Our Interaction with Other Teams, Labs and the Community
Consulting with previous iGEM SJTU-Software leader for further improvements: Our team is also responsible for coordinating communication between the previous captains of the iGEM Software Team. We have established a communication group for the iGEM Software Team captains, where all the previous captains are included. This allows captains to communicate and share their valuable experiences and lessons learned in leading their teams and carrying out projects. Through the harmonious communication and the inheritance of experiences among the captains, our team is able to learn from each other's strengths and weaknesses, and operate more efficiently under the leadership of the captains.
The 10th Conference of China iGEMer Community (CCiC) was successfully held in Haikou from 8 to 10 July 2023, where the HP team participated and presented their projects. CCiC is a nationwide summit independently initiated by China's iGEM teams, aiming to provide a resource-sharing platform for participating iGEM teams to promote mutual learning and exchange. The 10th CCiC was hosted by Hainan University and organized by State Key Laboratory of Marine Resource Utilization in South China Sea.
From 2013 to 2023, CCiC celebrated its 10th year, and this year was also the first time that it resumed to be held offline after the epidemic, with online and offline synchronization. The conference attracted 90 teams and 12 individuals, with a total of 1,548 attendees. The three-day conference included an opening ceremony, academic lectures, team report presentations, poster presentations, workshops, team discussions, and a closing ceremony.
The HP team participated in the opening and closing ceremonies of the conference and presented our project during the conference, sharing our current progress and future plans in detail and drawing on the advice of experts. In addition, the HP team benefited greatly from listening to many excellent academic lectures and sharing by other teams, and introduced our work plan to other iGEM teams before the presentation. The HP team also had extensive and in-depth exchanges with iGEMers from all over the country, and gained a lot of inspirations and constructive opinions.
During this year's summer break, our team visited the Shanghai National Center for Applied Mathematics - AI for Science Workstation for educational purposes. Here, the instructors provided insights into the current state of protein research and shared their perspectives. They emphasized that the practical activities would revolve around the application of large-scale modeling techniques for protein research. This primarily included research in the fields of deep learning evolution, encompassing subjects such as schizophrenia, deep-sea species, and Argonaute. Additionally, there was a focus on the application of protein language models, with a particular emphasis on optimizing Argonaute's thermal stability.
Throughout the visit, we engaged in discussions about research details, internship project procedures, and future trends in protein research. The workstation also offered us opportunities to learn and apply artificial intelligence technologies, particularly in the realm of biological sciences.
The Workstation organized and led us on visits to renowned AI-related companies, allowing us to firsthand experience the innovation and exploration at the forefront of science. This visitation not only facilitated face-to-face interactions with industry experts but also provided us with the opportunity to gain a deeper understanding of practical research work.
At the AI for Science research practice workstation, our interactions extended beyond visits, research, and learning; we also engaged in discussions with a seasoned team of industry-academic-research mentors. This exchange not only allowed our team to delve into the development trends and hot topics within the research field but also presented to us the value and prospects of scientific innovation from an investor's perspective.
During these discussions, we delved into ethical and societal challenges posed by artificial intelligence in biological science research, and explored paths towards sustainable development.
During the proposal period,all members of the HP team contacted several professors from the School of Life Science and Technology whose research are related to the field of synthetic biology in order to broaden our thinking.Through communication with the professors, we broke through the current thinking barriers and obtained more novel ideas.
Members first contacted the professors via email, and after receiving enthusiastic responses from the professors, we communicated with them offline.During the communication, Professor Liang Rubing mentioned her research team's current research content and existing limitations, and provided us with the idea of designing a metabolic interaction model between microbial communities.In the process of communicating with Professor Zhao Weishen, he suggested to us the idea of studying enzyme-directed evolution, and mentioned that such enzyme modification can improve the efficiency of product production or pollutant degradation, and is conducive to degrading pollutants existing in extreme environments.Not all valuable suggestions provided by other professors are listed.
During the proposal period, the iGEM team received great help from many professors in the college. They pointed out the problems in our thinking and the possible problems and limitations in different research directions. After several brainstorming sessions at team meetings, the team members finally determined the research direction. Once again, I would like to express my heartfelt thanks to many professors who have helped us during the proposal process.
In order to communicate work progress and solve problems encountered in research in a timely manner, the team regularly holds meetings to summarize the work of the previous period.
Early stage: In order to start the project, members from different groups such as front-end, back-end, and HP gathered together to propose research directions of interest and significance. By reviewing the literature, members analyzed the potential difficulties and limitations that may arise in future research in different directions, and then the members of HP collected and summarized the problems, communicated with professors in related research fields of the college to get suggestions.After several brainstorming meetings, the team determined the current research direction. The team leader coordinated and planned the tasks and work timeline of each group to carry out the next step of work.
Mid-term: Each group holds separate internal meetings, with the frontend, backend, HP, and design team leaders organizing tasks in an orderly manner. At the same time, all groups maintain close contact, hand over work content in a timely manner and cooperate. Most of the project work was completed during this stage.
Later stage: The team members gather together to summarize the current work content, check for deficiencies and make up for them, and sprint together towards the final deadline.
During the initial phases of our project, we engaged in extensive communication with experts from various fields while simultaneously conducting comprehensive literature and data research. We examined recent exceptional iGEM projects and drew upon the topics our team had explored in the previous two years, with a specific focus on the intersection of type 2 diabetes and GEM [1-4]. During this process, we conclude that:
A genome-wide metabolic model, known as iYLW1028, was developed for Actinoplanes sp. SE50/110. This model incorporated the GECKO (GEM with Enzymatic Constraints using Kinetic and Omics data) method, allowing the imposition of enzyme constraints within the GEM. This innovative approach enabled researchers to reduce spatial variability and enhance prediction accuracy. Consequently, it became feasible to quantitatively analyze acarbose production while considering the catalytic functions of various enzyme types.
However, we encountered challenges due to the scarcity of experimental data and the inherent inaccuracies in predictive models. Improving the precision of Kcat value predictions within our existing GEM model proved to be particularly arduous. As a result, we shifted our attention from predicting protein reaction activity to predicting protein thermal stability. After thoroughly exploring a range of protein language models, we concluded that employing PLM for protein thermal stability prediction held great promise. This decision marked the beginning of our journey in constructing DARWINS.
For more detail about constructing DARWINS, please refer to Engineering.
[1] Wang Y, Xu N, Ye C, Liu L, Shi Z and Wu J (2015) Reconstruction and in silico analysis of an Actinoplanes sp. SE50/110 genome-scale metabolic model for acarbose production. Front. Microbiol. 6:632. doi: 10.3389/fmicb.2015.00632
[2] Gu D, Jian X, Zhang C, Hua Q (2016) Reframed genome-scale metabolic model to facilitate genetic design and integration with expression data. IEEE/ACM Trans Comput Biol Bioinform https://doi.org/10.1109/TCBB.2016.2576456
[3] Lu, H., Li, F., Sánchez, B.J. et al. A consensus S. cerevisiae metabolic model Yeast8 and its ecosystem for comprehensively probing cellular metabolism. Nat Commun 10, 3586 (2019). https://doi.org/10.1038/s41467-019-11581-3
[4] Li, F., Yuan, L., Lu, H. et al. Deep learning-based kcat prediction enables improved enzyme-constrained model reconstruction. Nat Catal 5, 662-672 (2022). https://doi.org/10.1038/s41929-022-00798-z