the regulation between TFs and Genes contains only one-way interaction, the matrix of GRN is a rectangle.</br></br>
-
First of all, read the regulation relationship of TFs. Our software filters the documentation of RegulonDB on the head of all files and then reads the name of regulate and regulated TF, which is also the name of its genes, one by one. In the same time, our software numerates the genes and stores their names into an objects’ array of genetic data.  
+
First of all, read the regulation relationship of TFs. Our software filters the documentation of RegulonDB on the head of all files and then reads the name of regulate and regulated TF, which is also the name of its genes, one by one. In the same time, our software numerates the genes and stores their names into an objects' array of genetic data. </br></br>
-
The format of regulation database:
+
&nbsp;&nbsp;The format of regulation database:</br>
-
TF_name   TF_name   +/-/+-
+
&nbsp;&nbsp;&nbsp;&nbsp;TF_name &nbsp;&nbsp;&nbsp;TF_name &nbsp;&nbsp;&nbsp;+/-/+-</br></br>
-
 
+
<div align="center"><img src="https://static.igem.org/mediawiki/2013/6/69/USTC_Software_TT.jpg"/></div></br>
-
The regulation of TFs has been put into a square matrix whose row is the regulator and column is the one regulated by. To make our GRN as complete as possible, the regulation between TF and genes has joined into the matrix. The one-way interaction results that we must read the TF in order to fulfill the regulator before completing the TF to gene’s regulation in the same way of TF to TF. The format of regulation database:
+
The regulation of TFs has been put into a square matrix whose row is the regulator and column is the one regulated by. To make our GRN as complete as possible, the regulation between TF and genes has joined into the matrix. The one-way interaction results that we must read the TF in order to fulfill the regulator before completing the TF to gene's regulation in the same way of TF to TF. </br></br>
-
TF_name   Gene_name   +/-/+-
+
&nbsp;&nbsp;The format of regulation database:</br>
-
 
+
&nbsp;&nbsp;&nbsp;&nbsp;TF_name &nbsp;&nbsp;&nbsp;Gene_name &nbsp;&nbsp;&nbsp;+/-/+-</br></br>
-
At last, a regulatory matrix whose row represents regulate gene (TF) and whose column represents gene regulated by (TF+Gene) has been output into a file called “old_GRN” in root directory. The values in GRN matrix are regulations in which “1” means positive activation, “-1” means repression and “0” means no relationship. There have been some regulations both positive and negative identified regulations are determined by the experimental environment. As a result, our software picks out those uncertain genes and stores them into a file named “uncertain_database”.
+
<div align="center"><img src="https://static.igem.org/mediawiki/2013/4/47/USTC_Software_TG.jpg"/></div></br>
-
The format of uncertain database:
+
At last, a regulatory matrix whose row represents regulate gene (TF) and whose column represents gene regulated by (TF+Gene) has been output into a file called “old_GRN” in root directory. The values in GRN matrix are regulations in which “1” means positive activation, “-1” means repression and “0” means no relationship. There have been some regulations both positive and negative identified regulations are determined by the experimental environment. As a result, our software picks out those uncertain genes and stores them into a file named “uncertain_database”.</br></br>
-
?   Gene_name->Gene_name
+
&nbsp;&nbsp;The format of uncertain database:</br>
 +
&nbsp;&nbsp;&nbsp;&nbsp;? &nbsp;&nbsp;&nbsp;Gene_name->Gene_name</br></br>
The question mark represents the unknown regulation between regulator and regulated-by whose names presented afterward. Users could replace the question mark with the data known in past experiment. (“+” rep positive, “-” rep negative). Our software will replace the values in matrix automatically. But if not rewrote, our software will regard those regulation as unknown.
The question mark represents the unknown regulation between regulator and regulated-by whose names presented afterward. Users could replace the question mark with the data known in past experiment. (“+” rep positive, “-” rep negative). Our software will replace the values in matrix automatically. But if not rewrote, our software will regard those regulation as unknown.
Line 81: Line 113:
                 </div>
                 </div>
-
<div class="jobs_trigger"><strong> Fetching Gene Info</strong></div>
+
<div class="jobs_trigger" id="fgi"><strong> Fetch Gene Info</strong></div>
<div class="jobs_item" style="display: none;"><p align="justify">
<div class="jobs_item" style="display: none;"><p align="justify">
-
All gene information has been deposited into a file named gene_info which could be downloaded here[]. In order of picking out the genes in GRN as fast as possible, all genetic information are stored in a “map”. “Map” is just like a dictionary yet its words are names of genes and its descriptions of words are replaced by genetic information. By using binary tree method, it is very fast to searth the “word” wanted in the “dictionary”. As tested, the speed of binary tree method built-in “map” function is 720 times faster than traversal method.
+
All gene information has been deposited into a file named gene_info which could be downloaded <a class="content" href="http://regulondb.ccg.unam.mx/menu/download/datasets/files/Gene_sequence.txt">here</a>. In order of picking out the genes in GRN as fast as possible, all genetic information are stored in a “map”. “Map” is just like a dictionary yet its words are names of genes and its descriptions of words are replaced by genetic information. By using binary tree method, it is very fast to search the “word” wanted in the “dictionary”. As tested, the speed of binary tree method built-in “map” function is 720 times faster than traversal method.</br></br>
-
The format of Gene Info database:
+
&nbsp;&nbsp;The format of Gene Info database:</br>
-
ID_assigned_by_RegulonDB   Gene_name   Left_end_position   Right_end_position   DNA_strand   Product_type   Product_name   Start_codon_sequence   Stop_codon_sequence   Gene_sequence
+
&nbsp;&nbsp;&nbsp;&nbsp;ID_assigned_by_RegulonDB &nbsp;&nbsp;&nbsp;Gene_name &nbsp;&nbsp;&nbsp;Left_end_position &nbsp;&nbsp;&nbsp;Right_end_position &nbsp;&nbsp;&nbsp;DNA_strand &nbsp;&nbsp;&nbsp;Product_type &nbsp;&nbsp;&nbsp;&nbsp;Product_name &nbsp;&nbsp;&nbsp;Start_codon_sequence&nbsp;&nbsp;&nbsp;  Stop_codon_sequence &nbsp;&nbsp;&nbsp;Gene_sequence</br></br>
-
 
+
<div align="center"><img src="https://static.igem.org/mediawiki/2013/4/45/USTC_Software_GI.jpg"/></div></br>
-
The label of the map vector is gene name which will be picked out based on the names read in regulation matrix before. It is really fast using the binary tree method to find the specific genetic information and store them into a specific object. Those information includes gene ID, left position, right position, gene description and gene sequence. The gene ID is used to link to RegulonDB’s gene details; The left position is used to find its specific transcription unit; The right position is used to figure out the base amount; The description of genes is used to distinguish the RNA and protein; The sequence is used to predict the regulation by alignment.
+
The label of the map vector is gene name which will be picked out based on the names read in regulation matrix before. It is really fast using the binary tree method to find the specific genetic information and store them into a specific object. Those information includes gene ID, left position, right position, gene description and gene sequence. The gene ID is used to link to RegulonDB's gene details; The left position is used to find its specific transcription unit; The right position is used to figure out the base amount; The description of genes is used to distinguish the RNA and protein; The sequence is used to predict the regulation by alignment.
</p>
</p>
Line 94: Line 126:
                  
                  
                  
