I am pleased to be a part of the teaching team for the Doctoral Training Programme in Synthetic Biology. It was great to have the opportunity to talk to these 12 bright, enthusiastic students, who will have the opportunity to learn from and work with researchers at Bristol, Oxford, and here at Warwick over the next four years. Interestingly, while most had a strong background in Biology, they were new to the practical aspects of actually engineering organisms to express new behaviours encoded directly in foreign genes. Teaching has also given me an excellent opportunity to myself learn about some great software for gene design.
The first session I provided was early in their studies, as a Taster Session, and for this I had to assume very little knowledge of gene synthesis. What paid off for me was to base the session on GenoCAD. This is a great tool, with excellent, professional quality tutorial material, easy to follow. The emphasis on grammars for bio-constructs meant that I could explain the many ways that components such as promoters, RBS sequences, and coding sequences can be placed on DNA. The interface is very intuitive, and by having at every point all the choices for any type of component made the assembly of a whole synthetic product easy. From there, the transition to performing simulations was also very easy.
While preparing to teach this — at the last minute, as is usually the case — I found some problems and without much hope, emailed to the published email contact address. I was then surprised to very quickly hear back from Jean Peccoud himself, who acknowledged that I had come across a glitch in the standard example files and provided an immediate work-around in the form of new examples I could download and use. All in all, a very pleasant experience. The next day the students were unleashed on the system and spent a few hours designing some constructs and simulating them. Jean later told me that the 12 students had put the servers on the highest load they had seen – though on our end, it seemed the servers were perfectly able to cope with it.
This week I had a second session with the students, 4 months into their studies. By now they had learned a lot more, about genetic engineering, about cell biology, and also about mathematical foundations. So I was keen to tell them a little bit more about computational (as opposed to mathematical) modelling. My starting point was Luca Cardelli’s very readable paper about Abstract Machines in Biology, with which I could emphasize the different views one has of on the one hand the dynamics of gene regulation and on the other hand the understanding of metabolic networks, which helped to introduce the importance of stochastic simulations. The topics that followed were the systems developed to capture the process-oriented view of gene regulation, namely Stochastic Pi Calculus and Stochastic Petri Nets.
We had an extensive hands-on session with Visual GEC, which again seems to have worked very well as a teaching tool. The set of inbuilt examples showed how a high-level specification (written in GEC) is mapped to a low-level description (expressed in LBS) which in turn can be seen as a Petri Net. The fact that these three descriptions are explicit is, in my opinion, very important: many other tools I have seen would hide the low-level representations from users, which keeps them from really understanding the inner workings of the tools. Here, they got a better understanding of simulation, rather than taking it as some kind of magic. The user experience here, however, seemed less intuitive that GenoCAD, and with very few kinds of templates for regulation motifs (for example, it doesn’t seem that we can model promoters with multiple regulation mechanisms) we were left hoping for a more comprehensive tool. It seems. however, that GEC is being re-implemented, so I am looking forward to the next version and hope it can address some of the shortcomings of this one.
Talking of re-implementation, I was initially hoping to use Eugene/CLOTHO in my teaching. However, after several attempts at getting the software to work, I discovered that this tool too is being re-implemented, and the current version is basically not being maintained. So, hopefully when this new version is released there will be a third system that can be used to provide yet another vehicle for developing a better understanding of what synthetic biology is about.
I am happy to talk about these issues, and share the resources I developed – do get in touch if interested!