Some time ago I came across this article “Biology’s next microscope, mathematics’ next physics” that gave me many ideas of how to address the collaboration that I was proposing in my grant, and that has been going on for years in our lab. And as far as it seems trivial and normal for me now, I realized this collaboration between modelers and biologists still is not a common topic among scientists.
I came to work as a postdoctoral associate here in the US back in 2008 and was specially delighted with this new experience. At first, I was very intimidated with the mathematicians from our lab, specially when they start with their equations and numbers – I was never a big fan of math at school. But after a while I started to understand that it was not about the math per se, but about what the math could bring me! A mathematician mind functions in a very different way than ours, biologists. We tend to complicate things, to be very prolix. They are incredibly sharper, they go straight to the point! Sometimes I think that they simplify too much things, but in the end it’s like they are summarizing things in a box and arrow diagram by the end of our paper.
Practically, despite helping with data analysis and interpretation of results, a typical new project starts with a review of the literature, as any other project. I am the type of scientist that often gets lost between an infinite number of Pubmed open pages in my browser, but who isn’t? Whenever things start to get too complicated it’s time to ask for help. We present all the science collected so far and they transform all this information in equations of a simplified model of what should be happening the micro system that you are working on. Often there is a lot of information that they don’t use but also they ask for other info that you would not think about – does this neuron X has receptors for Y? What happens with X if you increase Y? Then you look for the answers of this questions in the literature (or at least what you can get) and they build a mathematical model from all of it.
It’s show time! We start to run simulations now! Block X, see what happens with Y, fine tune the model to best fit what it is known in the literature. Now we get to the fun part – ask questions about your hypothesis! I predict that if we increase Y, X will decrease and we won’t have a response. Put this info in your model, run the simulations and see what happens. If your hypothesis is correct in the model, it’s time to go to the bench and run the experiments – of course the model is not correct 100% of the time. But my PI use to say that we learn more adjusting the model than when the model is always right on its predictions.
It is fun and very useful! I am sure that even after I get a permanent position I will continue this fruitful collaboration with mathematicians. And I suggest you try to collaborate with them too, you will not regret it!