For more details of the best new basic part, please visit the Part Improvement page.
We have completed the Competition Deliverables including the Wiki, Project Promotion Video, Team Presentation and the judging form.
Link to our presentation (pdf)
On our attribution page we have explained the work that each of the team members did, and where our advisors assisted and taught us key skills. Though much of the team members' work overlapped, we have discussed the areas each member focussed on. We have also mentioned all the people that contributed in a small way to our project, whether it be academics and people in industry who agreed to an interview as part of our human practices, or those in Cambridge who helped us get access to lab space and equipment. We have also acknowledged the departments, colleges and companies who have helped us finance our project - without their support it would not have been possible.
On our description page we have clearly detailed the motivations behind our project - to work towards solving the issues caused by variability in synthetic biology. We have detailed how our solution - a biological integral feedback controller - tackles this to reduce variability in biological systems. We have also explained why we felt this was an important project to work on, applying the framework of effective altruism to potential projects to help ensure we could make the maximum impact possible. We then go on to discuss potential applications of our project, to clearly demonstrate with examples the wide-ranging utility of our project. This has been presented in an accessible way, with figures to help aid explanation of key concepts within the project, and references for those who wish to know more background.
We have 3 parts to our contribution: Wet lab, Dry-lab and creative: Our wet-lab contribution consists of characterising 3 JUMP DVs from the distribution kit - pSC101 (low copy no.), pBBR1 (low-med copy no.), and pUC (high copy no.) in varying concentrations of ammonium sulphate. Uploaded on the pSC101 registry page, we recorded an increase in sfGFP fluorescence in the low copy number plasmid up to 540mM [NH4]2SO4, and the same in pBBR1 (low-med copy no.) up to 360mM [NH4]2SO4. This can be used to help future iGEM teams determine how to optimise their protein expression depending on the copy number of their DV. Our dry-lab contribution is using Computer Aided Design to 3D print our own shaking incubator rack, with our method/instuctions available for future iGEM teams to follow. This decreases the cost from the market value ~£50-200 to only ~£14 for the material (provided you have free access to a 3D printer). Shaking incubator racks are designed to improve aeration of bacterial liquid culture when they are shaking so these should be considered essential kit for synthetic biologists. Finally, we know that many teams are tight on budget but we believe that the iGEM community spirit shouldn’t be missed out on due to this! We created instructions on how to DIY your own team t shirts which brings the cost down significantly for t-shirt personalisation and can ensure teams don’t feel like they have to miss out on the creative aspect of iGEM in order to prioritise their project.
Demonstrate engineering success by going through the engineering design cycle Across the course of the project, we have gone through multiple rounds of the engineering design cycle, as we have developed our protocols and parts. On this page we have included 3 examples of this. Firstly, the development of MegaT - our reporter construct. We built and tested our original version of MegaT, but found the fluorescence levels did not match our expectations well, so had to redesign and improve it in order to make the output gene level match the feedback reporter level in the circuit. Secondly, we build a negative feedback circuit, to test the effectiveness of this control against our integral controller. The original design and mother machine experiments had significant issues with mother machine imaging parameters, and expression levels. To fix this we redesigned the circuit to have stronger expression, and improved our method of collecting data with the mother machine to avoid phototoxicity. The third example of engineering success here is in our use of the downstream site integration provided by JUMP cloning. For this we had to develop and test adapter DNA sequences to allow integration of our assembly into the level 2 plasmid.
Collaborating with other iGEM 2022 Teams
We engaged extensively with a number of different iGEM teams over the course of the summer, across different aspects of the iGEM competition. With team Freiburg, we collaborated to verify our data interpretation and modelling. We did this by each looking at the other team's growth curve data, and interpreting the data without prior biases or expectations. This allowed us to verify that trends we were seeing in the data were reliable and not due to our analysis being biased by prior expectation of a particular result. We also hosted a virtual meetup, in the form of a quiz for other iGEM teams. This gave teams from around the world the opportunity to meet up, discuss their experiences and have fun answering questions. With the Bath iGEM team, we collaborated to improve the viability of applications of both our projects. By applying our circuit to their PhoBac system, we have together demonstrated how both our circuits can work in the real world, with our circuits interchangeable outputs allowing us to easily design systems that would otherwise be exceptionally complex. We also collaborated with Vienna, UCL and Sheffield on the coding workshop series as part of our education efforts.
