USTC-Software/term

Today’s meeting was about choosing our team leader. Xiaoshan Sun and Chenkun Wang were the candidates for the leader and they all gave wonderful speeches. It was really hard to choose one from them. At last, Chenkun Wang won most of our votes and from then on he was our captain. He is a third-year student majoring in physics and also the former chairman of student union. We all believed that he would be a good leader since he was humorous and experienced. And Xiaoshan Sun became our vice-captain and would assistant Chenkun Wang with his work.
After the vote, Wang told us his hope for the 2013 software team and he said he would plan for the next meeting then emailed us what we need to prepare and our schedule for the whole semester. It was encouraged and satisfied to have such a good leader and we were ready to help him build the team to be a warm family.
At the end of this meeting, we chose Pei Zhang to be the person who was in charge of finance in the future. He is a freshman but has a cool head and speaks really fluent English. It seemed there were many difficult things for him to do, like getting sponsorship. But we all trusted him to fulfill his responsibility.
Our team finally had a leader and we all tried to figure out what we could do to make our team more united. I was waiting for the next meeting and starting to work.

Today we had the regular group meeting and everybody came on time. The meeting aimed at solving some current problems and made some assignments.
The first thing was about the 2013 iGEM software team. It was duty and responsibility that required us to perform more like a family rather than just a team. Everybody should devote to the project and know one’s own ability that can contribute the most to the project. Then we all agreed that every week every one of us should prepare at least one page slide to show the work in that week.
Then the second important thing was that everyone should have the awareness of one’s position in the team. Each member had different roles and by cooperating could we finally make the program go well. Those roles included the core roles like writing codes in C++ and Java, and other things like designing poster, starting wiki and learning related software.
The third thing discussed in the group meeting was about brain storm. To make the project more creative and innovative, every several days there will be a brain storm hold for everybody to bring up new things to enrich the contents of the overall project. By now we already had some good ideas like the construction of proteins, the signal process in biological circuits, the biological compiler for simple vivo system, computational biology and so on. Then we all had assignments to read journals and publications about these topics and will present results in the next week’s group meeting.

There were five main parts of this meeting and everyone was well-prepared for their own topics.
The first part was about how to score each person’s contribution. It was not really necessary to give every contribution a score, but we all agreed that in this way could we be more energetic and hardworking.
The second part was the core part of this meeting. Everyone showed their research results in front of the whole group via PPT. Our captain, Wang, had done plenty of work. He had talked to our leader teacher, Mr. Hong together with Pei Zhang and then registered our team. He was giving three lectures. One lecture was on mathematic about modifying cancer cells. One was a topic we had raised before, biological calculation and the last one was his newly-raised idea, logical circuits. It was a really interesting idea and we’d like to do more research on it.
Gen Li and Shuai Ma also did great work in poster and wiki, they read a lot and by then they had already been familiar with requests of posters and wikis of high quality. Tianhao Zhang was a guy who was really interested in computer and he did much research in universal algorithms and compiler. Xiaoshan Sun this week read paper about protein structures prediction, the topic he raised last meeting. However, he found this topic was not suitable for our team. He really had an objective attitude towards things since it was not easy to deny his own topic.
Xingyu Liao had always been an assiduous student, we could see from his report that he had learned a large part of C++ already and still was working hard in genetic path. Jinyang Li had thought of a new and meantime attractive idea, minimal core genome, a really hot topic in synthetic biology and he also introduced Python to us, software with powerful uses.
Shaohua Chen is a freshman majored in biology same as Pei Zhang, they both shared valuable ideas with us since they knew more biology than we did. Weichao Peng, who was really good at computer, gave a speech on visualization. Chen Shao, a talented girl majored in math, whose presentation was more about compressive sensing, told us many about machine learning and data mining.
The third part came the brainstorming, we decided to delete the protein structures prediction and add data compression to our topic list. Then we exchanged experiences about searching literature.
The last part was about simplifying everyone’s presentation because this time we took too much time. This was really a long meeting but we really learned lots of things and would do better job in the coming days.

