Summary

Collaboration is such an important practice within the scientific (and wider) community and this can be demonstrated by the mutlitude of ways in which collaboration can improve an iGEM project and an iGEM experience. Our team fully appreciate the power of bringing together different teams to share their expertise for the benefit of a shared objective, project, or mission. Collaboration helps us to problem solve, to bring people closer together, and to open up new channels of communication.

Over the course of the iGEM competition our team engaged in multiple collaborations with teams from around the world. This included hosting a successful international event to build community amongst iGEM teams and working with multiple different teams to strengthen our respective projects and promote good scientific practice. Here we have documented these collaborations.

Collaboration with Freiburg iGEM Team

"chAMBER meets AdaptR - the start of a new chAptR"

Motivation

This collaboration, following a discussion about our different approaches to modeling and data collection, can be summarised as a collaboration in model and data verification. The Freiburg team shared with us raw growth curve data from their lab, and likewise, we exchanged this for growth curve data we had obtained in the lab. Using their data, we used curve fitting algorithms to fit their growth curves to equations (namely the logistic growth curve equation) and assessed the goodness of fit and parameters associated with the fits. The successful fits and the equations associated with this could then be used by the Freiburg team in their simulation efforts whereas the poor fits would point to interesting questions and prompt discussion about why those fits were not appropriate. Subsequently, we engaged in further discussions with the team about what we could infer from their data and any insights we had obtained from working with it; this included speculating on how they originally modelled the system and on if any of their results appeared anomalous. This collaboration sought to improve our respective projects in multiple ways:

• To remove bias in the data fitting process. Fitting their data using curve fitting algorithms without prior assumptions on what we were trying to fit the data to means that these fits were less biased. Consciously and subconsciously, it is natural in scientific disciplines to want your data to agree with your model predictions however this is in some ways problematic in the sense that alternative fits, which may be better, are ignored. The Freiburg team can also engage in curve fitting with their data to their original model but now with an additional unbiased perspective of how best to represent their data.
• To identify anomalous results. By working with each other’s data sets we were able to identify errors that may have otherwise been missed.
• Improve transparency and honesty in reporting our results by allowing each other access to our raw, unprocessed data.
• The discussions we engaged with each other allowed for new perspectives and thoughts to be discussed which is important in any scientific discipline. Listening to thoughts from an external iGEM team was valuable and gave us new things to think about.
• The collaboration provided a curve fitting service to both teams which accelerated and enhanced the data analysis process.
• It was fun and exciting to learn more in depth about another iGEM team’s project and have insight into what has been going on in the iGEM labs from around the world! This increased the sense of community felt by each team. This also added to our excitement about attending the Jamboree and meeting the Freiburg team and all the other teams; this provided us with a motivational boost in the last few weeks of the competition!



Method:

The growth curve data provided by Freiburg was in the form of fluorescent measurements vs time for a range of different assays (e.g. induced/ uninduced). The fluorescence was assumed to be proportional to cell population. The rate of change of population growth was then calculated for each data time point such that the data could be fit to the logistic growth curve commonly used to describe cell population growth:
$$dN/dt=a\left(\frac{b-N}{N}\right)x$$
Here N is the proportional to population, a is the maximal rate of growth and b is the carrying capacity of the population. These were assumed to both be greater than zero. The data was then fitted to the logistic growth curve using the method of non-linear least squares and the parameters a and b extracted along with measures of the goodness of fit (SSE, R^2 value, adjusted R^2 value, RMSE). The success of the fits was then discussed and, where the fits were poor, alternative fits were considered.

Thank you to the Freiburg team for the collaboration; we deemed this to be highly valuable to our project and an important exercise in model/ data verification that is good practice for any scientific project.


See the Freiburg Team's wiki for their documentation of the collaboration, and to learn about their exciting project!



Results- Cambridge Fits to Freiburg Data:




Below are figures showing the fits vs plotted data for the Freiburg Team's data, as fitted and plotted by our team. It is noted that fit 4 is not included (see the parameter table) since this fit was poor, due to suspected problems with this data series).















































Example data for the fits for the Freiburg Team's data, as fitted by the curve fitting tool in MatLab and handled in Excel:





Results- Freiburg Fits to Cambridge Data:

The fits were made to the growth curves acquired as part of the JUMP DV characterization project. Below we have included the fitted parameters and model details as obtained by the Freiburg team for our data series:


Conclusions and Further Discussion Points:

Reassuringly, both teams fits were consistent with the fits each time had obtained for their own data. Thus this collaboration successfully added extra validity to the fits which each team had used for their data and planned to use for subsequent modelling efforts.
The two team's approach to fitting was also different; we used different curve fitting tools and algorithms, and although we both fitted to the logistic equation, we used different forms of the equation and plotted different variables. The fact that two different approaches yielded consistent results was reassurring; it gave us confidence in our methods, validated assumptions made in the fitting process and removed any systematic bias that would've arisen if we had adopted identical approaches. This said, the following points were raised and discussed:


• The Freiburg team suggested that it would be better to fit the data to the logistic growth curve only up to the ‘lag phase’ period.
• The size of the well plates may affect results and that it should be fit in two parts (this was about your fluorescence measurements, because we think the decline)
• The poor fit of Freiburgs data series 4 suggested something had gone wrong with that experiment (and indeed this had been suspected by the team previously).
• Could other fits be more representative?
• What are is the physical significance of the parameters obtained (discussed in our introduction above)? What do these mean in the context of our systems?
• What is the most appriate choice of variables to plot?


