1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Best background for computational and quantitative biology

  1. Sep 19, 2015 #1
    After two years of Biology i switched to Computer Science and i wish to gain the necessary skills to become a "computational biologist/neuroscientist"

    (Here we don't have the "major/minor" system, we can choose one subject and, at least, 2 "external" exams)

    The problems are:
    1) math level (just algebra, probability, integrals and some differential equations)...no vector, real and complex analysis

    2) no physics (!) just one optional and little general exam.

    Maybe it's the wrong degree for someone interested in simulation, system biology and computational biology.

    Better move to PHYSICS?

    (Physics Bs and Ms have mandatory courses like C, numerical analysis, computational physics.
    There is a specific Ms in Biophysics and one could integrate with some CS courses)
    Thank you!
  2. jcsd
  3. Sep 19, 2015 #2
    In our CompSci Bs there's an AI exam and possibilities to obtain a degree thesis in AI applied to biomedical analysis...bat i think that this is not enough
  4. Sep 20, 2015 #3
    I wouldn't describe myself as a computational neuroscientist -- maybe more of a computational cognitive neuroscientist, but I've done my fair share of neuron modeling, so maybe I can give some input.

    In my experience, it's common for computational neuroscientists/biologists to come from fields outside of biology or neuroscience, so that's not a problem. Computer science is common (and useful), but you're right that the math background provided by a Comp. Sci degree is generally not enough. I won't recommend a degree program, but I'll describe some of the math that I use and that I see used often in the literature:

    - As a minimum, you should be comfortable with the material covered in a standard calculus sequence up to multivariable (and maybe some vector) calculus, as well as linear algebra. This generally means three calculus courses and one linear algebra course (or two, depending on how much theory you want)

    - Neuron models (I can't really talk about anything else in Comp. Bio) are generally systems of nonlinear differential equations. They are studied both by simulating their behavior, which means solving them numerically, and by analyzing their qualitative behavior as dynamical systems. If your school has a course named something like "Non-linear dynamics", then take it. If not, buy Strogatz' "Non-linear dynamics and chaos" and work through it. Beyond that, take a course or two in differential equations just so you're comfortable with them and know how to solve them numerically.

    - Take a good calculus based statistics sequence. It will serve you well in any STEM field. Beyond that, a course in stochastic processes would probably be useful. I'm a big believer that any scientist in a quantitative field should take a good, rigorous course (or two!) in mathematical statistics (say, on the level of Casella and Berger's "Statistical Inference"), but maybe this is getting to be too much to fit into your schedule.

    - I wouldn't say that you necessarily have to take a course in numerical methods or scientific computing, but you should definitely be completely comfortable with some programming environment (Matlab is popular, but I don't like it because it isn't open source; I use R and Python). As you learn new numerical methods or computational models, practice implementing them using your language of choice.
    Last edited: Sep 20, 2015
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook