Difference between revisions of "Team:SEU/Description"

 
(33 intermediate revisions by 4 users not shown)
Line 1: Line 1:
 +
{{SEU/Header}}
 
<html>
 
<html>
 +
    <meta charset="utf-8">
 +
    <link href="https://2019.igem.org/Template:SEU/Home_CSS?action=raw&ctype=text/css" rel="stylesheet" type="text/css">
 +
    <link href="https://2019.igem.org/Template:SEU/Home_CSS2?action=raw&ctype=text/css" rel="stylesheet" type="text/css">
 +
    <div id="fullpage" class="fullpage-default">
 +
        <div>
 +
            <div class="section-inner">
 +
                <div class="about-section">
 +
                    <div class="row justify-content-center">
 +
                        <div class="col-md-6">
 +
                            <div class="row">
 +
                                <div class="col-md-6.col-md-offset-3">
 +
                                    <br><br><br><br>
 +
                                        <h2>Project Description</h2>
 +
                                        <h3>Motivation</h3>
 +
                                        <p><font size="4">Artificial intelligence is one prevailing research field in recent years due to its successful application in areas like medical care, computer vision and natural language processing. Most of artificial intelligence systems are implemented with silicon-based chips which have high area complexity and power cost when facing large systems. Is it possible to use bio-chemical materials to implement artificial intelligence systems more efficiently?</font></p>
 +
                                        <h3>Preliminaries</h3>
 +
                                        <p><font size="4">DNA strands have been proved a powerful medium to perform computation. Previous researches [1], [2] have shown some interesting applicatoins of such materials, which implemented a "probabilistic switch" and a pattern recognition machine, respectively. Computation systems in this field are usually based on DNA strand displacement reactions. [3] and [4] provide vivid demonstration of how such reactions proceed in molecular level.</font></p>
 +
                                        <h3>Our Work</h3>
 +
                                        <center>
 +
                                      <img src="https://static.igem.org/mediawiki/2019/f/f3/T--SEU--description.png">
 +
                                        </center>
 +
                                        <p></p>
 +
                                        <p><font size="4">In this project, we have employed DNA reactions to implement neural networks which are basic components in artificial intelligence systems. Specifically, using molecular concentrations as input and output signals, we propose and demonstrate computation models for weighted summation (including multiplication and addition) and activation functions (e.g, ReLU and Sigmoid) which are arithmetic operations in neural networks. Also, the classical backpropagation algorithm which is utilized in the training process can be similarly constructed (which has not been studied in molecular computing to our best knowledge). Based on these models, we synthesize basic neural networks with DNA strand displacement reactions.</font></p>
  
−
<!-- 解决官网bug的 CSS -->
+
                                        <p><font size="4">To help other researchers obtain the proposed DNA-based artificial neural networks, we develop a software tool to automatically generate DNA reaction models and relevant DNA sequences according to the required parameters of neural networks, which may help them with further bio-computer design and bio-robot design. </font></p>
−
<style>
+
−
        /*消除官网坑爹样式*/
+
−
        body {
+
−
            margin-top:-6px;
+
−
            height: 0%;
+
−
            width: 100%;
+
−
        }
+
−
        #HQ_page p {
+
−
            font-size:16px;
+
−
            font-family:"微软雅黑";
+
−
        }
+
−
        #globalWrapper{
+
−
            margin: 0;
+
−
            padding: 0;
+
−
        }
+
−
        #content{
+
−
            width: 100%;
+
−
            padding: 0;
+
−
            margin: -12px 0 0;
+
−
        }
+
−
        #top_title{
+
−
            display: none;
+
−
        }
+
−
        #sideMenu{
+
−
            display: none;
+
−
        }
+
−
      #content,
+
−
      .vs-content,
+
−
      .vs-wrapper section {
+
−
        background-color: transparent !important;
+
−
      }
+
  
−
</style>
+
                                        <p><font size="4">For more details, please browse our <b> <a href="https://2019.igem.org/Team:SEU/Contribution">Contribution Page.</a></b></font></p>
−
    <!-- Mobile Specific Metas================================================== -->
+
                                        <h3>References</h3>
−
<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1">
+
                                        <p><font size="4">[1] Wilhelm, Daniel, Jehoshua Bruck, and Lulu Qian. "Probabilistic switching circuits in DNA." Proceedings of the National Academy of Sciences 115.5 (2018): 903-908.</font></p>
−
 