                  
-
             <div class="jobs_trigger"> <strong>Fetching Promoter Info</strong></div>
+
             <div class="jobs_trigger" id="fpi"> <strong>Fetch Promoter Info</strong></div>
-
        <div class="jobs_item" style="display: none;"><p align="justify">All promoter information has been deposited into a file named promoter_info which could be downloaded here[]. But we also need transcription unit information because the information files about promoter do not contain all genes’ names backward. “TU Info” file, which can be downloaded here[], contains the starting position of each TU and its promoter name. Our software picks out the starting position into a integer array. Using the left position picked out in gene info, our software would find out which unit the gene belongs to through dichotomy method and then stores the name of promoter into corresponding object.
+
        <div class="jobs_item" style="display: none;"><p align="justify">All promoter information has been deposited into a file named promoter_info which could be downloaded <a class="content" href="http://regulondb.ccg.unam.mx/menu/download/datasets/files/PromoterSet.txt">here</a>. But we also need transcription unit information because the information files about promoter do not contain all genes' names backward. “TU Info” file, which can be downloaded <a class="content" href="http://regulondb.ccg.unam.mx/menu/download/datasets/files/TUSet.txt">here</a>, contains the starting position of each TU and its promoter name. Our software picks out the starting position into a integer array. Using the left position picked out in gene info, our software would find out which unit the gene belongs to through dichotomy method and then stores the name of promoter into corresponding object.</br></br>
-
The format of TU info database:
+
&nbsp;&nbsp;The format of TU info database:</br>
-
Operon_name   Unit_name   promoter_name   Transcription_start_site ......
+
&nbsp;&nbsp;&nbsp;&nbsp;Operon_name &nbsp;&nbsp;&nbsp;Unit_name &nbsp;&nbsp;&nbsp;promoter_name &nbsp;&nbsp;&nbsp;Transcription_start_site ......</br></br>
-
 
+
<div align="center"><img src="https://static.igem.org/mediawiki/2013/1/1e/USTC_Software_TI.jpg"/></div></br>
-
The principle of fetching information of promoters is same as fetching genes’s. Our software stores the promoter information from the file named “promoter_info” in a “map” which could be used to pick out the promoter sequence by searching promoter name through binary tree method.
+
The principle of fetching information of promoters is same as fetching genes's. Our software stores the promoter information from the file named “promoter_info” in a “map” which could be used to pick out the promoter sequence by searching promoter name through binary tree method.</br></br>
-
The format of Promoter Info database:
+
&nbsp;&nbsp;The format of Promoter Info database:</br>
-
Promoter_ID_assigned_by_RegulonDB   Promoter_name
+
&nbsp;&nbsp;&nbsp;&nbsp;Promoter_ID_assigned_by_RegulonDB &nbsp;&nbsp;&nbsp;Promoter_name</br></br>
-
 
+
<div align="center"><img src="https://static.igem.org/mediawiki/2013/8/8a/USTC_Software_PI.jpg"/></div></br>
-
The sequence of promoter will be used in the alignment method in next module which could make a prediction of exogenous genes’ regulation pattern.
+
The sequence of promoter will be used in the alignment method in next module which could make a prediction of exogenous genes' regulation pattern.
</p>          </div>   
</p>          </div>   
                  
                  
                  
                  
                
                
-
<div class="jobs_trigger"> <strong>Integration</strong></div>
+
<div class="jobs_trigger" id="Integration"> <strong>Integration</strong></div>
<div class="jobs_item" style="display: block;"><p align="justify">                     
<div class="jobs_item" style="display: block;"><p align="justify">                     
-
Our software integrates all information we picked out about genes and generates a file named “all_info” —— all information about genes —— for the output graphical interface’s reading. In the meanwhile, the array of objects containing all information has been stored in computer memory which greatly improve the computing speed of our software.
+
Our software integrates all information we picked out about genes and generates a file named “all_info” —— all information about genes —— for the output graphical interface's reading. In the meanwhile, the array of objects containing all information has been stored in computer memory which greatly improve the computing speed of our software.</br></br>
-
The format of all_info database:
+
&nbsp;&nbsp;The format of all_info database:</br>
-
No.   promoter_sequence   gene_sequence   gene_name   ID   left_position   right_position   promoter_name   description
+
&nbsp;&nbsp;&nbsp;&nbsp;No. &nbsp;&nbsp;&nbsp;promoter_sequence &nbsp;&nbsp;&nbsp;gene_sequence &nbsp;&nbsp;&nbsp;gene_name &nbsp;&nbsp;&nbsp;ID &nbsp;&nbsp;&nbsp;left_position &nbsp;&nbsp;&nbsp;right_position &nbsp;&nbsp;&nbsp;promoter_name &nbsp;&nbsp;&nbsp;&nbsp;description</br>
-
The fetching module generates three files: old_GRN, all_info and uncertain_database.
+
The fetching module generates three files: old_GRN, all_info and uncertain_database.</br>
</p>
</p>
           </div>
           </div>
Line 128: Line 160:
<div id="Alignment_Analyze">
<div id="Alignment_Analyze">
-
<h2>Alignment Analyze</h2>
+
<h2>Operon Theory and Regulatory Model</h2>
   <div id="jobs_container">
   <div id="jobs_container">
-
        <div class="jobs_trigger"><strong>An example</strong></div>
+
        <div class="jobs_trigger"><strong>Operon Theory</strong></div>
-
<div class="jobs_item" style="display: none;"><p class="bodytext"></p><p align="justify">We would like to start with a simple example. A cell operates like a basketball team. Imagine you are a manager of a team who wants to bring in some talent players making up a “big three” and build a champion-potential team this season. Before you pay the sky-high bills for the “big three”, you can evaluate the effect of the talent introduction. New members‘ records are good reference, but not the whole thing.
+
<div class="jobs_item" style="display: none;"><p class="bodytext"></p>
 +
                <p align="justify">In genetics, an operon is a functioning unit of genomic DNA containing a cluster of genes
 +
under the control of a single regulatory signal or promoter. The genes contained in the
 +
operon are either expressed together or not at all. Several genes must be both cotranscribed
 +
and co-regulated to define an operon.<br /><br />
 +
The first time "operon" was proposed is in a paper of French Academic Science, 1960.
 +
The lac operon of the model bacterium E. coli was discovered and provides a typical
 +
example of operon function. It consists a promoter, an operator, three structural genes and
 +
a terminator. The operon is regulated by several factors including the availability of glucose
 +
and lactose.<br /><br />
 +
From this paper, the so-called general theory of the operon was developed. According to
 +
the theory, all genes are controlled by means of operons through a single feedback
 +
regulatory mechanism-repression. The first operon to be described was the lac operon in
 +
E. coli. The 1965 Nobel Prize in Physiology and Medicine was awarded to François Jacob,
 +
André Michel Lwoff and Jacques Lucien Monod for their discoveries concerning the operon and virus synthesis.<br />
 +
              </p>
-
There are various factors influencing the effect of introduction. Let’s carefully choose one of the most profound aspects and focus on the relationship of the members. In the original team, you are familiar with all players’ characteristics, their roles in the team and the coach’s style, i.e. you have the information of the original player interaction network.  
+
<div align="center"><img src="https://static.igem.org/mediawiki/igem.org/7/7d/USTC_Software_Figure_1.png" />
 +
<p align="center"><strong>Figure 1.</strong> Structure of Operon</p></div>
 +
<p align="justify">An operon is made up of several structural genes arranged under a common promoter and
 +
regulated by a common operator. It is defined as a set of adjacent structural genes, plus
 +
the adjacent regulatory signals that affect transcription of the structural genes. The
 +
regulators of a given operon, including repressors, corepressors and activators, are not
 +
necessarily coded for by that operon.<br /><br />
 +
As a unit of transcription, upstream of the structural genes lies a promoter sequence which
 +
provides a site for RNA polymerase to bind and initiate transcription. Close to the promoter
 +
lies a section of DNA called an operator.<br /><br />
 +
Operon regulation can be either negative or positive by induction or repression. Negative
 +
control involves the binding of a repressor to the operator to prevent transcription.
 +
Operons can also be positively controlled. An activator protein binds to DNA, usually at a
 +
site other than the operator, to stimulate transcription.
 +
</p>
 +
<div align="center"><img style="width:600px;" src="https://static.igem.org/mediawiki/igem.org/2/25/USTC_Software_Figure_2.png"/>
 +
<p align="justify"><strong>Figure 2.</strong> Regulation of Operon
 +
1: RNA Polymerase, 2: Repressor, 3: Promoter, 4: Operator, 5: Lactose, 6: lacZ, 7:
 +
lacY, 8: lacA. Top: The gene is essentially turned off. There is no lactose to inhibit the
 +
repressor, so the repressor binds to the operator, which obstructs the RNA polymerase
 +
from binding to the promoter and making lactase.Bottom: The gene is turned on.Lactose
 +
is inhibiting the repressor, allowing the RNA polymerase to bind with the promoter, and
 +
express the genes, which synthesize lactase. Eventually, the lactase will digest all of the
 +
lactose, until there is none to bind to the repressor. The repressor will then bind to the
 +
operator, stopping the manufacture of lactase.</p></div>
-
You know Alex is a good shooter and Bob is a strong centre forward. Carl is your target player. Carl is famous for his shooting skills and appears dominant in the court. In other words, Carl shows more obvious similarity with Alex but a low level of similarity with Bob. Then, in the new player interaction network, Carl might play a role 80% like Alex and 20% like Bob. He is similar with Alex and Bob, however different. That’s an analysis and prediction at the global point of view.
 
