Biomathematics major in undergrad for computational neuroscience in grad school?

  1. Nov 21, 2012 #1
    What do you think about a major such as bioinformatics versus a major in physics and minor in mathematics or a major in 'cell biology& neuroscience' w/ minor in computer science in terms of preparing me for computational neuroscience -- taking in mind that I've completed courses such as linear algebra and classical mechanics. I am interested in computational models of brain processes.

    Here is the syllabus for bioinformatics:

    R120:101 General Biology I (3-3-4)
    Chem 125 General Chemistry I (3-0-3)
    Math 111 Calculus I (4-1-4)
    HUM 101 English Composition: Writing, Speaking, Thinking I (3-0-3)
    BNFO 135 Programming for Bioinformatics (3-0-3)
    CS 107 Computing as a Career (1-0-1)

    2nd Semester

    R120:102 General Biology II (3-3-4)
    Chem 124 General Chemistry Laboratory (0-2-1)
    Chem 126 General Chemistry II (3-0-3)
    Math 112 Calculus II (4-1-4)
    BNFO 136 Programming for Bioinformatics II (3-0-3)

    SECOND YEAR: 1st Semester

    R120:201 (Foundations of Biology) (3)
    R120:202 (Foundations of Biology laboratory) (1)
    R120:352 Genetics (3)
    CS 241 Foundations of Computer Science I (3-0-3)
    Math 333 Probability and Statistics (3-0-3)
    HUM 102 English Composition: Writing, Speaking, Thinking II (3-0-3)

    2nd Semester

    R120:356 Molecular Biology (3)
    Chem 243 Organic Chemistry I (3-0-3)
    Social Science (GUR) (3)
    BNFO 240 Principles of Bioinformatics II (3-0-3)
    Econ 201 Economics (3-0-3)
    CS 207 Computing and Effective Communication (1-0-1)

    THIRD YEAR:1st Semester

    Phys 111 Physics I (3-0-3)
    Phys 111A Physics I Laboratory (0-2-1)
    * BNFO 340 Data Analysis for Bioinformatics (3-0-3)
    HUM 211 The Pre-Modern World (3-0-3)
    HUM 212 The Modern World (3-0-3)
    Hist 213 The Twentieth-Century World (3-0-3)
    CS 431 Database System Design and Management (3-0-3)
    Elective (Free) (3-0-3)

    2nd Semester

    Math 337 Linear Algebra (3-0-3)
    Mgmt 390 Principles of Management (3-0-3)
    Elective SP (Specialty Elective) (3-0-3)
    CS 435 Advanced Data Structures and Algorithm Design (3-0-3)
    IS 350 Computers, Society and Ethics (3-0-3)
    PE (Physical Education) (1)

    FOURTH YEAR:1st semester

    * BNFO 482 Databases and Data Mining in Bioinformatics (3-0-3)
    Eng 340 Oral Presentations (3-0-3)
    Eng 352 Technical Writing (3-0-3)
    Elective SP (Specialty Elective) (3-0-3)
    Elective SP (Specialty Elective) (3-0-3)
    Elective (Free) (3-0-3)
    PE (Physical Education) (1)

    2nd Semester

    BNFO 491 Computer Science Project (3-0-3)
    HSS Cap (Capstone Seminar:GUR) (3-0-3)
    Elective GUR (Lit/Hist/Phil/STS:GUR) (3-0-3)
    Elective SP (Specialty Elective) (3-0-3)
    Elective (Free) (3-0-3)
    CS 407 Professional Development in Computing (1-0-1)
  2. jcsd
  3. Nov 21, 2012 #2
    There's a notable lack of differential equations/non-linear dynamics and probability/statistics in your schedule, the former being absolutely necessary. Honestly, a huge number of those courses don't really seem all that related to your goal. In computational neuroscience, mathematics tends to be far more important than biology. Organic chemistry, for instance, will probably not be terrible useful.
  4. Nov 21, 2012 #3
    Hmm, well then should I take these mathematical classes on the side of bioinformatics? I considered bioinformatics in the first because I thought it would prepare me for computational neuroscience in terms of the computer science aspect. It feels like in all of the majors there is going to be a number of things you will never really use, such as in a physics major.
  5. Nov 21, 2012 #4
    If you can program well in a language or two, and have some experience building neural networks and spiking models, you're fine; there's no need to study (e.g.) data structures. A huge number of your courses are either not helpful, or are replacing courses that would be more helpful. Regarding the biology, a single good course is neurobiology/neuroscience should be more than enough to get you in the door as far as comp. neuroscience goes; you can pick up the rest later. A fair number of people working in the field have never taken a single biology course in their lives.