In today’s group meeting everybody presented something one had read or learn in this week. The abstracts were as follows.
For Chenkun Wang, the team learder, his presentation was about using electric circuits to simulate the behavior of cells. The basic ions like Ca2+ and biological parts like proteins could be regarded as basic elements of a circuit so that analyzing a circuit like this is the same as analyzing an equivalent circuit consisting of electric elements. Apart from this, he also presented something about NP=P problem and its applications in DNA calculation. Finally he talked about some regular affairs about the registration fee.
For Xiaoshan Sun, his topic was about machine learning and system control. He gave us an example of that and also showed us something of a team from Korea in 2012. Then was Haoran Zheng, he centered on data mining, aiming to show that this method was not only useful in developing a searching engine but also helpful in other aspects.
Tengyang Xie’s presentation was on the C++ programming language and some already mature software like Proto and TinkerCell. He thought that we might learn something from the already existed software to make our software optimized. Tiaohao Zhang also gave us some examples in computational tools for biology with respects to the numeric. He showed that like MATLAB it was quite feasible for people to develop biological models based on existing packages.
Then Pei Zhang and Shaohua Chen gave their opinions on some topics. As to developing searching engine, even though last year we saw overwhelming interests in developing that kind of software, Pei Zhang thought it was not that easy as it seemed. Then as to biological compiler, Shaohua Chen kept doubted that this topic might not be practical because of lacking enough knowledge. Then he proposed other possible topics like setting up a website for software, thus the process of searching biological software would be expedited.
Xingyu Liao gave summaries of 2012 Johns Hopkins team, 2011 Welesley team and 2012 NYC-HUNTER team. Shuai Ma’s work was mainly on the DNA compression for storing more sequences. Jinyang Li gave a demo of using machine learning to find the minimum genome.
New topics kept being brought up in the group meeting. We all were very satisfied with this week’s work and determined to keep on working hard.

As our project reached a new stage, we had a meeting today. In today’s meeting, every one gave some opinions or advice about the ongoing project.
At first, our team leader Chenkun Wang showed us an overall schedule for the following several weeks. As middle exams coming near, we all have courses to review and exams to prepare, so time left for iGEM became a bit less. Chenkun told everyone to overcome the difficulties and pull through this period. Then Chenkun gave an introduction to signals in biological system. Those signals included electrical signals and chemical signals. But still problems remained unsolved like how to detect them and how to identify noises.
Then our vice-captain, Xiaoshan Shun told us something on system biology and basics about minimal core genome. In his presentation, some pictures about the definition of minimal core genome made the concept quite clear for us. In this week, Shuang Li’s work was mainly about Eugene. That was a kind of language used to store biological constructs. Then she talked about the structure of the language and showed some examples as well.
After that, it came to me. My work these days was about machine learning. I watched the first two lessons of the open courses on sina.com. Then I presented my notes to everyone in the room. The first lesson was an introduction to machine learning, what is supervised learning and what is unsupervised learning. Moreover, the second was about regression, methods in machine learning. Finally, it was Pei Zhang’s turn. His talk was about C++ and digital filter.
After this week, my work still kept on. I had to finish the online courses to find something useful for our project.

In today’s group meeting we all sat together to discuss something. Chenkun Wang, Shuai Ma and Xinyu Liao gave us something they learned and thought this week.
First we tried to make a schedule of the date of the next group meeting. Then Chenkun Wang showed us some bottlenecks we were faced up with this period. Along with some confusion, these problems were the ones we have to solve in the following weeks. Then Chenkun showed us something about the database. Among all the database on the internet, GDB, EMBL, Genbank and some others may probably be the database used in our software. Apart from the database, Chenkun also talked about some algorithms and methods commonly used in machine learning and data mining. All these technics can give rise to new topics like identifying different genes, selecting genes and grouping genes.
Shuai Ma’s talk today was mainly about things in data mining. The first thing was on scoring multi-sequences. This technic has to solve the problems like finding the similarities between different genes and selecting those sequences that can determine the functions of a gene. The second part was also on some comparing algorithms.
Xinyu Liao’s work was still about learning C++ and machine learning. He also searched the web of knowledge to find the paper that might be helpful for our project.
Finally, Chenkun asked us some open problems like what to do next and what methods to use. We shall keep working on these questions and finally make a decision the topic of this year.