-
[ Pic.1 Alex, Bob and new member Carl]
+
      </div>
-
 
+
-
Just like the basketball team example, researchers often need to insert exogenous genes(new players) into a cell(original team) to achieve a specific goal(win the champion). In the past, the behaviors of exogenous genes are mainly speculated by wet lab experiments. Now we are trying to give an answer before wearing laboratory gloves.
+
-
</p>
+
-
                </div>
+
-
<div class="jobs_trigger"><strong>Models</strong></div>
+
<div class="jobs_trigger"><strong>Regulatory Model</strong></div>
-
<div class="jobs_item" style="display: none;"><p align="justify">
+
<div class="jobs_item" style="display: none;"><p align="justify">Regulation of gene expression includes four levels. We choose the transcriptional level to simulate the regulation both for its significance and model simplification.</p>
-
Regulatory Model
+
                <div align="center"><img style="width:600px; height:auto;"src="https://static.igem.org/mediawiki/igem.org/8/87/USTC_Software_Figure_3.png" />
-
Regulation of gene expression includes 4 levels:
+
                <p><strong>Figure 3.</strong>Regulation of gene expression.<br />Our regulation model is built based on the operon theory.<br /> The promoter region is regarded as the main regulatory region.</p></div>
-
Level of DNA rearrangement.
+
      </div>
-
Level of transcriptional regulation.
+
-
Level of translation.
+
-
Level of post-translation
+
-
This year we focus on the level of transcriptional regulation both for the importance of the level [FIXME: do i need to describe why it is important?] and model simplification. By carefully examining the lac operon system, which is widely considered as the first discovery of the gene regulation system, we constructed our regulation model with functional units called “Regulation Unit” [FXIME: regulation or regulatory?]. A regulation unit consists of two segments. The first one is a promoter sequence and the second one is a protein coding sequence.
+
-
 
+
-
[Pic. 3 Promoter Sequence, Protein Coding Sequence and Regulation Unit]
+
-
 
+
-
In our model, the promoter segment is regarded as the main regulated region. A transcription factor is a protein that binds to specific DNA sequences, thereby controlling the flow of genetic information from DNA to mRNA. The binding sites are promoter regions of DNA adjacent to the genes that they regulate. At first, according to the lac operon system, regulation units as a regulatory target with the same promoter are supposed to have same behavior. But we found it is insufficient because there are units with the same promoter showing different properties. Then we took the genes regulated by transcription factors into consideration. The different properties of two units are first owing to their promoters. If they have the same promoter, their protein coding sequences are supposed to make the difference. By taking this method, it turns out that this model works better.
+
-
 
+
-
[Pic. 4 Regulated Region]
+
-
 
+
-
A unit in the network regulates another through the transcription factor. That is, the product of the protein coding sequence of the unit is a transcription factor and the transcription factor regulates the promoter of the another unit.  
+
-
</p>
+
-
                </div>
+
                  
                  
                  
                  
-
             <div class="jobs_trigger"> <strong>Prediction Model</strong></div>
+
             <div class="jobs_trigger"> <strong>Similarity and Homology</strong></div>
-
        <div class="jobs_item" style="display: none;"><p align="justify">The basic idea behind the prediction model is deceptively simple: the more similar two sequences are, the more likely they have similar behaviors. In fact, it is extremely difficult to predict an exogenous gene’s behavior because of the complexity of the problem, random noise of the system and the coupling of biosystems.  
+
        <div class="jobs_item" style="display: none;"><p align="justify">The sequence similarity is obtained by sequence alignment. It is defined as the proportion of the common subsequence in the aligned sequence. Any two sequences share a certain
 +
similarity. It should be noted that similarity and homology are two different concepts.<br /><br />
 +
As with anatomical structures, homology between protein or DNA sequences is defined in
 +
terms of shared ancestry. Two segments of DNA can have shared ancestry because of
 +
either a speciation event or a duplication event. The terms “percent homology” and
 +
“sequence similarity” are often used interchangeably. As with anatomical structures, high
 +
sequence similarity might occur because of convergent evolution, or, as with shorter
 +
sequences, because of chance. Such sequences are similar but not homologous.
 +
Sequence regions that homologous are also called conserved.<br /><br />
 +
In our project, we use similarity to connect the exogenous gene with the original network.
 +
Because there is a good chance that the exogenous gene is not homologous with the
 +
genes in the network.</p>         
 +
      </div>
 +
        <div class="jobs_item" style="display: none;"><p align="justify">The GRN matrix is the mathematical description of gene regulatory network in which “1” represents “enhance”, “-1” represents “repress” and “0” represents “no regulatory relationship”. The units(RU) in x-axis regulate the units in y-axis. A row can be seen as a vector containing all the information of the target(corresponding unit in the y-axis). Similarly, a column can be seen as a vector containing all the information of the regulator(corresponding unit in the x-axis).</p>         
 +
                </div>
 +
        <div class="jobs_item" style="display: none;"><p align="justify">The sequence similarity is obtained by sequence alignment based on Needleman-Wunsch algorithm[FIXME: wiki link here]. The Needleman-Wunsch algorithm performs a global alignment on two protein sequences or nucleotide sequences. It was the first application of dynamic programming to biological sequence comparison.<br /><br />
-
Advanced alignment algorithm is selected to reduce the complexity. Sequences which contain all the information of the species are the entity of the gene regulatory network. Sequence similarity is an essential concept to the prediction model. The selected alignment algorithm can significantly reduce the complexity of the problem and makes it possible to give a reliable prediction from a global point of view.
+
When dynamic programming is applicable, the method takes far less time than naive methods. Using a naive method, many of the subproblems are generated and sovled many times. The dynamic programming approach seeks to solve each subproblem only once. Once the solution to a given subproblem has been computed, it is stored to be looked up next time.<br /><br />
-
We designed a random method to filter the noise in sequence alignment. There are no totally different sequences. Even the similarity of any two random sequences is not zero. Filtered results are more significant and reliable to the following steps.
+
Like the Needleman-Wunsch algorithm, of which it is a variation, Smith-Waterman is also a dynamic programming algorithm. But it is a local sequence alignment algorithm. The famous BLAST(Basic Local Alignment Search Tool) is improved from Smith-Waterman algorithm. Although local algorithm has the desirable property that it is guaranteed to find the optimal local alignment, we decided to choose the global one because we regarded the segment sequence as a unit.<br /><br />
-
Coupling of biosystem is also simulated at some level. When predicting exogenous gene’s behavior, all the units in the original gene regulatory network are taken into consideration.
+
Sequences are aligned with different detailed methods in different situations. In the regulated side, what we care about is the DNA sequence. In the regulating side, it is the amino acid sequence. When it comes to predict the regulated behavior, we use a DNA substitution matrix to align promoter and protein coding sequences. In the prediction of regulating behavior, the substitution matrix BLOSUM_50 is used to align the amino acid sequences translated from protein coding sequences.<br /><br />
-
Given that the exogenous gene may have never been inserted into E. coli before, all possible reactions in gene regulatory network are reserved to be filtered.
+
-
 
