Let’s compile a Biotech for IT folks book and publish it!

IT people are the dominant high tech tribe today and especially on the web. But biotechnology (BT) is the next infotech so no wonder that the IT crowd is growingly curious about everything biotagged on the one hand, while they are usually not too savvy in DNA-RNA-protein-organelle-cell-tissue-organ-organism related matters on the other hand. Check for instance Tim O’ Reilly at Nature: science meets bored tech-savvyness to find new things.

And what can biotech bloggers do in order to meet the growing demands: well here is a little conversation from my twitter channel in the last 20 minutes:

Biotech for IT folks

5 thoughts on “Let’s compile a Biotech for IT folks book and publish it!

  1. while I’m glad that IT is showing interest on bio, we need to make sure that we both learn the basics from each other first. They need to learn the basics of biology from us and we, biologists, need to learn at least some basics from them (ie perl) to be able to communicate our needs.

    I’m glad this is getting started as IT is urgently needed in bio nowadays, specially in genetics/genomics. Until a few months ago I worked in a lab making (and analyzing) microarrays. They created quite an amount of data -sometimes impossible to open in a regular spreadsheet like Illumina data- that is impossible to go through without bioinformatics knowledge.

    Luckily sometimes our lab was able to work with software companies that tried to help us analyze, but most of the time it was analysis done by students in the informatics dept.

    The biggest problem I found when working with these students is that most of the time they didn’t know exactly what we, biologists, were/might be interested in. Pathways? relation between pathways and expression? copy numbers? tissue types?

    This is because a lot of people I worked with took for granted their biology knowledge and did not try to explain the biology behind the experiment, and if they did, made it so complicated for the IT person to understand.

    Same thing happens when the beautiful program analyzes your data but you aren’t sure of what it exactly means!

    Basic, clear communication is the first step to be taken…anyway, i probably already rambled too long…

  2. I’m enrolled in a bioinformatics program and am quite surprised at how far the field has progressed despite a sizable IT presence. Many sophisticated algorithms have been designed to tackle sequencing challenges and for processing the large arrays of biodata. Many of these were designed a decade ago or more.

    The challenge I see is that there hasn’t been much in the way of sophisticated tool development that integrates the various data sets that high-throughput lab automation are now starting to generate. The complexity of this data climbs quickly when only adding a few parameters which in turn drives hardware requirements rather quickly as well.

    Another challenge is that much of traditional IT today is focused on business or engineering data for manufacturing or fields like telco. Biological data is a much different beast. It’s applied mathematics, genetic algorithms, statistics, hidden markov models, etc. The data sets and algorithms involved in biology have a much steeper learning curve than those produced by IT driven markets such as finance, insurance, etc. Bringing this crowd together with biologists isn’t going to happen overnight.

    I think it’s going to take a generation for the multi-disciplined bioinformaticians to evolve but when they do, I suspect much more biological work will be done at a computer workstation linked to supercomputing centers than in a lab.

  3. Jim: I agree; it happening but, definitely not overnight.

    What I have been seeing is a lot of Biologists turned into bio-statisticians after not being able to explain how X affects Y in more than one way to IT/the software companies.

    I do hope the next generation to be able to use our current data, and to be able to analyze in a fast, efficient way (For us, one in-house array took overnight to be analyzed…and when you’re handling 80 samples…)

    So much data, so many interesting correlations and answers to be found!

  4. I agree whole heartedly. We need those intimately familiar with the Semantic Web, and intelligent database design to help medical researchers. This is absolutely necessary in order to help facilitate key medical research breakthroughs. There are huge bottlenecks in the way data is processed and shared. There needs to be an ongoing dialogue between IT professionals and medical researchers in order to develop high level collective intelligence platforms. This is something I am endeavoring to do.

Comments are closed.