The only chemist who does a triple integral without complaining

August 10, 2010 11:15 PM

I studied mathematics and some computer science as an undergrad, and I loved it. I feel that my real passion, however, is in experimental sciences or engineering. What are some graduate school options for me?

I went through some twists and turns with my undergraduate education, but ended up with a decent mathematics education. However, my small experience in chemistry classes and my current job in an interdisciplinary lab lead me to believe that I would be more fulfilled with experimental science (particularly chemistry) than hardcore theory, number crunching, or code monkeying. I also feel that I have an engineering mindset - I like design and making things work. However, my school had no engineering program.

I'm wondering if there are any graduate school options that would give me the opportunity to use my math education in a more hands-on, physical setting. I have looked at robotics programs and I am also enamored with instrumentation design. My current job is in biological image processing, and I find it pretty enjoyable, but I don't want to spend the rest of my life in front of a computer. Are there any fields, or specific programs, in experimental sciences or engineering that are amenable to mathematicians? Personal anecdotes would be greatly appreciated. Thank you!

posted by scose to education (9 answers total)

What you describe is not that far off from one of the thought processes I went through recently; I just graduated (okay, sort of a lie - I will in September, but classes are done) with a BS in computer science. I wasn't really interested in grad work in computer science; I love programming, I love systems, and I love AI, but doing that as a grad student wasn't exciting. And I didn't have the grades for it anyway.

I ended up switching into linguistics, which I had a minor in, but it turns out I could've gotten into a lot of programs without the minor. People going into graduate school in linguistics come from one of three backgrounds: they have an undergraduate degree in linguistics, they have an undergraduate degree in the humanities (often something as vague as "English" or "Education"), or they have an undergraduate degree in the sciences like you and I. They tell me that's highly desirable; we have a different mindset, and come in with experience in logical thinking, technical writing, statistics, data analysis, all those things.

I'm speaking specifically to linguistics, but all those skills are relevant - and hugely valuable - to graduate school in general. If you're willing to be your lab's programmer, or to work in a lab that does computational stuff, that opens up your options a lot further. Depending on the field, this transition may or may not involve more work. In my case, even without the linguistics minor I have, starting grad school would not have involved extra "catch-up" coursework. On the other hand, I have a friend doing a PhD in computational chemistry who I believe took some crash course chemistry classes when he was starting; they accepted/hired him for his background in math and CS, and were willing to do the training he'd need to be useful to his advisor.
posted by spaceman_spiff at 11:55 PM on August 10, 2010

Did you consider fluid mechanics / aerodynamics, maybe as part of an applied physics program? Lots of opportunities to be had as a mathematician, but also suitable for the experimentally-minded like me. You mention engineering skills as well, so that would be a match too.
posted by swordfishtrombones at 12:17 AM on August 11, 2010 [1 favorite]

There are a few kick-ass applied physics graduate programs in the US that are spread over several departments. The program is centered around the physics department, but the students work in a wide variety of research groups. The applicants come from physics, mathematics, and engineering. The point is to have an interdisciplinary group. Michigan and Cornell both have these programs, they are very very competitive (they admit 8-10 students per year out of hundreds of applications). I figure there should be more of them.
posted by copperbleu at 12:20 AM on August 11, 2010

Oh, and I don't want to steal your thunder, but rather want to help avoid disappointment. Many graduate students in applied sciences and engineering will do a triple integral without complaining. Point is, having a mathematics and programming background *will* be useful, but will only give you a medium-sized advantage.
posted by copperbleu at 12:38 AM on August 11, 2010

Have a look at nuclear engineering or a related field. There's a lot of simulation work that would be right up your alley as a mathematician/programmer as well as chemical aspects that could cover your appetite for laboratory work. As an added bonus, I hear job prospects in the US are good at the time, as there is a labor shortage.
posted by Dr Dracator at 2:00 AM on August 11, 2010

There's plenty of instrument design and engineering in many experimental environmental/atmospheric chemistry/physics/geology groups. Those kinds of groups generally also have a balance between full-on experimental work (design a sampling rig for a rocket or a balloon) and modeling.
posted by janell at 2:53 AM on August 11, 2010

Neuroscience. Computational neuroscience involves lots of statistics and probability (often combined in one gooey mess, aka information theory) and lots of interesting cross-talk with the EE/CS/physics/statistics/machine-learning crowd.

And you can test some of your theories with experiments (of which, there is often quite a bit of "engineering" involved, from building your test rig, debugging and problem solving when things aren't going as you expected, to the data analysis in the end). These experiments run the gamut from in vivo neuron recordings to in vitro slice and cell culture experiments.
posted by scalespace at 10:25 AM on August 11, 2010

Thanks for the suggestions, everyone! Spaceman_spiff: how is your friend liking computational chemistry? I'm thinking about applying for that.

One thing I'm taking away from the suggestions is that it would be a good idea to find programs with a computational aspect, and then contact the professors to see if there are opportunities for the computational people to work on some more hands-on projects too.
posted by scose at 7:07 PM on August 13, 2010

(just a p.s. to second scalespace. The people I've known in neuro have split their time between building instruments, computational analysis and modeling, and poking around in brains using electrophysiology and microscopy. Sounds like a really good fit.)
posted by en forme de poire at 7:53 PM on August 29, 2010