+
-
Using the innovated methods above, we are trying to challenge the difficulties and obtain a global perspective of the relationship between the exogenous gene and the original gene regulatory network.
+
-
</p>          </div> 
+
-
               
+
-
            <div class="jobs_trigger"> <strong>Mathematical Description of The Network</strong></div>
+
-
        <div class="jobs_item" style="display: none;"><p align="justify">The GRN matrix is the mathematical description of gene regulatory network in which “1” represents “enhance”, “-1” represents “repress” and “0” represents “no regulatory relationship”. The units(RU) in x-axis regulate the units in y-axis. A row can be seen as a vector containing all the information of the target(corresponding unit in the y-axis). Similarly, a column can be seen as a vector containing all the information of the regulator(corresponding unit in the x-axis).
+
-
</p>          </div>   
+
-
 
+
-
            <div class="jobs_trigger"> <strong>Sequence similarity</strong></div>
+
-
        <div class="jobs_item" style="display: none;"><p align="justify">The sequence similarity is obtained by sequence alignment based on Needleman-Wunsch algorithm[FIXME: wiki link here]. The Needleman-Wunsch algorithm performs a global alignment on two protein sequences or nucleotide sequences. It was the first application of dynamic programming to biological sequence comparison.
+
-
 
+
-
When dynamic programming is applicable, the method takes far less time than naive methods. Using a naive method, many of the subproblems are generated and sovled many times. The dynamic programming approach seeks to solve each subproblem only once. Once the solution to a given subproblem has been computed, it is stored to be looked up next time.
+
-
 
+
-
[Pic. 5 Dynamic programming and naive method]
+
-
 
+
-
Like the Needleman-Wunsch algorithm, of which it is a variation, Smith-Waterman is also a dynamic programming algorithm. But it is a local sequence alignment algorithm. The famous BLAST(Basic Local Alignment Search Tool) is improved from Smith-Waterman algorithm. Although local algorithm has the desirable property that it is guaranteed to find the optimal local alignment, we decided to choose the global one because we regarded the segment sequence as a unit.
+
-
 
+
-
Sequences are aligned with different detailed methods in different situations. In the regulated side, what we care about is the DNA sequence. In the regulating side, it is the amino acid sequence. When it comes to predict the regulated behavior, we use a DNA substitution matrix to align promoter and protein coding sequences. In the prediction of regulating behavior, the substitution matrix BLOSUM_50 is used to align the amino acid sequences translated from protein coding sequences.
+
The promoter similarities of the query unit and subject units are stored in a vector. The protein coding similarities are stored in another vector. These vectors are prepared to be used in the new network construction.
The promoter similarities of the query unit and subject units are stored in a vector. The protein coding similarities are stored in another vector. These vectors are prepared to be used in the new network construction.
</p>           
</p>           
-
          </div>     
+
        </div>     
-
     </div><!--jobs container-->
+
     </div>
-
</div>
 
-
<div id="New_Network_Construction">
+
<h2 id="fa">Forward Analysis</h2>
 +
<div class="jobs_trigger" id="cng"><strong>Construct New GRN</strong></div>
 +
  <div class="jobs_item" style="display: none;">
 +
    <h3 id="ui">1 User Input</h3>
 +
    <p align="justify">
 +
      Some genes' regulation could be get from experiment. So, if users could get the unknow regulation between new gene and old ones, they could manually set the interactions which do not need model. Those regulations will be used in later simulation.
 +
    </p>
 +
    <h3>2 Simalarity Analysis</h3>
 +
    <p align="justify"><div id="sequence"><h4>2.1 Sequence</h4></div></br>
 +
      <div id="nwa"><h5>2.1.1 Needleman-Wunsch Algorithm</h5></div>
 +
      The Needleman-Wunsch algorithm was first published in1970 by Saul B. Needleman and Christian D. Wunsch. It performs a global alignment of two sequences and is mostly used in bioinformatics to align protein or nucleotide sequence. Our software applied this algorithm in the alignment of DNA and amino acid sequences.<br/><br/>
-
<h2>New Network Construction</h2>
+
      The Needleman-Wunsch algorithm is one kind of dynamic programming and It was the first attempt in biological sequence comparison of dynamic programming.<br/><br/>
 +
      Here is an example of Needleman-Wunsch algorithm. S(a,b) is the similarity of character a and character b. The scores of characters are shown in the similarity matrix. We assume this matrix was
 +
    </p>
 +
      <div align="center"><img src="https://static.igem.org/mediawiki/2013/5/52/USTC_Software_DNA_S_M.png"/></div>         
 +
      <p>And we uses linear gap penalty, denoted by d, here, we set the gap penalty as -5.Then the alignment:</p>
 +
      <p align="center"><strong><em>
 +
                          A: AGACTAGTTAC<br/>
 +
                          B: CGA - - - GACGT
 +
      </em></strong></p>
-
  <div id="jobs_container">
+
      <p>would have the following score:</p>
-
        <div class="jobs_trigger"><strong>Filter</strong></div>
+
      <p align="center"><strong><em>
-
<div class="jobs_item" style="display: none;"><p class="bodytext"></p><p align="justify">Once the similarity vectors are calculated, the next step is to filter them. As explained in the previous part, there is random noise in sequence alignment. In order to filter these meaningless values, a certain amount of random sequences are generated for each query-subject alignment. Normally, 100 is sufficient. Because the sequence length will influence alignment result, random sequences are fixed at the same length as the query one. Then align these random sequences with the subject sequence. The statistic result of these random similarities will be used as a threshold. If the original similarity is lower than the threshold, it is abandoned. In this case, the original value is usually short of statistical significance.
+
      S(A,C)+S(G,C)+S(A,A)+(3)+S(G,G)+S(T,A)+S(T,C)+S(A,G)+S(C,T) = -3+7+10-(3x5)+7+(-4)+0+(-1)+0 = 1
-
</p>
+
      </em></strong></p>
-
                </div>
+
-
+
-
<div class="jobs_trigger"><strong>Construct A New Regulated Vector</strong></div>
+
-
<div class="jobs_item" style="display: none;"><p align="justify">
+
-
If there are units in the original network having the same promoter as the exogenous one, the first step is to pick them out. Positive and negative regulations of these units are counted separately and distinguished into “positive group” and “negative group”. Then compare the exogenous one with these units. The similarities have already been calculated and stored in the corresponding positions in the similarity vector. The similarity mentioned here is the similarity of protein coding sequences as explained in the model part. The next step is to calculate the average similarity of each group. The exogenous unit is supposed to have the larger one’s direction(positive or negative). The weighted average regulation value of the chosen group whose weight is the sequence similarity is the new element’s value. It means regulatory intensity.
+
-
If there is no unit having the same promoter as the exogenous one, given that the promoter is the main regulatory region, the promoter similarity is used as the weight. And the weighted average of the regulation of the whole column is the new element’s value
+
      <p align="justify">To find the highest score of alignment, in this algorithm, a two dimensional matrix F with sequences and scores was allocated. The score in row i, column j is denoted by Fij. There is one column for each character in sequence A and one row for each character in sequence B. Therefore, if we align sequences with sizes of n and m, the amount of memory taken up here is O(n,m).<br/><br/>
-
</p>
+
-
                </div>
+
-
+
-
               
+
-
               
+
-
            <div class="jobs_trigger"> <strong>Construct A New Regulating Vector</strong></div>
+
-
        <div class="jobs_item" style="display: none;"><p align="justify">The construction of the new regulating vector is achieved in a way similar to the one described above. By calculated the weighted average regulation of a row, the program gives the regulatory intensity that the exogenous unit regulate the corresponding unit in the network.
+
-
</p>         </div>  
+
-
               
