Predicting the risk of relapse after stopping imatinib in chronic myeloid leukemia

IMODay1To escape the Montreal cold, I am visiting the Sunshine State this week. I’m in Tampa for Moffitt’s 3rd annual integrated mathematical oncology workshop. The goal of the workshop is to lock clinicians, biologists, and mathematicians in the same room for a week to develop and implement mathematical models focussed on personalizing treatment for a range of different cancers. The event is structured as a competition between four teams of ten to twelve people focused on specific cancer types. I am on Javier Pinilla-Ibarz, Kendra Sweet, and David Basanta‘s team working on chronic myeloid leukemia. We have a nice mix of three clinicians, one theoretical biologist, one machine learning scientist, and five mathematical modelers from different backgrounds. The first day was focused on getting modelers up to speed on the relevant biology and defining a question to tackle over the next three days.
Read more of this post

Are all models wrong?

GeorgeEPBoxGeorge E. P. Box is famous for the quote: “all models are wrong, but some are useful” (Box, 1979). A statement that many modelers swear by, often for the wrong reasons — usually they want preserve their pet models beyond the point of usefulness. It is also a statement that some popular conceptions of science have taken as foundational, an unfortunate choice given that the statement — like most unqualified universal statements — is blatantly false. Even when the statement is properly contextualized, it is often true for trivial reasons. I think a lot of the confusion around Box’s quote comes from the misconception that there is only one type of modeling or that all mathematical modelers aspire to the same ends. However, there are (at least) three different types of mathematical models.

In my experience, most models outside of physics are heuristic models. The models are designed as caricatures of reality, and built to be wrong while emphasizing or communicating some interesting point. Nobody intends these models to be better and better approximations of reality, but a toolbox of ideas. Although sometimes people fall for their favorite heuristic models, and start to talk about them as if they are reflecting reality, I think this is usually just a short lived egomania. As such, pointing out that these models are wrong is an obvious statement: nobody intended them to be not wrong. Usually, when somebody actually calls such a model “wrong” they actually mean “it does not properly highlight the point it intended to” or “the point it is highlighting is not of interest to reality”. As such, if somebody says that your heuristic model is wrong, they usually mean that it’s not useful and Box’s defense is of no help.
Read more of this post