Bioinformatics From Different Background With Different Education

Question

[Fabian Bull]

I am following the internet bioinformatics community for quite some time now, checking forums, social networks, twitter and stackexchange.

I often wonder how much the skill level (specially in theory) is differing between people. Why is that?

I learned (theory):

1. Advanced bayesian statistics
2. Introduction to basic bioinformatic algorithms
3. Machine learning applied to biological problems

What I didnt learn (practice):

1. Use BLAST or use any other program
2. Talk to a biology guy
3. Software development

What did you learn? Did you visit any university or did you switch from biology or cs? Did your courses differ hugely? Should there be more practical courses?

[Pierre Lindenbaum]

community wiki ?

[Lyco]

Could you briefly explain what 'community wiki' means, or where I can read anything about it? I checked the FAQ (in vain) and don't see any other help page or documentation linked on the Biostar pages

[Keith Callenberg]

I think these questions are too vague. I think this post should have been narrowed down to something like: "what proportion of courses in a bioinformatics degree should be theory as opposed to application?" or "What proportion of students in bioinformatics programs come from computational as opposed to biological backgrounds?" I think these questions have both already been posted however.

Answer 1

[Nicolas Rosewick]

In bioinformatics the most important is to understand the biological question. So talking to biology guy is very important.

[The Original Gtk]

I would add that biology includes theory, lots of it. "Talk to a biology (person)" is, barely, scratching the surface of what a bioinformaticist should expect of themselves. How they learn it, and what they are most expert in, of course, will vary widely

Answer 2

[Stevelor]

I studied bioinformatics at university. I learned a lot of theoretical stuff....a lot about background of algorithm, tools, patterns etc.... and a bit about biology.[?] The most important thing i think is to understand the biological background if you work with biological data....You can do so much more with your data if you are really understand what is behind it!!!!Not only using some tools :/[?][?]I am really happy to have close contact to biologist, chemists etc. here at a research institute....This was completely missing at university!!

Of course you should never forget: Nothing makes more happy than coding :D ^^

Cheers!

[Fabian Bull]

Thats pretty much the same I am experiencing now. Just a little to late. :D

[Stevelor]

y?? what are you doing?? And which way did you go??

Answer 3

[Akk]

I come from a computer science / scientific computing background. I did numerical analysis, which at my department boiled down to the development of parallel numerical methods for the solution of large scale partial differential equations describing fluid flow (the Navier-Stokes equations). Lots of C, Fortran 77 and Fortran 90 programming on massively parallel machines, and linear algebra.

What I learnt from all that, which I still have use for, is object-oriented analysis and design, computer architecture, many programming languages (being multilingual helps thinking outside of the restrictions of a particular language), the nuts and bolts of basic data structures, and just generally algorithmic thinking.

I've worked in bioinformatics for little over 10 years since, either as a software developer or as a system engineer. Although I have no higher degree in biology, the places where I've worked have allowed me to do what I do best (API development, algorithm design, and database design) together with people that I've learnt the necessary bits of genetics and biology from, including users of the software I produce.

From a software development point of view, a gene (or whatever you have) is just another type of data. Yes, it's important to understand where it's coming from and what it's going to be used for, especially if one is developing a database schema and API for storing and retrieving the data (or a web site, even), but there's a layer below which the problem turns into pure computer science and where biology will make things messier rather than clearer.

Sorry for the long post, I'm rambling.

[Fabian Bull]

Thaty pretty much my point of view. :D