+
                                        <p></p>
−
    <!-- CSS==================================================================== -->
+
                                        <p><font size="4">[2] Cherry, Kevin M., and Lulu Qian. "Scaling up molecular pattern recognition with DNA-based winner-take-all neural networks." Nature 559.7714 (2018): 370.</font></p>
−
  <link rel="stylesheet" href="https://2019.igem.org/Template:SEU/CSS?action=raw&ctype=text/css"/>
+
                                        <p></p>
−
        <link rel="stylesheet" href="https://2019.igem.org/Template:SEU/STYLE?action=raw&ctype=text/css"/>
+
                                        <p><font size="4">[3] <a href="https://www.youtube.com/watch?v=zjC_0PW9A4c&feature=youtu.be&tdsourcetag=s_pctim_aiomsg">DNA Strand Displacement.</a></font></p>
−
 
+
                                        <p></p>
−
    <!-- Custom Fonts=========================================================== -->
+
                                        <p><font size="4">[4] <a href="https://www.youtube.com/watch?v=42FCzoJt8Pg&feature=youtu.be">DNA Join Circuit.</a></font></p>
−
        <link href="https://2019.igem.org/Template:SEU/Font?action=raw&ctype=text/css" rel="stylesheet" type="text/css">
+
                                        <p></p>
−
<script src="https://2019.igem.org/Template:SEU/JS?action=raw&ctype=text/css" type="text/javascript"></script>
+
                                </div>
−
 
+
                            </div>
−
    <!-- Style_CSS============================================================== -->
+
                        </div>
−
<div class="wrap-body">
+
                    </div>
−
<div class="header">
+
                </div>
−
<div id='cssmenu' >
+
−
<ul>
+
−
  <li class="has-sub"><a href='https://2019.igem.org/Team:SEU'><span>Home</span></a></li>
+
−
  <li class=' has-sub'><a href='https://2019.igem.org/Team:SEU/Team'><span>TEAM</span></a>
+
−
  <ul>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Team'><span>Team Members</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Collaborations'><span>Collaborations</span></a>
+
−
</li>
+
−
  </ul>
+
−
  </li>
+
−
<li class='active'><a href='https://2019.igem.org/Team:SEU/Description'><span>PROJECT</span></a>
+
−
<ul>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Description'><span>Description</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Design'><span>Design</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Experiments'><span>Experiment</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Notebook'><span>Notebook</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Contribution'><span>Contribution</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Results'><span>Results</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Demonstrate'><span>Demonstrate</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Improve'><span>Improve</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Attributions'><span>Attributions</span></a>
+
−
</li>
+
−
</ul>
+
−
</li>
+
−
  <li class=' has-sub'><a href='https://2019.igem.org/Team:SEU/Parts'><span>PARTS</span></a>
+
−
<ul>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Parts'><span>Parts Overview</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Basic_Part'><span>Basic Parts</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Composite_Part'><span>Composite Parts</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Part_Collection'><span>Part Collection</span></a>
+
−
</li>
+
−
</ul>
+
−
</li>
+
−
  <li><a href='https://2019.igem.org/Team:SEU/Safety'><span>SAFETY</span></a></li>
+
−
  <li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Human_Practices'><span>HUMAN PRACTICES</span></a>
+
−
<ul>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Human_Practices'><span>Human Practices</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Public_Engagement'><span>Education & Engagement</span></a>
+
−
</li>
+
−
</ul>
+
−
</li>
+
−
<li class=' has-sub'><a href='https://2019.igem.org/Team:SEU/Entrepreneurship'><span>AWARDS</span></a>
+
−
<ul>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Entrepreneurship'><span>Entrepreneurship</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Hardware'><span>Hardware</span></a>
+
−
</li>
+
−
<li class=' has-sub'><a href='https://2019.igem.org/Team:SEU/Measurement'><span>Measurement</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Model'><span>Model</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Plant'><span>Plant</span></a>
+
−
</li>
+
−
<li class='has-sub'><a href='https://2019.igem.org/Team:SEU/Software'><span>Software</span></a>
+
−
</li>
+
−
</ul>
+
−
</li>
+
−
<li><a href="https://igem.org/2019_Judging_Form?team=SEU"><span>JUDGING FORM</span></a>
+
−
</ul
+
−
></div>
+
−
</div>
+
−
  <section id="container">
+
−
    <div class="wrap-container clearfix">
+
−
      <div id="main-content">
+
−
        <div class="wrap-content">
+
−
          <article>
+
−
            <div class="art-header">
+
−
              <h1 class="entry-title">Project Description</h1>
+
−
            </div>
+
−
            <div class="art-content">
+
−
 