+
-
               
+
-
               
+
-
<div class="jobs_trigger"> <strong>A Supplementary Game: Test of The Model</strong></div>
+
-
<div class="jobs_item" style="display: block;"><p align="justify">                   
+
-
The behavior similarity of two units can be described by the dot product of two regulated vectors or two regulating vectors. A more intuitive way is using the vectorial angle to measured the similarity of two behaviors. But there are some zero vectors in the gene regulatory network which usually means the units either play the role of target or the regulator.
+
-
[Pic. 4 GRN matrix, target vector, regulator vector and their dot product]
+
      As the algorithm going on, Fij was calculated to be the optimal score by the principle as following:<br/>
 +
      Basis:
 +
      </p>
 +
      <p align="center"><strong><em>Fi0 = d*i<br/>F0j = d*j</em></strong></p>
 +
      <p>Recursion:</p>
 +
      <p align="center"><strong><em>Fij = max(F(i-1,j-1) + S(Ai,Bj), F(i-1,j) + d, F(i,j-1) + d)</em></strong></p><br/>
 +
      <p>The pseudo-code of this algorithm would look like this:</p>
 +
      <br/>
 +
      <div id="pseudo"><p>
 +
      <strong> for</strong> i = 0 <strong>to length(A)</strong><br/>
 +
      &nbsp;  F(i,0) <-- d*i<br/>
 +
      <strong> for</strong> j = 0 <strong>to length(B)</strong><br/>
 +
      &nbsp;  F(0,j) <-- d*j<br/>
 +
      <strong>for</strong> i = 0 <strong>to length(A)</strong><br/>
 +
      &nbsp;  <strong>for</strong> j = 0 <strong>to length(B)</strong><br/>
 +
      &nbsp; {<br/>
 +
      &nbsp; &nbsp;  Match  <-- F(i-1,j-1) + S(Ai,Bj)<br/>
 +
      &nbsp; &nbsp;  Delete <-- F(i-1,j) + d<br/>
 +
      &nbsp; &nbsp;  Insert <-- F(i,j-1) + d<br/>
 +
      &nbsp; &nbsp;  F(i,j) <-- <strong>max</strong>(Match, Insert, Delete)<br/>
 +
      &nbsp; }
 +
      </p>
 +
      </div>
-
We have tested the hypothesis by analyzing all 1748 regulation units of Escherichia coli, K-12, recorded in RegulonDB[FIXME: website link here]. By pairwise comparison of all these units, about 1.6 million sets of data was obtained. Each set of data consists of promoter sequence similarity, protein coding sequence similarity and behavior similarity. We hope to find some structure in the data that supports our hypothesis. And it is lucky enough to find there is a tendency showing the relationship between sequence similarity and behavior similarity(Pic. 2).
+
      <p align="justify">After the matrix F was computed, Fnm would be the maximum score among all possible alignment.<br/><br/>
-
[Pic. 2 Sequence similarity and behavior similarity]
+
      If you want to see the optimal alignment, you can trace back from Fnm by comparing three possible sources mentioned in the above code (Match, Insert and Delete). If Match, then Aj and Bi are aligned, if Insert, Bi was aligned with a gap and if Delete, then Aj and a gap are aligned. Also, you may find there are not only one optimal alignment.<br/><br/>
 +
      As for the example, we would get the following matrix by applying Needleman Wunsch algorithm:</p>
-
Sequence similarity is set as x axis and behavior similarity is set as y axis. Obviously sequence similarity is continuous-valued (from 0 to 1) and behavior similarity is discrete-valued. Values of behavior similarity determined by the dimension(N) of the vector are between -N and N. According to the result, promoter sequence similarity mainly distributes[FIXME] from 0.4 to 0.6, protein coding sequence similarity mainly distributes from 0 to 0.7 and behavior similarity mainly distributes from -3 to 5. As it is shown in Picture 4, high behavior similarity is partial to high sequence similarity. Peak value of behavior similarity, 17, appears where sequence similarity is 0.537. When behavior similarity value is fixed, for example, set behavior similarity as 8, it is obvious that the higher the sequence similarity is, the more intensive the dots are.
 
-
</p>
 
-
          </div>
+
      <div align="center"><img src="https://static.igem.org/mediawiki/2013/e/e2/USTC_Software_arrow_game.png"/></div>
-
  </div><!--jobs container-->
+
      <p>And the optimal alignment would be:</p>
-
</div>
+
 +
      <p align="center"><strong><em>- - AGACTAGTTAC <br/>
 +
                              CGAGAC - - GT - - -
 +
      </em></strong></p>
 +
      <div id="asg"><h5>2.1.2 A Supplementary Game</h5></div>
 +
      <p align="justify">The rows and columns in the GRN matrix can be regarded as vectors containing the regulated or the regulating information. The behavior similarity of two units can be described by the dot product of two regulated vectors or two regulating vectors. Biologists usually think the more similar two sequences are, the more likely they have similar behaviors. Whether the ratio of genes with similar behaviors is positively correlated with gene similarity is essential to our project. So we obtained 1.6 million sets of data by pairwise alignment of all the 1748 units in the GRN of K-12. Each set of data consists of gene similarity and behavior similarity. The result is analyzed and plotted in the figure. The linear fit shows that the ratio is positively correlated with the similarity.</p><br/>
-
<div id="Network_Model">
+
      <div align="center"><img style="width:600px; height:auto;"src="https://static.igem.org/mediawiki/2013/d/d0/USTC_Software_Simi-Ratio.png" />
-
<h2>Network Model</h2>
+
      <p><strong>Figure 4.</strong>Linear fit of ratio-similarity relationship.</p></div>
 +
      <p align="justify">Although there are examples that a slight change in DNA sequence will significantly change the property of the gene, for example, sickle-cell disease, the influence is usually determined by the location and scale of the mutation. So the result is still convincing to some degree.</p>
-
<div id="jobs_container">
+
    <div id="filtering"><h4>2.2 Filtering</h4></div>
 +
    <div id="rn"><h5>2.2.1 Random Noise</h5></div>
 +
    <p class="bodytext"></p><p align="justify">Normally, the similarity of two sequences will not be zero. Some computational
 +
experiments were carried out to study the random sequence similarities. We randomly
 +
chose a gene in the network and generated 1000 random sequences. The alignment result
 +
indicates that the random sequence similarities are Gauss distributed. The result suggests
 +
that some similarities are out of statistic significance.</p>
 +
<div align="center">
 +
<img src="https://static.igem.org/mediawiki/igem.org/8/89/USTC_Software_Figure_4.png" />
 +
<p><strong>Figure 5.</strong> Random similarity distribution</p></div>
 +
    <div id="filter"><h5>2.2.2 Filter</h5></div>
 +
    <p align="justify">We need the genes highly similar to the exogenous one to interact with it. The program will
 +
align the exogenous gene(query) with genes in the network(subject) and get the original
 +
similarities. In order to filter meaningless low values, a certain amount of random
 +
sequences are generated for each query-subject alignment. Normally, 100 is sufficient.
 +
Because the sequence length will influence alignment result, random sequences are fixed
 +
at the same length as the query one. Then align random sequences with the subject
 +
sequence. The statistic result of these random similarities is used as a threshold.<br />
 +
<div align="center">Threshold = μ + xσ</div><br />
 +
In the formula, μ is the average random similarity. σ is the standard deviation. x is used to
 +
control the filter determined by machine learning. If the original similarity is lower than the
 +
threshold, it is abandoned. It is usually means the original value is usually short of
 +
statistical significance.<br /><br />
 +
An example about filtring and consistency is presented in “Example”.
 +
</p>
 +
    <div id="rc"><h4>2.3 Regulation Calculation</h4></div>
 +
    <p align="justify">If there is a three-unit network and they interact with each other as it is shown in the figure.
 +
The regulation is described by the GRN matrix.</p>
 +
<div align="center"><img src="https://static.igem.org/mediawiki/igem.org/8/8a/USTC_Software_Figure_5.png" />
 +
<p align="center"><strong>Figure 6.</strong> Example network and its GRN matrix.</p></div>
-
               
