Team:HFUT-China/Contribution
Contribution
Based on the concept of “Internet Plus” and the methodology of software engineering, BioDesigner is devised as a software platform whose theme includes genetic engineering and synthetic biology. The whole system can be described as three parts, which are information retrieval, gene chain designing and entertainmen.
Firstly, the information retrieval section is the highlight of BioDesigner Dolphin. We designed a website crawler to collect data from the iGEM’s official website using Python BeatifulSoup. Then we sorted and cleaned the data to make it available for later use. To analyze and to let computers “understand” the data, we implemented a series of machine learning and statistical learning algorithms. For example, we used Latent Dirichlet Allocation (LDA) to automatically classify the documents, TF-IDF (Term Frequency-Inverse Document Frequency) to excavate keywords out of documents and Word2Vec to generate vectors which contain semantic meaning of words. We reorganized the analyzed data and saved it into database. Users can input keywords and at the meantime select multiple tracks which lets the system search for results in teams only under these certain tracks. Selecting no track means searching under all tracks. After a quick glance at the summarized information of teams, by clicking on a specific team will lead to a well-designed page with detailed information and the link of its wiki of the team on it. Furthermore, recommended teams will also be shown. Recommendation is made based on the semantic similarity and the number of shared parts between two teams. Searching for biobricks is also available in our system. We provide a convenient and powerful interface for quick information and literature retrieval. Other teams can benefit a lot when looking for information for its great convenience and function.
Secondly, the designing section lets users design their own gene chain by simply dragging and dropping biobricks listed on the left. In order to help users design highly dependable and qualified gene chains, recommended biobricks will also be listed according to the existing biobricks on the chain. Visible interface is provided in this part, and the easy operation as well as the friendly environment of it should not be neglected. With the designing function, participants can be worried no more about designing gene chains.
Last but not least, the entertainment part is newly designed and is another highlight of our project. Wearing VR glasses and holding handles, players can experience an amazing and immersed trip in the game world. The basic idea of the game is to shoot biobricks, which means collecting them, to form a gene chain. The whole gene chain will not be accomplished unless biobricks are collected in a certain order, which can be interpreted as biological feasibility of the gene chain. Actually, this part is mainly designed for the popularization of synthetic biology. Adding entertainment elements to it can let players more willing to play it, and at the meantime obtain knowledge about synthetic biology.
The three parts of our project mentioned above are devised to meet the demand of other teams and provide comprehensive, convenient and high-quality services. With our system, the threshold of scientific research will lowered, and synthetic biology is consequently further developed.
Based on the concept of “Internet Plus” and the methodology of software engineering, BioDesigner is devised as a software platform whose theme includes genetic engineering and synthetic biology. The whole system can be described as three parts, which are information retrieval, gene chain designing and entertainmen.
Firstly, the information retrieval section is the highlight of BioDesigner Dolphin. We designed a website crawler to collect data from the iGEM’s official website using Python BeatifulSoup. Then we sorted and cleaned the data to make it available for later use. To analyze and to let computers “understand” the data, we implemented a series of machine learning and statistical learning algorithms. For example, we used Latent Dirichlet Allocation (LDA) to automatically classify the documents, TF-IDF (Term Frequency-Inverse Document Frequency) to excavate keywords out of documents and Word2Vec to generate vectors which contain semantic meaning of words. We reorganized the analyzed data and saved it into database. Users can input keywords and at the meantime select multiple tracks which lets the system search for results in teams only under these certain tracks. Selecting no track means searching under all tracks. After a quick glance at the summarized information of teams, by clicking on a specific team will lead to a well-designed page with detailed information and the link of its wiki of the team on it. Furthermore, recommended teams will also be shown. Recommendation is made based on the semantic similarity and the number of shared parts between two teams. Searching for biobricks is also available in our system. We provide a convenient and powerful interface for quick information and literature retrieval. Other teams can benefit a lot when looking for information for its great convenience and function.
Secondly, the designing section lets users design their own gene chain by simply dragging and dropping biobricks listed on the left. In order to help users design highly dependable and qualified gene chains, recommended biobricks will also be listed according to the existing biobricks on the chain. Visible interface is provided in this part, and the easy operation as well as the friendly environment of it should not be neglected. With the designing function, participants can be worried no more about designing gene chains.
Last but not least, the entertainment part is newly designed and is another highlight of our project. Wearing VR glasses and holding handles, players can experience an amazing and immersed trip in the game world. The basic idea of the game is to shoot biobricks, which means collecting them, to form a gene chain. The whole gene chain will not be accomplished unless biobricks are collected in a certain order, which can be interpreted as biological feasibility of the gene chain. Actually, this part is mainly designed for the popularization of synthetic biology. Adding entertainment elements to it can let players more willing to play it, and at the meantime obtain knowledge about synthetic biology.
The three parts of our project mentioned above are devised to meet the demand of other teams and provide comprehensive, convenient and high-quality services. With our system, the threshold of scientific research will lowered, and synthetic biology is consequently further developed.