Our project started from the effective altruism principle of maximising the good we could do with it, which naturally led us into a foundational project, as discussed at the start of our human practices. Since our project was foundational, we made sure to consider a broad range of people to discuss our ideas with, across academia and industry. This spread of people with different backgrounds allowed us to ensure we were both building a project based off of the latest research in the field, and that would be useful and relevant in industrial and other applications. Our discussions have shown our project could be very beneficial - people from many different companies and industry sectors were interested in the project and could see viable use cases of it. We have also considered potential adverse effects of the project and ensured that these are both minimal and greatly outweighed by potential benefits.
As a foundational project, there are many potential use cases of our integral feedback controller in the real world. With our implementation we have tried to make the most of these wide use cases by designing a range of circuits with different strengths and properties, so that different ones can be used for different applications. To allow for this as easily as possible, we have proposed different possible ways of using our circuit, with a focus on flexibility and ease of use, exemplified by the version of megaT that allows for direct insertion of genes of interest, allowing use of our circuit with just 1 cloning step. We have also proposed many potential use cases, and schemes of how to decide which of our circuits may be most appropriate for each use case. We’ve also considered how safety considerations will impact implementation of our circuit, and along with the Bath iGEM team developed a full real world use case of the circuit, which serves as a demonstration of how it would integrate with other circuits.
Discussions with academics and industry experts in our initial human practices work showed us many things we needed to consider further in the design of our project, such as the potential metabolic cost of our circuit. We therefore made sure that within our factorial design we had included strategies to lower metabolic cost, with different circuits allowing us to find the optimal balance between metabolic cost and stability of the circuit. Discussions with industry also led us to implementing better ways to perturb the circuit to ensure we would be able to fully demonstrate the adaptation of the circuit. We also had to prioritise what was feasible with the equipment and time available, which stopped us from using certain perturbation types, and caused us to stick with the integral controller, rather than improving it with addition of a proportional controller. By focusing on parts of the project that seemed most relevant in discussions, such as the ammonia perturbation, we were able to make a far more transferable and reliable controller than otherwise. We also decided to develop a new library of pBADs in response to our discussions highlighting problems with wild type pBAD.
We successfully engineered two improved versions of the araBAD promoter, which are parts of a new collection of pBAD_APs. All the designed pBADs have truncated length of less than 160 base-pairs, with some even showing higher performance than the 300 bp wild-type counterpart. We envisioned that the significant reduction in the promoter’s size would be of huge frugal advantage for future circuit designs. With less than 150 bp, the promoter can be synthesised by oligo-annealing, which is significantly faster and cheaper than ordering DNA parts. This could also be of great benefit for future iGEM teams from regions where DNA synthesis and delivery are often slow, allowing them to speed up project timelines.
Our modelling efforts were extensive, varied, and critical to our project. We used deterministic ODEs of varying complexity to verify our circuit design would behave as intended and used them to fit our data to to better characterise our circuit. We used stochastic simulations to determine the extent to which noise affected our circuit behaviour, and applied global sensitivity analysis to determine the most effective ways in which to perform our perturbation experiments. Furthermore, we used the tools of control theory and analysis to evaluate the stability of our system. This involved detailed mathematics and thorough theoretical calculations that required significant amounts of time and work. Our work with the mathematics of control theory is something that makes our modelling efforts unique within the iGEM competition, and we believe also can contribute in a meaningful way to the field of cybergenetics. Using this modelling strategy allowed us to establish the best part combinations to use in the lab to build the best possible circuits. All our modelling has been extensively documented and explained on the modelling page.