+
−
              <h2>Description</h2>
+
−
              <p>This project is based on one of our previously published article [1]. Artificial intelligence is one prevailing research field in recent years, but most of the implementations are on traditional silicon-based computers or chips. Is it possible to use biochemical materials to implement such systems? Our previous paper provides one possible method, but it is validated by only simulations. In this project, we aim to implement such a system in wet experiments. Also, to aid the design of such systems, we will develop a small software to automatically generate required DNA topological structures. </p>
+
−
              <p></p>
+
−
              <p>In our system, the concentrations of some input DNA species will be regarded as the input to the neural network. Some mathematical calculations are performed in solutions (weighted summation, activation, etc.) and the output of the neural network is the concentration of some certain DNA strands, similarly. </p>
+
−
              <p></p>
+
−
              <p>There are various possible applications of this technology. For example, as it utilizes only DNA, a type of bio-friendly material, with small modifications it may be integrated to other biosystems to create biochemistry robots.</p>
+
−
 
+
−
              <h2>Preliminaries</h2>
+
−
              <p>DNA strands have been proved a powerful medium to perform computation. Previous researches [2], [3] have shown some interesting applicatoins of such materials, which implemented a "probabilistic switch" and a pattern recognition machine, respectively.</p>
+
−
              <p></p>
+
−
              <p>In this project, we plan to utilize a similar approach to conduct our experiment, implement a neural network using biochemical materials and validate our previous theory.</p>
+
−
 
+
−
              <h2>References</h2>
+
−
              <p>[1]C. Fang, Z. Shen, Z. Zhang, X. You and C. Zhang, "Synthesizing a Neuron Using Chemical Reactions," 2018 IEEE International Workshop on Signal Processing Systems (SiPS), Cape Town, 2018, pp. 187-192.</p>
+
−
              <p></p>
+
−
              <p>[2]Wilhelm, Daniel, Jehoshua Bruck, and Lulu Qian. "Probabilistic switching circuits in DNA." Proceedings of the National Academy of Sciences 115.5 (2018): 903-908.</p>
+
−
              <p></p>
+
−
              <p>[3]Cherry, Kevin M., and Lulu Qian. "Scaling up molecular pattern recognition with DNA-based winner-take-all neural networks." Nature 559.7714 (2018): 370.</p>
+
−
              <p></p>
+
 
             </div>
 
             </div>
−
          </article>
 
 
         </div>
 
         </div>
−
      </div>
 
−
    </div>
 
−
  </section>
 
−
<hr class="line">
 
−
<script src="https://2019.igem.org/Template:SEU/JS?action=raw&ctype=text/css" type="text/javascript"></script>
 
−
</div>
 
 
</html>
 
</html>

Latest revision as of 06:41, 19 October 2019









Project Description

Motivation

Artificial intelligence is one prevailing research field in recent years due to its successful application in areas like medical care, computer vision and natural language processing. Most of artificial intelligence systems are implemented with silicon-based chips which have high area complexity and power cost when facing large systems. Is it possible to use bio-chemical materials to implement artificial intelligence systems more efficiently?

Preliminaries

DNA strands have been proved a powerful medium to perform computation. Previous researches [1], [2] have shown some interesting applicatoins of such materials, which implemented a "probabilistic switch" and a pattern recognition machine, respectively. Computation systems in this field are usually based on DNA strand displacement reactions. [3] and [4] provide vivid demonstration of how such reactions proceed in molecular level.

Our Work

In this project, we have employed DNA reactions to implement neural networks which are basic components in artificial intelligence systems. Specifically, using molecular concentrations as input and output signals, we propose and demonstrate computation models for weighted summation (including multiplication and addition) and activation functions (e.g, ReLU and Sigmoid) which are arithmetic operations in neural networks. Also, the classical backpropagation algorithm which is utilized in the training process can be similarly constructed (which has not been studied in molecular computing to our best knowledge). Based on these models, we synthesize basic neural networks with DNA strand displacement reactions.

To help other researchers obtain the proposed DNA-based artificial neural networks, we develop a software tool to automatically generate DNA reaction models and relevant DNA sequences according to the required parameters of neural networks, which may help them with further bio-computer design and bio-robot design.

For more details, please browse our Contribution Page.

References

[1] Wilhelm, Daniel, Jehoshua Bruck, and Lulu Qian. "Probabilistic switching circuits in DNA." Proceedings of the National Academy of Sciences 115.5 (2018): 903-908.

[2] Cherry, Kevin M., and Lulu Qian. "Scaling up molecular pattern recognition with DNA-based winner-take-all neural networks." Nature 559.7714 (2018): 370.

[3] DNA Strand Displacement.

[4] DNA Join Circuit.