How I Ended Up Modelling Cell Division as an Undergraduate
As a third-year Biochemistry student, Filip Roch joined the Volkov lab to explore how chromosomes stay attached to microtubules during cell division. By combining cell biology with computational modelling, he studies molecular interactions that are too small and fast to observe directly.
As a third year Biochemistry BSc student, I have now been working with the Volkov lab for about 18 months. The main aspect that motivated me to choose QMUL for my undergraduate studies was the approachability of the lecturers and researchers I saw during the open days. After the first year of studying, I was eager to join a research lab. To learn by doing and contribute to scientific work - something I’ve always dreamed of. By a lucky accident, I spent some of my free time during my first-year learning to code (partially with the QMML society) and one my lecturers, Dr Volkov, happened to be looking for someone to create a computational model of the problem his group was working on.
Studying biology at the molecular level can be challenging because many of the processes we care about happen on scales that are both too small and too fast to watch directly. One of the topics we investigate in the Volkov Lab is the attachment of mitotic spindle microtubules to chromosomes during cell division. At the heart of this process is a protein called Ndc80c, stemming from the centre of the chromosome and attaching to the microtubule, which then shortens and pulls the chromosome apart. This step is crucial for successful division. In human cells there are 46 chromosomes that need to be correctly split apart, with exactly one copy going to each of the daughter cells. Ndc80c is known to play a role in signalling whether chromosomes are attached correctly. Even a single incorrect attachment is enough to halt mitosis and promote reattachment of the chromosome to a different microtubule.
The dynamic character of this interaction is what makes it so difficult to study; proteins are too small to observe them with visible light, and typical structural biology methods like X-ray diffraction or CryoEM only give us “still images”. This makes it difficult to draw conclusions about a protein as dynamic and mobile as Ndc80c. To address this limitation, I have been developing a computational simulation of the Ndc80c – microtubule interaction. The approach I’m using is called Brownian dynamics, which allows us to boil down this complex system (where a small section of a microtubule can have ~1 million atoms) to a small set of “balls and springs”. This approach allows us to run thousands of different simulations every day and makes it easier to map the experimental data to the computational results. Probably one of the most satisfying moments was when the model was working well enough that I could start stress testing it on a supercomputer – around a thousand lines of handwritten code were enough to recreate a tiny model of reality and became useful in understanding the experiments more intuitively. You can read more about Dr Volkov’s research here.
Filip Roch, BSc Biochemistry, 3rd Year, 2026.