Team:UESTC-Software/Implementation
<!DOCTYPE html>
What can we do
Difficulties increase as synthetic biology networks expand, limiting the possibility to design more complex biology systems. There are challenges at every step of the way, from the characterization of biological elements to the designing and construction of systems. The effective integration of current research results of synthetic biology is a key step to promote its development. We note that most of the current approaches to protein design are de novo synthesis, requiring targeted modifications or resets of amino acid sequences that require corresponding modifications and resets of DNA. The workload of this approach is undoubtedly huge.
In addition, during the first decade of synthetic biology, most of the research focused on single-module design. Integrating different single modules to perform more complex functions is another key. At present, many excellent databases (CATH, SCOP/SCOP2, Uniprot) have detailed and accurate records of protein domain for protein supersecondary structure , and we assume that it may be feasible to study proteins with more complex functions by using domain as a module. At the same time, the research of domain will also play an important role in the development of synthetic biology in the future. The structural domain research can not only provide appropriate bricks for the design of new proteins, but also offer clues to the evolution of life.
This project establishs a new database by performing data cleaning, data integration, data specification, and data transformation on the domain information in the SCOP, CATH, and Uniport databases.
Through 3DZD, we extracted the 3D features of the domain and then clustered the extracted features according to Kmeans so that domains with similar traits are clustered into one category. Finally we built a domain parts database, processed the domain model in the database, and polished it to make domains' entity models.
Target group
|
If you are an expert in the field of protein domain, you can search for the protein domain you need based on the 3D shape classification of the domain in our database. We also provide the probability of binding between different types of domains and the functional information about the proteins that each domain participates in the construction for your protein design more convenient. |
|
If you are a biological researcher, CPD3DS provides a platform for you to design protein from the domain level. We classify the protein according to the shape of the protein, so that you can refer to the data to perform protein design by replacing protein domains with similar shapes. |
|
If you are an educator in the field of Biology, you can use CPD3DS to get vivid domain figure and protein figure which may help with promoting the understanding of microcosm. |
|
If you're a learner or interested in synthetic biology, you can get a better understanding of domain through our project. |
Other Challenge
It is complex to design proteins at the domain level. At present, what we can do is to verify the splicing of protein domains at the theoretical level, and it would be more expensive to verify it in experiments.
In addition, how to connect between protein domains is also a question worth exploring, which needs more experimental data to conduct research.
Further more, whether there are related activity and corresponding function after the binding of protein domains is also urgently needed to be verified. Regarding research on protein domains, we still have many difficulties to solve.