We dedicated significant amounts of time, thought and effort to producing and delivering a variety of educational content. We wanted to produce content that would go beyond the timeline of our project; creating a YouTube video series (with in excess of 100 followers) and posters educating people about synthetic biology, making our coding workshop notebook accessible and making informative powerpoint presentations that can be reused by future iGEM teams to deliver educational talks about synthetic biology. We also had a goal of ensuring our educational efforts were tailored to our audiences' needs: we conducted surveys to determine what videos people would benefit the most from, incorporated Q and A sessions into our school visits, and provided feedback forms in our coding workshops to help plan future instalments in the series. To ensure our content was appropriate for high school age students wOur educational efforts targeted a variety of ages and backgrounds: our surveys reported a huge diversity of educational backgrounds of our audiences, we engaged with children both preschool and primary school ages at the CHaOS event, spoke to students applying to and arriving at university, with the aim of introducing them to synthetic biology, and likewise with students at secondary school and sixth form. Our education efforts were thorough in their planning and execution, and have a legacy beyond this summer in the resources we created and in the children and students we inspired.
With our education efforts we attempted to include as many people from a variety of backgrounds as possible, by attending different types of events, such as Cambridge Hands On Science (CHaOS) or university open days. We also produced a variety of online content, such as Youtube series, coding notebooks and powerpoint presentations that can be used in the future to help with synthetic biology education efforts. We tried to ensure that as much as possible our media was interactive, through participating in events where discussion was key such as CHaOS events, or by ensuring that we received feedback and questions on talks and videos. This allowed us to make sure that the content we were teaching was relevant and engaging, and promoted open discussion about the values and risks of synthetic biology, which is necessary for mainstream understanding and support.
As a foundational project, our work has the potential to shape many areas of synthetic biology, and indeed society. We have therefore had extensive discussions with both academics and industry leaders and stakeholders, to understand how best to improve the project for their use cases. This meant our human practices discussions, particularly with industry, led heavily into potential applications for our project. This helped us pick out new applications that we may not have realised were relevant before, such as survival of ammonia perturbation. This led directly into our lab work, such as part of our contribution, characterising copy number response to ammonia stress, which made our project much stronger overall.
Whilst it is an extensive and important body of work in its own right, our modelling work was crucial in allowing us to design and produce the best possible wet lab project. We first utilised modelling with deterministic ODEs to establish if our circuit design was feasible, before moving on to other modelling strategies, to give us more detailed information. This allowed us to determine how stable our circuit is and test perturbations. We could then take this data and use it to predict which parts, from our multifactorial selection, would be most likely to give the best performing circuit. This allowed us to prioritise our lab work sensibly, for the best chance of success.
We successfully engineered a greatly improved version of the araBAD promoter, which are parts of a new collection of pBAD_APs. The best pBAD, AP2, has a truncated length of less than 132 base-pairs. We envisioned that the significant reduction in the promoter’s size would be of huge frugal advantage for future circuit designs. With higher performation than the 300bp wild-type counterpart. With less than 150 bp, the promoter can be synthesised by oligo-annealing, which is significantly faster and cheaper than ordering DNA parts. This could also be of great benefit for future iGEM teams from regions where DNA synthesis and delivery are often slow, allowing them to speed up project timelines. Additionally, this promoter has better performance characteristics than the wild-type promoter, creating an even greater incentive to use this as an alternative.
As part of our assembly design, we required a polycistronic transcription unit, to match the output of our protein of interest with a feedback protein. This is not easily supported in the JUMP assembly format, so to make this easier, we created a reporter including the RBS, CDS and terminator, with the overhangs of a terminator part, which we called MegaT. This allows us to use our interchangable parts for the promoter, first RBS and first CDS, while keeping with the standard JUMP format, and having a polycistronic assembly. We suggest this would be a useful model for other teams wishing to use the JUMP assembly format to create polycistronic assemblies.
We successfully engineered two improved versions of the araBAD promoter, which are parts of a new collection of pBAD_APs. All the designed pBADs have truncated length of less than 160 base-pairs, with some even showing higher performance than the 300 bp wild-type counterpart. We envisioned that the significant reduction in the promoter’s size would be of huge frugal advantage for future circuit designs. With less than 150 bp, the promoter can be synthesised by oligo-annealing, which is significantly faster and cheaper than ordering DNA parts. This could also be of great benefit for future iGEM teams from regions where DNA synthesis and delivery are often slow, allowing them to speed up project timelines.