Results
This year, our dry-lab was able to successfully model the lead sequence optimization process for secretion via T3SS in Java. We implemented and considered many biological functions in our code and assigned biological activity values for each output sequence, which allowed us to visualize how the system was optimizing the sequence across generations. Building and studying this model gave us insight into the function of T3SS and prepared us to design sequences that would efficiently secrete.
The current program takes a sequence of 20 nucleotides and optimizes it to obtain the highest possible activity value. This is useful for optimizing the lead sequence as it allows for lead sequence to achieve a higher biological activity rate. By doing so, one could splice together the optimized lead sequence and another strand in order to help a protein or other substance proliferate.
In the below graph, the optimized sequence for each generation is generated and then the optimized biological activity value for each generation is then also generated. The average activity column is included due to the fact that the program starts off with a random population value and then mimics the spreading of that population and also somewhat mimics crossover and natural selection as well. These processes were added in order to mimic what may go into optimizing a lead sequence. Due to the structure of the code, at some point, the code fails to optimize the code further then. At 0.99 biological activity, the program can not properly calculate relative to a value close to 1 so it begins to spit values that are not optimized.
Before deciding on our DiGI-T3 lead-sequence optimizer, we had other ideas for modeling some part of the therapeutic T3SS design cycle. The first idea that the team wanted to pursue was a support vector machine (SVM) model to determine a statistical probability of secretion based on protein fingerprinting criteria. However, our meeting with Dr. Lesser taught us that secretion does not guarantee proper folding, and that this folding is more often the issue. Her recommendation was to focus instead on nanobodies, which almost always fold correctly due to their tiny size.
For human practices, our team was involved in many outreach events and discussed with professionals of diverse fields. Our outreach events at the Center of Science and Industry (COSI), Boonshoft Museum of Discovery and WestFest at the Ohio State University allowed us to educate kids and parents about the goals of our project while promoting the ideas of synthetic biology. Our discussions with professionals guided our approaches to both dry-lab and wet-lab during brainstorming and conceptualization. Other professionals gave us great direction in the proposed implementation of our project and made us more aware of our project’s place in the world, specifically for its potential use as a therapeutic for GI cancers. Additionally, we obtained valuable data on the experiences of GI cancer patients and their perspectives on our project idea.
Unfortunately, due to struggles with securing a lab space, many of our wet-lab goals had to be delayed. We still made sure to document our expected plans and wet lab engineering process for future reference. The work we did with dry-lab and human practices should prepare our team to pursue a similar project with a heightened focus on wet-lab activities. Better yet, our work could be used by other iGEM teams to develop a project of their own using T3SS.