The idea for this series was born out of our partnership with the Vienna iGEM team. Because their team was not very familiar with modelling, we thought it would be a good idea to run a series of workshops to help them develop the skills they would need to build their own model.

To grow our collaborations and expand our reach on the wider iGEM community, we reached out to the Cambridge team – who was going to run an introductory workshop to Python – to ask if they would like to join our collaboration, allowing all of us to work together to introduce the iGEM community to more complex topics in modelling. They agreed that combining forces would allow us to have a wider impact, and so, our series Coding with Biology was born.

Together we designed a short course to cover three areas:

• A basic introduction to coding on Python
• How to create a simple flux model and code it into Python
• How to design more complicated models

After advertising the 2nd episode, the Sheffield iGEM team reached out to us because they were also interested in running an episode.

Sheffield’s bonus workshop was run on a different and newer programming language, Julia. We chose this language because it combines the speed of compiler languages (e.g. C++, FORTRAN) with the interpretability of modern languages (e.g. Python, R), and it allowed us to introduce the idea that there are many programming languages out there and that each has their strengths and weaknesses.

Thus, between all of us, we collaborated to make 4 workshop lectures, each hosted by a different team and designed to build off the previous session.

The Coding with Biology series was designed to have each team run a separate workshop, to allow each of us to bring our expertise to the table. To ensure that we were collaborating and that we were providing the best possible experience to our attendees, we run surveys before and after each workshop.

The surveys run BEFORE the workshops helped us to identify our attendees’ previous experience in programming before the workshop, allowing us to tailor each session to our audience.

The surveys run AFTER each workshop were always followed by a debrief meeting between the teams. This allowed us to brainstorm how to build up from the feedback and create a two-way dialogue with our audience. Some key examples of this can be seen in the graphic below:

---

Another important point of collaboration between all our teams came regarding the advertisement of each of our sessions. We ensured that all four teams were promoting the events on their social media to attract a larger audience. We also came up with the idea to promote the event through some of the UCL departments’ mailing lists, allowing us to reach students from the Biochemical and Chemical Engineering departments at UCL.

Our course was attended by > 60 people with an average of 83% saying that they found the sessions to be successful. Our session, in particular, saw the highest level of “success”, I.e., the highest number of people who left the workshop feeling like they had learnt what we had sought out to teach, which in our case was how to build and code your own model.

We were responsible for the 2nd episode of the series: learning how to build your own model and run it in Python.

The episode was divided into 2 sections:

---

1) How to create a model.

For this first part of the workshops, we counted with the help and support of Prof. Chris Barnes from UCL, who presented a few slides on the importance of modelling biological systems and how one would go about creating a model.

You can find the recording of the session here. The passcode is 4^m0o5=

---

2) How to run it in python.

The second section of the workshop focused on how to write your model in a Python script.

You can find a link to the Jupyter notebook that we used here.