Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

Bioinformatics and Computational Biology

By iGrrrl in Technology
Thu Mar 15, 2001 at 02:28:33 PM EST
Tags: Science (all tags)

These fields may not have a place in the mind of the average computer and programmer type, but perhaps they should. Several years ago I helped put together a career paths series on "things you can do with a biomedical Ph. D. that don't necessarily involve academia." When we tried to find someone to speak on Bioinformatics we had trouble filling the slot. The usual response from the Human resources offices of the many Boston area biotech firms ran along the lines of: "We'd love to send someone to talk to you, but we don't have anyone. If you know someone interested in a bioinformatics job, please send them our way."

What do these words mean?

The best definition of bioinformatics I've run across comes from UCLA's graduate program:

Bioinformatics is the study of the inherent structure of biological information and biological systems. It brings together the avalanche of systematic biological data (e.g. genomes) with the analytic theory and practical tools of mathematics and computer science.
Computational biology seems a bit harder to define. It stretches from modeling single molecules to creating computational neural networks that help researchers understand real networks of living nerve cells. Neither of these things are the same as evolutionary computing. However, all kinds of computational thinking can be applied to biological problems, such as can be seen at NASA's Evolutionary Cell Computing Project.

People get into this fusion of biology and software from different angles. Most often researchers have a background in biology and teach themselves the computer science and programming skills. Sometimes a person with experience with computers finds themselves working in biological applications. In the past few years degree programs have sprung into existence, but they vary in emphasis. Some stress computation, and some stress biology and biophysics. (A quick Google search will turn up programs from all over the world.) No matter how people come to the field, there seems to be demand for more.

What does this really mean in practical terms?

There are academic applications, corporate jobs, and even practical work for people who have an interest in computing and biology. The biggest problem for academics and biotechnology-based pharmaceutical companies lies in making sense of the wealth of genomic data. One big relational data problem has not been solved, which is how to correlate genetic data with biochemical data. Proteomics is the word replacing genomics as the hot new field.

If it seems esoteric and academic, there are medical applications. The data generated by gene chip experiments are provocative -- knowing what genes change in cancer or other disease states. Combining genomics with proteomics means that not only will there be diagnostics for disease, biomedical researchers may be able to develop a clue as to what the genes actually do. But these applications hang on finding a way to sift through and correlate information, to find patterns which might emerge from a wealth of data.

Computational biology also has impacts outside of medicine. An attempt to model a biological system, the sense of smell, led to the development of an artificial nose geared toward the detection of land mines. Essentially, it's an attempt to re-create a dog's nose. The Tufts artificial nose grew out of two biologists interested in olfaction. Computer models of nerve cell circuits in the salamander became the basis for the neural networks of an artificial chemosensing device.

That all sounds pretty esoteric

My prejudice is clearly academic. Bioinformatics and computational biology could be rich fields for someone interested in both what computers can do, and what can be done with computers. For those of more practical mind, there are eleven job openings listed at Aventis Pharmaceuticals alone. Two hundred and eight US jobs show up on a quick monster.com search, admittedly ranging from sales rep to Ph.D.-level specialist, including software engineers and UNIX system administrators. The problems tend to be interesting, and the toys are often up-to-the-minute. There are database and relational database problems. There are pattern-finding computational problems. There could be a career you might not have considered before

Other Links of interest:

  • Human Genome project bioinformatics page
  • The Open Lab at bioinformatics.org
  • Sponsors

    Voxel dot net
    o Managed Hosting
    o VoxCAST Content Delivery
    o Raw Infrastructure


    Related Links
    o Google
    o UCLA's graduate program
    o evolutiona ry computing
    o Evolutiona ry Cell Computing Project
    o Google search
    o Proteomics
    o gene chip
    o Tufts artificial nose
    o My prejudice
    o Aventis Pharmaceuticals
    o monster.co m search
    o Human Genome project bioinformatics page
    o The Open Lab at bioinformatics.org
    o Also by iGrrrl

    Display: Sort:
    Bioinformatics and Computational Biology | 11 comments (6 topical, 5 editorial, 0 hidden)
    Caught In Two Minds (1.00 / 1) (#4)
    by dave.oflynn on Thu Mar 15, 2001 at 12:46:53 PM EST

    I'm torn with this article. The subject is interesting, partly because it's a rapidly developing field, and partly because it will have a lot of influence on the future of society.