    Your math coursework is not nearly enough. How can you expect to study neuron models if you have no experience working with dynamical systems (i.e. the language of computational neuroscience)?
  6. Nov 21, 2012 #5
    Hmm that comes as a surprise, wouldn't someone generally need a major or at least a minor in computer science to run simulations of the brain?

    About taking a good course on neurobiology/neuroscience courses, I'll have to take Bio 1 & 2 prior to that which shouldn't be a problem.

    Thanks, I really didn't expect much mathematics for computational neuroscience, at least for my area of interest, that is modeling the human brain and running simulations. I have the option of doing a biomathematics major . here is the list for course synopsis

    Edit: Just realized, in the title where it says "biomathematics," I meant bioinformatics
  7. Nov 21, 2012 #6
    Computer science is a extremely theoretical field; if you're only interested in simulation, majoring/minoring is comp. science is not at all necessary (it wouldn't necessarily hurt, but a huge number of those courses are unnecessary and unrelated to your interests). A huge number of scientists in the field are "home grown" programmers who became fluent in a language or two along the way.

    Brain modelling draws enormously from non-linear dynamics (unless you're doing it at an extremely simplistic and unrealistic level (e.g. artificial neural networks)), so you'll need to be very comfortable with differential equations. The biomathematics major you linked to looks like it's pretty comprehensive, though quite a few of the biology courses are unnecessary (nothing wrong with that, though; everything else is there and you may find that other areas of mathematical biology interest you).
  8. Nov 21, 2012 #7
    Just to throw in some thoughts here. My EE program has a bio-medical specialty that is pretty much just brain modeling classes. The classes required for those courses are Diffy Q, signal processing, probability/stats and programming but no biology courses.
  9. Nov 22, 2012 #8
    Don't do it. I do bioinformatics (which is awesome!) but it's only worth it at the M.S PhD level. Otherwise you are missing out on a little useful undergrad courses. You need more differential equations, more physics, more programming, more linear algebra, less biology, and some research experience.

    Edit. Nearly forgot, more Stats!

    As you can see, more mathematics than biology!
    Last edited: Nov 22, 2012
  10. Nov 22, 2012 #9
    What did you study in undergrad?
  11. Nov 22, 2012 #10
    I completed a degree in Mathematics and English with two minors in Physics and Comp Sci.
  12. Nov 24, 2012 #11
    Thanks for your input Drummingatom and Marnemath.

    Thanks a lot, you've helped quite a bit. So I'm going to stick with a double major in biomathematics and physics. I've visited the mathematics library in my university and looked at all of the books about computational neuroscience, you were quite right -- lots of physics and especially mathematics.

    One question I have though is whether or not I should do an applied option in physics or the professional option? I am thinking I should do the professional option even though it will delay my graduation, the applied option doesn't even contain classical mechanics or electromagnetism. From what you have told me so far I think it would be the obvious choice to go with the professional option, it would also leave me open to the possibility of quantum computing if I ever choose to enter the field.
  13. Nov 25, 2012 #12
    Number-Nine, disregard the past question. I have a more important one for you and the others.

    I won't be able to complete a double major in biomathematics and physics (BS). What do you guys think about the importance of completing a biomath major versus a physics major? I spent a lot of time yesterday looking at peer reviewed articles in computational neuroscience and I don't see a whole lot of physics being applied. It is all just math really. However, I realized just how important problem solving really is, something that physics trains you well for. The thing is though, I should pick up a lot of problem solving abilities even more-so in computational biology/neuroscience from the biomathematics major.,

    Also what do you guys think of the importance of experimental courses in physics in terms of preparing you for computational neuroscience (or how would grad school look at it)? I was thinking of double majoring in biomathematics and physics (BA), the BA in physics will give me a lot of freedom in terms of classes. In addition, I will also be able to skip the experimental classes that I don't enjoy a whole great deal.
    Last edited: Nov 25, 2012
  14. Nov 25, 2012 #13
    I don't imagine you'll use most of the physics you'd study, and I really can't see how physics laboratories would help you do research in computational neuroscience. Some neuroscience research borrows from techniques is, say, classical mechanics (e.g. mean field theories), but "problem solving practice" in the abstract is not enough of a reason to major in physics, IMHO. Study as much math as you can, and get some experience in neuroscience over in the biology department.
  15. Nov 25, 2012 #14
    Great, thanks! I was leaning toward biomathematics anyways so you've cleared my guilt for not choosing a BS in physics. I like physics and everything but the curriculum sucks up the fun I find in it, I would rather study it in my spare time for leisure. Granted, I'm still planning to finish the classical mechanics sequence and the electromagnetism sequence as suggested by someone on this forum (which will also satisfy my BA in physics).
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?