This collaboration gave us lots to consider in terms of our approach to fitting data. It gave us further confidence in our approach whilst also encouraging us to scrutinize our analysis in greater detail. We really enjoyed discussing our results with the Freiburg team, learning about their project and talking with them about life and their iGEM experience; we thank them for their friendship and support!






Grand iGEM Quiz- Hosting an International Event

On the 27th of August we hosted an exciting international virtual meetup in the form of a friendly quiz which provided participants with fun and healthy competition.

We chose to host such an event for a number of reasons. Firstly, we recognize as a team the importance of maintaining a healthy work-life balance for looking after mental health and wellbeing. We understand, first hand, that the iGEM competition can be stressful and for many participants involved long hours in the lab or thinking hard about synthetic biology problems. Therefore, we thought a fun quiz would be the perfect event for iGEM-mers to relax and take a well deserved break where they can engage with fun questions that are not directly related to their synthetic biology projects. Writing and hosting this quiz also proved to be a great way for us to take our minds of some of the stressful aspects of our project.

Secondly, a quiz is great for team bonding and we thought it would be a great opportunity for teams to boost their sense of team by showing off their knowledge outside of synthetic biology. This would ultimately strengthen the ability of teams to work together on their project!

Thirdly, it served as a great opportunity to meet teams from all around the world! In between answering questions about drosophila genes and ignobel prizes we chatted about life in different countries and about our mutual excitement for the jamboree. We all left we a renewed excitement for the Jamboree and an eagerness to meet some of the lovely people we spoke to!

We have included pdf versions of our quiz slides and answers so that future iGEM teams can easily host a similar quiz or for anyone to test their knowledge in their spare time!



Download the quiz slides and answers here to have a go at the quiz!

The Grand iGEM Quiz 2022- Questions

The Grand iGEM Quiz 2022- Answers




Collaboration with Bath iGEM Team




After engaging with the UniofBath iGEM team via social media, and learning about their project through their promotional video and outreach efforts, we decided that our respective projects would benefit from collaboration. There was a clear opportunity to incorporate the antithetic integral controller motif that we were working with into their project. The addition of a control motif would act to overcome potential issues around maintaining expression when taking their ‘PhoBac’ from the lab into cold soil. Team members from UniofBath then visited our lab in Cambridge to discuss the collaboration further, and for each team to familiarise themselves with the other team’s project. After this in-person meeting and thorough subsequent brainstorming, we designed circuits that combined our motif with their proposed system. UniofBath came away with designs for a more practical and commercially viable ‘PhoBac’ product, and we came away with insight into how to develop our system for potential real-world application.




We have discussed the implementation of our project in the context of Bath's system in ourImplementation Page





Below is a circuit diagram showing how our circuits would work together:











Following the circuit design stage, we sought to find the appropriate parts to make the circuits work, with the intention of incorporating parameters associated with these parts into our models, and simulating the operation of our joint circuit. Here we encountered challenges in the lack of well-characterized, safe bacillus orthologous parts. For our circuit to work, we needed to find antithetic molecules produced by bacillus, that are orthogonal to the rest of the circuit. In failing to find the appropriate parts, we here put a call to action for the characterization of more safe bacillus orthologous parts. This highlights further the challenges associated with working with non-model organisms in synthetic biology; a challenge that ADAPTR would face if it were to be implemented in other non-model organisms besides bacillus. Furthermore, both teams acknowledge the challenges we would face in cloning our two circuits together if we were to have found the appropriate parts. Despite the clear challenges associated with combining ADAPTR and PhoBac in the lab, this collaboration was of great value to both teams. As discussed, it provided UniofBath with a circuit design that would overcome a challenge they faced in implementation and introduced them to the merits that using control theory in synthetic biology can have on outcome, and it provided us with a clearer vision of how our circuit would be used in real-world applications and served as practice for how we could think about applying ADAPTR in a variety of novel systems. Both teams further benefited from our thought-provoking discussions, from gaining new perspectives, and from learning about each other’s iGEM experiences. We look forward to seeing the UniofBath team again in Paris, and seeing how their project has progressed!











Coding Workshop

We collaborated with iGEM Vienna, iGEM UCL and iGEM Sheffield in our coding workshops. See the Coding Workshop section on our Communication page to find out more!



Synthetic Biology Book

We have collaborated with Shym_NIL_NIS Team, a high school team in Kazakhstan, to create a synthetic biology book to explain our project to the general public. For more information, please refer to the SynBio book section on our Communications page.

See below for a photo of our collaboration meeting!