Undergrad projects in maths and biology

[PhDP]

Hello

I'm thinking about creating an online group of undergrad biology/mathematics students to create projects in theoretical/mathematical biology (I'm not talking about statistics; stats are well established in biology).

The fact is, there are so many opportunities to use mathematics in biology, sometime, even basic models have not been analysed. But most biologist have little training in maths, and they often make simple mistakes (the worst part; it gets published). It's not really better in mathematics departments; very few have researchers interested in biology.

So, there are very few opportunities for undergrad student, at least in many universities, to do research in biomaths/theoretical biology. This is a little odd; so many articles in biology journals rely on maths. On the brighter side, because researches in those domains don’t require much equipment, and because many subject have barely been explored, it's a fertile ground for undergrad research projects.

Of course, many aspects of biomaths/theoretical biology have been analysed by many scientists, and it would be hard for a bunch of undergrad to do research without spending an unrealistic amount of time. On the other hand, many models have not been analysed by anyone, there's just not enough people with the proper training to do this job.

I don't want to create a monolithic group where everyone would do the same thing; I would like to simply assemble many motivated undergrads, with different backgrounds, it could be biologists with little training in maths, or mathematicians with little training in biology. Motivated people. Anyone could discuss project ideas, and teams could be assembled on specific projects, with the objective of publishing, of doing serious research. Of course, if would also be an opportunity to learn and to share knowledge, articles, et cetera...

I know if I'm looking for people with true hybrid training in biology and mathematics, I won't find many. But I know many biologists have read articles in journals and are frustrated by their inability to understand the maths behind theory, and I'm sure many mathematicians would like to find new opportunities.

I really think people don't realize how much work there is to be done in this domain. I've recently submitted a "Letter to the Editor" after I saw that a model, published in 3 serious journals (including "Science" and "PNAS") had a serious construction flaw (it's not even a matter of opinion and interpretation; it's just plain incoherence that can be demonstrated with basic algebra).

I would really like to have some feedback on this idea, and anyone interested can contact me by email or MSN (it's the same GMAIL address).

 

[J77]

One problem is that the term "biomathematics" covers many subfields.

For me, I think of dynamical systems.

However, even then, you have medical fields, infection models, population dynamics, neuroscience...

When I've searched through the literature, I've found it quite hard to find a specific field, which required the use of advanced techniques, to work on.

 

[Moonbear]

The number of students with "hybrid" training are increasing. The usefulness of mathematical modeling (along with the necessity of using math and physics for things like electrophysiological recording from cells and developing better methods of interpreting MRI data) have made this a more popular area than for previous generations. It's true that until now, there has been a gap between the fields where someone either has a strong biological background OR a strong mathematical background and then struggled to put together meaningful connections with the other field. This is changing with students receiving training across both disciplines (often with mentors in two departments to balance their training).

If you saw a flaw in a published mathematical model, it also would have been helpful to contact the author of the article directly. They either would have appreciated the insight, or helped you see why something was there that looked inconsistent at first but is explainable (I don't know which). BTW, PNAS shouldn't considered too serious of a journal in biological sciences...around here, we call it Pain N the AS*. The papers are invited by members, and not given as rigorous of peer review as other journals get, so people like to slip in more speculative ideas that they can't get published elsewhere. I consistently find major flaws in the studies reported there, and yet they get cited over and over. It can be interesting for thought-provoking new hypotheses, but not very useful for sound conclusions based on strong experimental evidence.

 

[Chris Hillman]

BTW, PNAS shouldn't considered too serious of a journal in biological sciences...around here, we call it Pain N the AS*. The papers are invited by members, and not given as rigorous of peer review as other journals get, so people like to slip in more speculative ideas that they can't get published elsewhere. I consistently find major flaws in the studies reported there, and yet they get cited over and over.

Indeed, my pet peeve is "approximate entropy", which was published in PNAS and has widely been applied in cardiology (!) and other areas of applied dynamical systems, but which is IMO pretty much snake oil. And potentially deadly, given its sometimes application to tricky medical diagnoses.

 

[PhDP]

J77,

About dynamical systems... if you look at mathematical analysis of the Rosenzweig–MacArthur model, or more recently, of many ratio-dependent models, you'll find them. But some other system of differential equations have not been analysed at all. In fact, when I thought about trying to find people do work on mathematical/theoretical projects, I was thinking about a recent system of differential equations in ecology that was not analysed by anyone (... I was also thinking about a well-known gene involved in the development).

Moonbear,

I'm getting a true "hybrid" training; a little more than 1/3 of my B.Sc in math, a little less than 2/3 in biology, and I can say; I could work in either math or in biology, but there's no opportunity at my university to do both. I think its unfortunate considering how much work has to be done in this area.

If you saw a flaw in a published mathematical model, it also would have been helpful to contact the author of the article directly. They either would have appreciated the insight, or helped you see why something was there that looked inconsistent at first but is explainable (I don't know which).

You're probably right. I have not contacted the authors, but I've contacted a biomathematician to be sure my reasoning was correct. It was quite ironic, as the mathematician in question had already busted a mathematical model by the same authors :)