Difference between revisions of "Team:SEU/Contribution"
| Line 29: | Line 29: | ||
<p style="font-size=36px">Through the comprehensive use of life science and information science knowledge, we have obtained the results of this experiment. In the whole process, we encountered many difficulties and challenges, but after careful thinking and practice, we finally successfully overcome these. In addition, we have also summarized some information that may be helpful to other teams, hoping to make some contributions to the iGEM community. </p> | <p style="font-size=36px">Through the comprehensive use of life science and information science knowledge, we have obtained the results of this experiment. In the whole process, we encountered many difficulties and challenges, but after careful thinking and practice, we finally successfully overcome these. In addition, we have also summarized some information that may be helpful to other teams, hoping to make some contributions to the iGEM community. </p> | ||
| − | <p style="font-size=36px">1.We propose molecular computation models for arithmetic operations in artificial neural networks as well as relevant reaction kinetic analysis. </p> | + | <p style="font-size=36px">1. We propose molecular computation models for arithmetic operations in artificial neural networks as well as relevant reaction kinetic analysis. </p> |
<center><a href="https://2019.igem.org/Team:SEU/Model" class="buttonContri">Model</a></center> | <center><a href="https://2019.igem.org/Team:SEU/Model" class="buttonContri">Model</a></center> | ||
| − | <p style="font-size=36px">2.Based on these models, we implement artificial neurons, which is the basic element of neural networks, with DNA reactions. Also, we achieve backpropagation training process for DNA-based neural networks.</p> | + | <p style="font-size=36px">2. Based on these models, we implement artificial neurons, which is the basic element of neural networks, with DNA reactions. Also, we achieve backpropagation training process for DNA-based neural networks.</p> |
<center><a href="https://2019.igem.org/Team:SEU/Demonstrate" class="buttonContri">Demonstrate</a></center> | <center><a href="https://2019.igem.org/Team:SEU/Demonstrate" class="buttonContri">Demonstrate</a></center> | ||
| − | <p style="font-size=36px">3.To help with experiment, we develop a software tool which can generate DNA reactions and relevant DNA sequences accroding to the input size of the neural networks. Researchers can directly use this tool to abtain their exepected DNA-based neural networks and conduct experiments.</p> | + | <p style="font-size=36px">3. To help with experiment, we develop a software tool which can generate DNA reactions and relevant DNA sequences accroding to the input size of the neural networks. Researchers can directly use this tool to abtain their exepected DNA-based neural networks and conduct experiments.</p> |
<center><a href="https://2019.igem.org/Team:SEU/Software" class="buttonContri">Software</a></center> | <center><a href="https://2019.igem.org/Team:SEU/Software" class="buttonContri">Software</a></center> | ||
Revision as of 08:16, 3 October 2019
Contribution
Through the comprehensive use of life science and information science knowledge, we have obtained the results of this experiment. In the whole process, we encountered many difficulties and challenges, but after careful thinking and practice, we finally successfully overcome these. In addition, we have also summarized some information that may be helpful to other teams, hoping to make some contributions to the iGEM community.
1. We propose molecular computation models for arithmetic operations in artificial neural networks as well as relevant reaction kinetic analysis.
2. Based on these models, we implement artificial neurons, which is the basic element of neural networks, with DNA reactions. Also, we achieve backpropagation training process for DNA-based neural networks.
3. To help with experiment, we develop a software tool which can generate DNA reactions and relevant DNA sequences accroding to the input size of the neural networks. Researchers can directly use this tool to abtain their exepected DNA-based neural networks and conduct experiments.