 
-
 
-
<div class="jobs_trigger"><strong>Network Model Abstract</strong></div>
 
-
<div class="jobs_item" style="display: block;"><p align="justify">Network analysis includes finding stable condition of network, adding new gene, finding new stable condition and changes from original condition to new condition. We use densities of materials to describe network condition. If all material densities are time-invariant, we can say the network condition is stable.</p>
 
-
      </div>
 
-
 
+
<p align="justify">If D is the exogenous unit, we can obtain three similarity data sets of D with the units in the
-
        <div class="jobs_trigger"><strong>Hill Equations</strong></div>
+
original GRN:
-
<div class="jobs_item" style="display: none;"><p class="bodytext"></p><p align="justify">Regulation relationship in genetic network includes positive regulation, negative regulation, positive-or-negative regulation and no regulation. We store regulation relationship in matrix R. Rji means the unit in line j and row i. For the material of original network, Rji=1 means material i enhance material j, Rji=-1 means material i repress material j, Rji=0 means material i has no influence on material j, Rji=2 means material i enhance or repress material j. For the new material, Rji ranges from -1 to 1. Rji<0 means the possibility of positive regulation is Rji; Rji>0 means the possibility of negative regulation is –Rji; Rji=0 means there is no regulation from i to j.;
+
<li style="margin-left:40px;">Promoter sequence similarity</li>
 +
<li style="margin-left:40px;">Gene sequence similarity</li>
 +
<li style="margin-left:40px;">Amino acid sequence similarity.</li>
 +
<p>
 +
The construction is equivalent to add a new column and a row into the original matrix.</p>
 +
<div align="center"><img src="https://static.igem.org/mediawiki/igem.org/9/97/USTC_Software_Figure_6.png" />
 +
<p><strong>Figure 7.</strong> Mathematical Equivalence</p></div>
 +
<p align="justify">When filling the column, D is compared with the regulators of the unit in each row. The
 +
regulations in the row are consider separately and marked as “positive group” and
 +
“negative group”. The average similarity of each group represents the distance between
 +
the exogenous unit and the group. D is supposed to have the larger one's regulatory
 +
direction(positive or negative). The regulatory intensity is the weight average regulation of
 +
the chose group. The weight here is the amino acid sequence similarity.<br /><br />
 +
There are two conditions when fill the new row:<br />
 +
1. There are units having the same promoter as the exogenous unit.<br />
 +
2. There is no units having the same promoter as the exogenous unit.<br /><br />
 +
In condition 1, the units sharing the same promoter with the new member are picked out,
 +
and the following steps are the same as the construction of the column. The difference is
 +
the similarity used here is the gene sequence similarity. As explained in the regulation
 +
model part, the promoter is the main regulatory region, but the following sequence is also
 +
considered. Now the promoter is the same, so what we focus on are the gene sequences.<br /><br />
 +
In condition 2, the process is almost the same as constructing the new column. Promoter
 +
similarity is used because it is the main region.</p>
 +
<div align="center">
 +
<img src="https://static.igem.org/mediawiki/igem.org/c/c5/USTC_Software_Figure_7.png" />
 +
</div>
 +
<p align="center"><strong>Figure 8.</strong> Construct New GRN</p>
 +
    <h3>3 Clustering</h3>
 +
    <p>
 +
      Cluster analysis or clustering is the task of grouping a set of objects in such a way that objects in the same group (called a cluster) are more similar (in some sense or another) to each other than to those in other groups (clusters). It is a main task of exploratory data mining, and a common technique for statistical data analysis, used in many fields, including machine learning, pattern recognition, image analysis, information retrieval, and bioinformatics.</br></br>
 +
      For get a better regulation, we use online database DAVID to cluster all the genes in our whole GRN. Avoid of supersoftless, we hope to create an online communication with DAVID. After getting the cluster of our genes, we multiply the genes simalarity with a factor if they are in the same cluster.</br></br>
 +
      Though the source code of this part has already done, we lack the experiment information to set a propriate factor. All source code were pushed up to our github.
 +
    </p>
 +
  </div>
 +
<div class="jobs_trigger" id="nm"><strong>Network Model</strong></div>
 +
  <div class="jobs_item" style="display: none;">
 +
<p align="justify">Network analysis includes finding stable condition of network, adding new gene, finding new stable condition and changes from original condition to new condition. We use densities of materials to describe network condition. If all material densities are time-invariant, we can say the network condition is stable.</p>
 +
<p class="bodytext"></p><p align="justify">Regulation relationship in genetic network includes positive regulation, negative regulation, positive-or-negative regulation and no regulation. We store regulation relationship in matrix R. Rji means the unit in line j and row i. For the material of original network, Rji=1 means material i enhance material j, Rji=-1 means material i repress material j, Rji=0 means material i has no influence on material j, Rji=2 means material i enhance or repress material j. For the new material, Rji ranges from -1 to 1. Rji<0 means the possibility of positive regulation is Rji; Rji>0 means the possibility of negative regulation is –Rji; Rji=0 means there is no regulation from i to j.
We use Hill equations to describe intensity of regulation. Equations are like following:
We use Hill equations to describe intensity of regulation. Equations are like following:
-
<br/>
+
<br/></br>
<img src="https://static.igem.org/mediawiki/2013/e/e0/USTC_Software_1.png" style="width:600px;"/>
<img src="https://static.igem.org/mediawiki/2013/e/e0/USTC_Software_1.png" style="width:600px;"/>
-
<br/>
+
<br/></br>
-
The left side of the equation is the derivative x(density) on t(time).”qi”,”pi”,”ri”,”mi”,”ni” are parameters, which determine the intensity of regulation."ri" is degradation rate. Mji is exponent. M is a matrix whose dimensions are equivalent to R's. Mji is 0 or ranges from 0.5 to 1.2 or ranges from -1.2 to 0.5. For the material of original network, if Rji=1,Mji ranges from 0.5 to 1.2;if Rji=-1, Mji ranges from -1.2 to -0.5; if Rji=2;Mji ranges from -1.2 to 0.5 or 0.5 to 1. These Mjis’ absolute values are given randomly by program. If Rji=0, Mji=0. For the new material,
+
The left side of the equation is the derivative x(density) on t(time).”qi”,”pi”,”ri”,”mi”,”ni” are parameters, which determine the intensity of regulation."ri" is degradation rate. Mji is exponent. M is a matrix whose dimensions are equivalent to R's. Mji is 0 or ranges from 0.5 to 1.2 or ranges from -1.2 to 0.5. For the material of original network, if Rji=1,Mji ranges from 0.5 to 1.2;if Rji=-1, Mji ranges from -1.2 to -0.5; if Rji=2;Mji ranges from -1.2 to 0.5 or 0.5 to 1. These Mjis' absolute values are given randomly by program. If Rji=0, Mji=0.  
-
<br/>
+
</br>For the new material,
 +
<br/></br>
<img src="https://static.igem.org/mediawiki/2013/6/64/USTC_Software_2.png"/>
<img src="https://static.igem.org/mediawiki/2013/6/64/USTC_Software_2.png"/>
-
<br/>
+
<br/></br>
</p>
</p>
-
                </div>
+
<p align="justify">
-
+
Stable condition is the condition in which densities are time-invariant. We store material densities in a vector and solve the differential equations with Euler's formula, which is like below
-
<div class="jobs_trigger"><strong>Find Stable Network Condition</strong></div>
+
<br/></br>
-
<div class="jobs_item" style="display: none;"><p align="justify">
+
-
Stable condition is the condition in which densities are time-invariant. We store material densities in a vector and solve the differential equations with Euler’s formula, which is like below
+
-
<br/>
+
<img src="https://static.igem.org/mediawiki/2013/e/e6/USTC_Software_3.png" style="width:600px;"/>
<img src="https://static.igem.org/mediawiki/2013/e/e6/USTC_Software_3.png" style="width:600px;"/>
-
<br/>
+
<br/></br>
We know the network will be stable at last, so every material density has a limitation.
We know the network will be stable at last, so every material density has a limitation.
</p>
</p>
-
                </div>
+
 