    However, the article itself doesn't seem to be saying anything beyond "there's not a lot of bioinformatics people around"... it reads like a job ad rather than a discussion piece. Maybe that's what a field this new needs... I dunno. As I said, I'm torn...

    OK, +1 Section Only. But just this once ;-).

    Protein folding and protein design (4.00 / 2) (#7)
    by mbrubeck on Thu Mar 15, 2001 at 03:39:47 PM EST

    My math department recently hosted a talk by a mathematician working for a company that does protein design. Unfortunately I don't remember the speaker's name or the company, but I can summarize the interesting points of the talk.

    Protein design is different from the more well-known protein folding problem. The protein folding problem attempts to determine the 3D structure of a given chain of amino acids. This is important because a DNA sequence only tells us the sequence of acids, while the physical structure of the protein determines its biological and chemical properties.

    The protein folding problem is very hard. The complexity of current algorithms is such that it is infeasible for proteins of more than about 30 amino acids. Because of the nature of the problem, approximations algorithms don't help ("a miss is as good as a mile"). However, our speaker described recent and ongoing research on improved algorithms that make the problem much more tractable for small proteins -- still no larger than 50 amino acids. Unfortunately this still leaves out most of the important proteins found in biology.

    Protein design is a sort of inverse to the protein folding problem. It is the problem of finding a chain of amino acids that has a particular set of physical or chemical properties. It is easier than protein folding in several ways -- for example, there are many solutions to a given protein design problem, while protein folding solutions are unique. The speaker described his work on protein design algorithms.

    There seems to be a lot of work going on in this area, and being actively applied to medicine and biotechnology. Sorry I don't have any more specific references. Hopefully I've given enough of a background that you can find more information on your own, if you're interested. This list of projects at Caltech's Mayo Group might be a good starting point.

    this is where i'm going (2.00 / 2) (#8)
    by StackyMcRacky on Thu Mar 15, 2001 at 04:06:38 PM EST

    my official training is as a microbiologist. i sold out and became a unix admin.

    some fo the guys i used to work with in the lab started their own company (don't laugh at the site, we're still kind of in cognito mode), and hired me on as the IT dept. after i get the basic infrastructure together, i'll be writing/designing all kinds of genomic data type programs. should be interesting.

    How you could help (3.50 / 2) (#9)
    by Alik on Thu Mar 15, 2001 at 06:04:07 PM EST

    Most folks here probably know of it, but just for the record, folding@home is YA distributed computing project that works on the big nasty evil protein folding problem. They've also got another one to analyze genomic data.

    curiously enough... (4.66 / 3) (#10)
    by heuermh on Fri Mar 16, 2001 at 11:31:43 AM EST

    The best way to get your feet wet in the field of bioinformatics would be to first read both human genome papers:

    Science 291(5507):1304-1351 -- the Celera paper
    Nature 409(6822):860-921 -- the public paper

    (both are available in full-text on the respective journal sites)

    Next, jump into some code.


    Curiously enough, I work in bioinformatics for Aventis.

    What does this mean? (none / 0) (#11)
    by cable on Fri Mar 30, 2001 at 04:17:39 PM EST

    First I never took any really complex biology classes, so I am barely understanding the basics here. I am trying to understand what this means. Excuse me if I get anything wrong, I'm sure it will be pointed out to me. When I wrote this, there was less than 10 posts/comments on the story.

    Does this mean that you can eventually make a Bio-Chip that can think? Or encode data onto DNA and Genomes for use in a computer system? Or is it more like cracking the code on DNA or Genomes and redesigning it to screen out the genes that cause cancer and illness, or adding ones that cause longer life and stronger bodies? Perhaps taking out the one that can cause obesity or that would slow down the metabolism that can cause obesity?

    Or is it something else that I am not quite grasping here?

    Only you, can help prevent Neb Rage!

    Bioinformatics and Computational Biology | 11 comments (6 topical, 5 editorial, 0 hidden)
    Display: Sort:


    All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
    See our legalese page for copyright policies. Please also read our Privacy Policy.
    Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
    Need some help? Email help@kuro5hin.org.
    My heart's the long stairs.

    Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!