Today’s group meeting was centered on some regular issues for registration and the project itself. First everyone should sign in online and make a schedule of one’s own. Then we moved on to talk about the how the project goes and the key points in finishing the project.
Our project this year was about the stability of a genetic network. First the user can add new genes into the existing network. Then, the software can generate a new network and provides the user the stability of the new network. Through calculation and analysis, the software also gave the user feedbacks about the properties of the new network. Here is something we should pay attention to. First is the type of network we chose. Then, what mathematical model shall we use to simulate and what database we may use. Besides, how to define the stability of a network and what the algorithms are. Finally, what user interface can be the perfect both for developers and users.
Then to make sure everything go well as we expected, we all have tasks to do. We have to learn how to design the code and the GUI. And also we have to make sure the software can have interfaces for its extension and connection with other software. When it came to the source code, we all decided to use the same standard of writing C++ to make the code beautiful.

iGEM chairman Dr. Gil Alterovitz is planning to pay a visit to USTC. So we all decide to give him some surprises about our project and wish to leave him a deep impression. In the group meeting today, we mainly discussed some problems unsolved at present.
The first one was about the proper application of the database KEGG. We decided to model the next work on the level of metabolic layer. The second thing was on the equations in the simulation. We thought that ordinary differential equation might be too restricted to be used here. Thus, we had to think out some new ways. Then Chenkun, the team leader, asked Jinyang to find a good place for us to work together in summer holiday. Besides, he would help everyone get a card for entering the life science building.
As for work assignments, in this group meeting we made everything clear. Shaohua Chen and Monan Zhou mainly work on wiki design. Shuai Ma’s work should be on Qt design and Jinyang Li, Xinyu Liao, Haosen Tan and Tianhao Zhang should concentrate on network analysis. Chen Shao, Pei Zhang and I have the task of finding out how to get the maximal concentration output. Our team leader, Chenkun Wang’s task was on the data fetching and analysis.
It was quite a long group meeting and we were all looking forward to meeting Gil several days later. .

Today’s group meeting was the last one in this semester. Since everyone has exams to take, the progress became slower this month. In order to finish the project on time, we made arrangement for the coming summer holiday.
Our team leader Chenkun Wang firstly showed us something deep in the project. First thing was about the database in the software. The metabolic pathway in KEGG is already a very comprehensive one so we should do our best to make full use of that. Then it was about the comparison of gene sequences. We would try to make a video of this feature so that this could be the highlight of our software. The last thing was the stability analysis and maximal output calculation. This part required efficient algorithms and computations.
Then we reassigned all the work for everybody and set deadline for everything. As for me, my task was about modular 3. I would cooperate with Pei Zhang and Chen Shao in finishing the structure of the modular. As for others, their work assignments were just like last time we assigned. The meeting required everyone to devote all the spare time to the project and overcome all the possible difficulties. After the group meeting I felt quite inspired and determined to do all my best in my role.

Today’s group meeting was mainly about writing the source code of our software. Before the start of the meeting, Chenkun Wang informed that we would have an ftp server to store all the slides of group meeting and all the codes. We all had to arrive 30 minutes earlier to attend group meeting every week in Room 429 at 7:00 Wednesday night.
Firstly, Chenkun showed us the class of TFTM, including the public and private variables and functions. Then he explained every detail to us about the definitions of the variables and functions. Then was explanation of the class called Regulation. He gave us some global variables to store important information of the system like the number of genes and the regulation matrix of genes. By his illustration, we came to understand the overall structure of the code. Then he gave some instructions on the interfaces of each class. Thus, we knew how to link all the classes together to make the software work.
Chen Shao was working on the code already in this week. At the same time, she kept learning how to write C++ more efficiently and thinking about how the next step went. She met some problems when writing the code and discussed together with us.
For me, I took note of all the things in the group meeting. I made up my mind to do some work to help my other teammates.

Weekly Report 2: Today we had a discussion together on algorithms in the software and the logo design. In our software, we have a lot of parameters to optimize. Thus, we have to find an algorithm that is most suitable not only practical but also efficient. Finally we decided to use particle swarm optimization as our method to optimize all the parameters. The next step for us was to read more paper on PSO, search the internet and finish the code of this part as fast as possible. As long as we had the optimization algorithm, then we would have the chance to test the code and improve the code. Then we discussed about the reverse prediction code. This part was independent to the algorithms so we could finish this at the same time. When talking about the logo, we all had our own visions of the logo. But the final decision had not been made because this discussion needed more time. What we want was a logo that not only presents the spirits of iGEM competition but also has features of our software. The summer holiday is coming and I am already well prepared for all my tasks.