-
+
 
-
               
+
 
-
               
+
  </div>
-
            <div class="jobs_trigger"> <strong>Find Changes From Original Stable Condition to New Condition</strong></div>
+
<div class="jobs_trigger" id="en"><strong>Evaluate Network</strong></div>
-
        <div class="jobs_item" style="display: none;"><p align="justify">Record the original stable condition, set new material density to 0 and this is the new initial density vector. Solve new equations and record density vectors before the new condition is stable and store these data in a text file.
+
  <div class="jobs_item" style="display: none;">
 +
<p align="justify">Record the original stable condition, set new material density to 0 and this is the new initial density vector. Solve new equations and record density vectors before the new condition is stable and store these data in a text file.</br></br>
To evaluate the new network, we introduce the grading system.
To evaluate the new network, we introduce the grading system.
-
<br/>
+
<br/></br>
<img src="https://static.igem.org/mediawiki/2013/3/32/USTC_Software_4.png" style="width:600px;"/>
<img src="https://static.igem.org/mediawiki/2013/3/32/USTC_Software_4.png" style="width:600px;"/>
-
<img src="https://static.igem.org/mediawiki/2013/b/bc/USTC_Software_5.png" style="width:600px;"/>
+
<img src="https://static.igem.org/mediawiki/2013/b/bc/USTC_Software_5.png" style="width:500px;"/>
-
<br/>
+
<br/></br>
-
"xi" and "Xi" are densities of material i, which is not the new material."ny" is the number of materials. The more new densities are close to the original, the less the influence the cell endues. In general, cells close to the original cell are more likely to survive than those who are far different from the original cell. That is the thought of the grading system.
+
"xi" and "Xi" are densities of material i, which is not the new material."ny" is the number of materials. The more new densities are close to the original, the less the influence the cell endues. In general, cells close to the original cell are more likely to survive than those who are far different from the original cell. That is the thought of the grading system.</br></br>
We did a lot of running and found that the “AbsValue” ranges from 0 to 370, so "ScoreA" ranges from 0 to 4.9.We get the integer part and store it in an array, which has five sections. Generate 100 or 200 matrix M from matrix R and run the original and new network for each M, so we can get 100 or 200 of "ScoreA"s. The section which has maximum "ScoreA"s is the eventual score.
We did a lot of running and found that the “AbsValue” ranges from 0 to 370, so "ScoreA" ranges from 0 to 4.9.We get the integer part and store it in an array, which has five sections. Generate 100 or 200 matrix M from matrix R and run the original and new network for each M, so we can get 100 or 200 of "ScoreA"s. The section which has maximum "ScoreA"s is the eventual score.
</p>           
</p>           
-
            </div>  
+
  </div>
-
 
-
 
-
 
-
    </div><!--jobs container-->
 
 +
<h2 id="ra">Reverse Analysis</h2>
 +
<div class="jobs_trigger" id="vg"><strong>Virtual Gene</strong></div>
 +
  <div class="jobs_item" style="display: none;">
 +
<p align="justify">Before reverse analysis, we use the same idea about constructing a new GRN. So we create a virtual gene which replace the gene what users want to get. In calculation, it means that we add a row and a column to the matrix of GRN.</p>
</div>
</div>
-
<div id="Predict">
+
<div class="jobs_trigger" id="er"><strong>Expression Range</strong></div>
-
<h2>Predict</h2>
+
  <div class="jobs_item" style="display: none;">
 +
<p align="justify">Before prediction, the expression of specific genes which the experimenter needs should be input into our software as well as the improvement or depression. The number of target gene is SIX at most.</br></br>
 +
It is a must that figuring out the strongest and weakest expression strength before inputting the extreme cases into the target expression. The way to find out the strongest and weakest expression is modeling the GRN's steady state by a large amount of random regulation from -1 and 1. We ran it for 1000 times to get the range of gene expression. On the other hand, the expression of genes unpicked by the users should be stable as much as possible. The initial strength of expression is calculated by modeling the original GRN with Hill's equation.
 +
</p>
 +
</div>
 +
<div class="jobs_trigger" id="pso"><strong>Particle Swarm Optimaztion</strong></div>
 +
  <div class="jobs_item" style="display: none;">
 +
<p align="justify">
 +
For getting the best regulation, our software uses PSO algorithm based on 30 particles to simulate the GRN's changing. First of all, the interactions of regulator and regulated-by have been put into those particles in random so that each particle will have the whole set of regulation. Secondly, the variance between target expressions and stable expression of new GRN have been regarded as the optimize requirements in PSO algorithm. As a result, the minimal variance of 30 particles is the global optimum and the minimal variance of the procession in one particle is the local optimum. Then, taking a step towards global and local optimum as well as considering the inertia and perturbation avoids falling into the sub-optimal condition.</br></br>
 +
At last, when the variance of expression reaches an acceptable range, our software picks out and saves the best global optimum particle following by the movement of those particles stop.</br></br>
 +
We constantly revises the factors in PSO algorithm by machine learning method for accurate simulation with a fast PSO particle-motion equation. At the same time, our software also filter the result of regulatory value which is more intuitive.
 +
</p>
 +
</div>
-
<div id="jobs_container">
 
-
               
+
<div class="jobs_trigger" id="lot"><strong>Locate Optimal Target</strong></div>
-
+
  <div class="jobs_item" style="display: none;">
-
<div class="jobs_trigger"><strong>Predict Abstract</strong></div>
+
<p align="justify">To improve the efficiency of choosing a suitable gene after getting a series of regulatory value, our software picks out some obvious regulation. The value of regulation is between -1 to 1 in which -1 means negative effect and 1 means positive effect. As a result, what our software has done is filtering out the absolute value which is lower than 0.9. Because the difference of regulatory intensity lower than 0.1 has very little effect to the stable expression, the final result of regulation is indicated by Boolean value.</br></br>
-
<div class="jobs_item" style="display: block;"><p align="justify">In some cases, importing exogenous gene is for enhancing or suppressing the expression of some specific genes in engineered bacteria itself. But it is hard to choose an appropriate regulatory gene. Our software analyzes the GRN forward as well as simulates by optimization algorithm backward for giving a reference of choosing to the users. Our software not only focused on the direct regulation but also focused on the global GRN. In the same time, controlling the expression of multiple genes in network has been realized by global prediction. What’s more, Particle Swarm Optimization (PSO) Algorithm makes it possible.</p>
+
The format of regulatory prediction in “Result”:</br>
-
      </div>
+
Gene_name->Gene_name    regulation(+/-)
 +
</p>        
 +
</div>
 +
<h2 id="reference">Reference</h2>
-
        <div class="jobs_trigger"><strong>Input Target</strong></div>
+
<p align="justify">
-
<div class="jobs_item" style="display: none;"><p class="bodytext"></p><p align="justify">Before prediction, the expression of specific genes which the experimenter needs should be input into our software as well as the improvement or depression. The number of target gene is SEVEN at most.
+
-
It is a must that figuring out the strongest and weakest expression strength before inputting the extreme cases into the target expression. The way to find out the strongest and weakest expression is modeling the GRN’s steady state by a large amount of random regulation from -1 and 1. On the other hand, the expression of genes unpicked by the users should be stable as much as possible. The initial strength of expression is calculated by modeling the original GRN with Hill’s equation.
+
-
</p>
+
-
                </div>
+
-
+
-
<div class="jobs_trigger"><strong>Particle Swarm Optimization</strong></div>
+
-
<div class="jobs_item" style="display: none;"><p align="justify">
+
-
For getting the best regulation, our software uses PSO algorithm based on 30 particles to simulate the GRN’s changing. First of all, the interactions of regulator and regulated-by have been put into those particles in random so that each particle will have the whole set of regulation. Secondly, the variance between target expressions and stable expression of new GRN have been regarded as the optimize requirements in PSO algorithm. As a result, the minimal variance of 30 particles is the global optimum and the minimal variance of the procession in one particle is the local optimum. Then, taking a step towards global and local optimum as well as considering the inertia and perturbation avoids falling into the sub-optimal condition.
+
-
At last, when the variance of expression reaches an acceptable range, our software picks out and saves the best global optimum particle following by the movement of those particles stop.
+
-
We constantly revises the factors in PSO algorithm by machine learning method for accurate simulation with a fast PSO particle-motion equation. At the same time, our software also filter the result of regulatory value which is more intuitive.
+
-
</p>
+
-
                </div>
+
-
+
-
               
