Team:UESTC-Software/Notebook
[4.1] Create teams: We have created a team with 15 people from biotechnology, biomedical engineering,
software and
other majors
[4.6 & 4.13] Team meeting: We held preliminary discussions for brainstorming to gather inspiration, and
ultimately determined the theme of continually improving BioMaster.
[4.15] Exchange for proposals on project with Senior Student Zhang Kaiyue, leader of UESTC-China in 2015 iGEM.
[4.25] According to the questionnaire, we decided to add some new databases. We went to Shangsheng Law Firm in
Sichuan to consult lawyer Liang about the copyright of database usage.
More
[5.3-5.4] 2019 Southwestern iGEM Exchange Conference of China was co-hosted by us and UESTC-China in Qingshuihe campus of UESTC. From this meeting, we got some suggestions on optimizing search and improving database mapping relationship. [Week 2&3&4] We set up five groups: database, art design, Human Practices, new function and web authoring. The project was officially launched. [6.1] We held a large-scale publicity science activity in Sichuan Science and Technology Museum. Students from the University of Hong Kong were invited to organize activities to exchange database technology with students from computer majors. Patrick suggested on adding docker to increase the portability of the project and prolong the project generation cycle. [Week1&2] Front-end and back-end groups began to learn knowledge about programming; database group searched for suitable databases; design group designed web logic; function group and Human Practices group explored the needs and collected ideas. [6.16] We interviewed Professor Zheng Xuelian, a senior teacher of UESTC-China, and discussed the development of synthetic biology. Under her suggestion, we added an education module to our project. [week 3] We utilized web crawler to get a small amount of database data, which validated our ideas and technologies. [Week 4] A new sequence alignment model was eatablished to improve integration accuracy. We learned about Amazon's S3 cloud database service and decided to use local MySQL + cloud database to store data. The new project was named “BioMaster 2.0”.
More
[week 1] We got the required data through download or API access, including 11 databases: UniProt, BRENDA, BioGRID, etc. Simultaneously, we collated the copyright agreements of various databases to ensure the legitimacy and rationality of our project. [week 2] We updated the team wiki information for 2018 . Besides, the database of BioMaster 1.0 was integrated, and useful information was migrated to BioMaster 2.0. Considering the performance and function, we decided to abandon think PHP and used Laravel to develop our project instead. [week 3] We constructed and configured the server to run demo. Besides, we compiled the prototype web page of Wiki, started to build the front-end web page prototype of BioMaster 2.0, and gradually completed the docking with the back-end. All pages for BioMaster 2.0 was designed. During this week, we started designing synthetic biology brochures and table games BioME. [7.20] We interviewed Professor Zeng Dong of our school about search optimizing. He recommended Elasticsearch as a search plug-in for us. [week 4] To solve the problem that some sequences in the old screening model were ignored, we decided to use "full sequence + feature annotated short sequence" as the BLAST object in the new model. In the pre-processing of iGEM part source data, we BLAST the corresponding short sequence according to the information given by feature, and screened the result of blast preliminarily. Finally, we got 3 groups of different tables for source data by 3 different sorting methods.
More
[week1] Inspired by SynBioHub and SBOL designer, we planed to devise visualization functions to display the auxiliary design functions in our own database. In the next month, we designed and compiled a web version of SBOL aided design tool according to the official standard of SBOL, which was used to realize sequence visualization. [9.7] We went to Shishi High School for iGEM propaganda and published our brochure. [week2] We finished the EC Number Prediction. The prediction tool used machine learning and constructed multiple binary classifiers. It combined three independent predictors based on subsequences, sequence similarities, and amino acid physicochemical features. It provided probabilistic enzymatic function by predicting the EC number. [week3] The display form of the information in the database was determined. We finished all database pages codes and added special effects by using bootstrap response framework and other plug-ins when designing the response of web pages. In addition, we realized the adaptive function of different screen sizes of web pages. We started to integrate all the new functions into the program code and did some UI beautification as well. [week4] We kept on writing drafts and adjusted web page layout. We started to write drafts for our presentation and design our PPT. The PC version of Bio-ME is finished, which enabled users to learn the basic knowledge of synthetic biology.
More
[week1] We continued to process the data of the screening model, merging and labeling 3 sets of tables. By compressing the range of independent variables, we set 36 groups of different parameters, draw the frequency distribution histogram and Poisson distribution fitting curve, and marked the reasonable threshold. [8.7] We went to Chengdu Junyu Biotechnology Co., Ltd. to visit the researcher, Mr. Zhang Yunfei, and consulted about the database function. He mentioned the importance of prediction in biology. Therefore, we decided to add EC number prediction function as a supplement to BioMaster 2.0 database on the basis of promoter prediction last year. [week 2] We further analyzed the screening results and fitted the tail thresholds of 36 groups of curves with normal distribution to obtain the median values, so as to determine the threshold of the formula. We used triple intersection as a criterion to evaluate the accuracy of the results, and drew the proportion of triple intersection under different parameters in the whole data set. Thus, the screening formula was successfully deduced when finding out a set of parameters with the best performance under reasonable thresholds. [week 3] We optimized the demonstration of interaction networks and used open source software platform Cytoscape to visualize interaction networks. According to Zeng Dong's suggestion, we adopted Elasticsearch to achieve full-text keyword search. Although we encountered many problems when trying to integrate Elasticsearch into BioMaster 2.0 framework, we never gave up. This work continued until week4, and we finally solved all these problems. [8.20-23] We went to Shenzhen to participate in the 6th CCiC and exchanged with the various Chinese iGEM teams. Taking the opinions of other teams into account, we decided to add some practical features to BioMaster, such as ID conversion and BLAST API. [week 4] After debugging for a long term, we successfully constructed the scoring equation of screening model. We obtained the best screening settings: 0.65 for threshold, 1-5-4 for coefficients. We also made a large number of random sampling tests on the obtained data from the new model, and especially enumerated the special cases such as data contamination. All the test results showed that the new model had a better screening performance.
More
[week1] We started to write drafts for our
presentation and designed our PPT. We completed the development of docker in BioMaster 2.0. Users were able to
download the blog from GitHub and ran it in docker by themselves. Without the support of remote server, they
could also experience the data search function of BioMaster.
[10.8-10.12] We invited CAU-China, Tongji-Software, SJTU-Software and other teams to try out the software and
collected their feedback.
[10.11] We went to Sichuan University and invited Professor Liang to use our software. She provided some
suggestions on the timeliness of data.
[week2] We added the update function to ensure data timeliness, and also simplified the maintenance of the
database. Besides, we finished our wiki and started to upload our web pages. We recorded our demonstration
video for BioMaster 2.0, finished our judging form and applied for four best awards.
[10.15] We held the Second iGEM Festival in UESTC.
[week3] Web debugging, database testing and rehearsal. We added some annotation for our codes and submitted
them to GitHub. The final version of our poster came out and we prepared for our trip to Boston.
[10.21] Wiki freeze
[10.31] Our journey began!
More