+
-
               
+
-
            <div class="jobs_trigger"> <strong>Filter</strong></div>
+
-
        <div class="jobs_item" style="display: none;"><p align="justify">To improve the efficiency of choosing a suitable gene after getting a series of regulatory value, our software picks out some obvious regulation. The value of regulation is between -1 to 1 in which -1 means negative effect and 1 means positive effect. As a result, what our software has done is filtering out the absolute value which is lower than 0.9. Because the difference of regulatory intensity lower than 0.1 has very little effect to the stable expression, the final result of regulation is indicated by Boolean value.
+
-
The format of regulatory prediction in “Result”:
+
-
Gene_name->Gene_name    regulation(+/-)
+
-
</p>        
+
Lei Z, Dai Y. Assessing protein similarity with Gene Ontology and its use in subnuclear localization prediction[J]. BMC bioinformatics, 2006, 7(1): 491.</br></br>
-
            </div>  
+
-
 
-
 
-
 
-
    </div><!--jobs container-->
 
 +
Ramoni M F, Sebastiani P, Kohane I S. Cluster analysis of gene expression dynamics[J]. Proceedings of the National Academy of Sciences, 2002, 99(14): 9121-9126.</br></br>
 +
Thieffry D, Huerta A M, Pérez‐Rueda E, et al. From specific gene regulation to genomic networks: a global analysis of transcriptional regulation in Escherichia coli[J]. Bioessays, 1998, 20(5): 433-440.</br></br>
 +
 +
 +
Eberhart R, Kennedy J. A new optimizer using particle swarm theory[C]//Micro Machine and Human Science, 1995. MHS'95., Proceedings of the Sixth International Symposium on. IEEE, 1995: 39-43.</br></br>
-
</div>
 
-
</div>
+
Jacob F, Perrin D, Sánchez C, et al. L'opéron: groupe de gènes à expression coordonnée par un opérateur [CR Acad. Sci. Paris 250 (1960) 1727–1729][J]. Comptes rendus biologies, 2005, 328(6): 514-520.</br></br>
 +
 
 +
 
 +
Needleman S B, Wunsch C D. A general method applicable to the search for similarities in the amino acid sequence of two proteins[J]. Journal of molecular biology, 1970, 48(3): 443-453.</br></br>
 +
 
 +
 
 +
Gama-Castro S, Jiménez-Jacinto V, Peralta-Gil M, et al. RegulonDB (version 6.0): gene regulation model of Escherichia coli K-12 beyond transcription, active (experimental) annotated promoters and Textpresso navigation[J]. Nucleic acids research, 2008, 36(suppl 1): D120-D124.</br></br>
 +
 
 +
Martınez-Antonio A, Collado-Vides J. Identifying global regulators in transcriptional regulatory networks in bacteria[J]. Current opinion in microbiology, 2003, 6(5): 482-489.</br></br>
 +
 
 +
 
 +
Salgado H, Moreno-Hagelsieb G, Smith T F, et al. Operons in Escherichia coli: genomic analyses and predictions[J]. Proceedings of the National Academy of Sciences, 2000, 97(12): 6652-6657.</br></br>
 +
 
 +
 
 +
Thieffry D, Salgado H, Huerta A M, et al. Prediction of transcriptional regulatory sites in the complete genome sequence of Escherichia coli K-12[J]. Bioinformatics, 1998, 14(5): 391-400.
 +
 
 +
</p>
 +
 
 +
<div class="jobs_trigger" style="display:none;"></div>
 +
<div class="jobs_item" style="display: none;"><p></p></div>
</body>
</body>
</html>
</html>

Latest revision as of 00:46, 29 October 2013

Header2


Methodologies

Methodologies

In order to simulate the GRN's working and analyze the changing after exogenous gene imported, some advanced algorithms and classical methods are employed in the software. These algorithms and methods include Binary Tree method, Needle-Wunsch Algorithm, Decision Tree method, Hill Equation and PSO Algorithm.


There are four parts of methodologies: Database, Operon Theory and Regulatory Model, Forward Analysis and Reverse Analysis.

Database

Abstract
Fetch Regulation
Fetch Gene Info
Fetch Promoter Info
Integration

Our software integrates all information we picked out about genes and generates a file named “all_info” —— all information about genes —— for the output graphical interface's reading. In the meanwhile, the array of objects containing all information has been stored in computer memory which greatly improve the computing speed of our software.

  The format of all_info database:
    No.    promoter_sequence    gene_sequence    gene_name    ID    left_position    right_position    promoter_name     description
The fetching module generates three files: old_GRN, all_info and uncertain_database.

Operon Theory and Regulatory Model

Operon Theory
Regulatory Model
Similarity and Homology

Forward Analysis

Construct New GRN
Network Model
Evaluate Network

Reverse Analysis

Virtual Gene
Expression Range
Particle Swarm Optimaztion
Locate Optimal Target

Reference

Lei Z, Dai Y. Assessing protein similarity with Gene Ontology and its use in subnuclear localization prediction[J]. BMC bioinformatics, 2006, 7(1): 491.

Ramoni M F, Sebastiani P, Kohane I S. Cluster analysis of gene expression dynamics[J]. Proceedings of the National Academy of Sciences, 2002, 99(14): 9121-9126.

Thieffry D, Huerta A M, Pérez‐Rueda E, et al. From specific gene regulation to genomic networks: a global analysis of transcriptional regulation in Escherichia coli[J]. Bioessays, 1998, 20(5): 433-440.

Eberhart R, Kennedy J. A new optimizer using particle swarm theory[C]//Micro Machine and Human Science, 1995. MHS'95., Proceedings of the Sixth International Symposium on. IEEE, 1995: 39-43.

Jacob F, Perrin D, Sánchez C, et al. L'opéron: groupe de gènes à expression coordonnée par un opérateur [CR Acad. Sci. Paris 250 (1960) 1727–1729][J]. Comptes rendus biologies, 2005, 328(6): 514-520.

Needleman S B, Wunsch C D. A general method applicable to the search for similarities in the amino acid sequence of two proteins[J]. Journal of molecular biology, 1970, 48(3): 443-453.

Gama-Castro S, Jiménez-Jacinto V, Peralta-Gil M, et al. RegulonDB (version 6.0): gene regulation model of Escherichia coli K-12 beyond transcription, active (experimental) annotated promoters and Textpresso navigation[J]. Nucleic acids research, 2008, 36(suppl 1): D120-D124.

Martınez-Antonio A, Collado-Vides J. Identifying global regulators in transcriptional regulatory networks in bacteria[J]. Current opinion in microbiology, 2003, 6(5): 482-489.

Salgado H, Moreno-Hagelsieb G, Smith T F, et al. Operons in Escherichia coli: genomic analyses and predictions[J]. Proceedings of the National Academy of Sciences, 2000, 97(12): 6652-6657.

Thieffry D, Salgado H, Huerta A M, et al. Prediction of transcriptional regulatory sites in the complete genome sequence of Escherichia coli K-12[J]. Bioinformatics, 1998, 14(5): 391-400.