The Interactome in the context of drug design

Wednesday, March 31th, 2010 by hinrich

Workshop Poster

Last week I attended the workshop "New Approaches in Drug Design & Discovery" with the topic "The Interactome: From Atom-Atom Contacts to Networks in Systems Biology". It took place at Schloss Rauischholzhausen in Germany and coincided with the first warm and sunny days of this year. The meeting was very well organized thanks to the whole group of Prof. Gerhard Klebe of the University of Marburg.

As molecular modeling is certainly not my area of expertise, the workshop was a great introduction to the issues and opportunities of current drug design approaches. However, in the context of the IWT-sponsored project on compound profiling using whole genome gene expression analysis that I am currently running, it is of high relevance.

Here are some statements / questions / provocative thoughts from the meeting that I find particularly remarkable:

Can we really reduce the complexity of biology to simple rules?
Systems Biology is about putting together rather than taking apart.
Pathway diagrams often tell us only who is interacting but not how they are interacting.
We have a lot of information about the structure of an interaction component but not much about the structure of the interaction.
The binding of a small molecule to a protein can induce the formation of a binding pocket that fits well the small molecule.
Bad news for automatic docking: quite some differences in structure of the binding pocket can be induced depending on the binding inhibitor compound.
The majority of compounds are not selective. A given compound is currently known to have on average 6 targets.
Only 10% of the kinome is covered by publications in PubMed.
A major disadvantage of RNAi technology is that you need to wait for protein depletion and that can take some time (in contrast to e.g., chemical probes).
You may not want to know too many details!
You should not generalize too early!

Posted in Science

Leaving the Trough of Disillusionment

Friday, March 19th, 2010 by hinrich

Affymetrix GeneChips

In a recent note I have been wondering about the half life of new 'omic technologies and their impact on data quality. I would like to revisit this thought in the context of Roy Amara's law. Roy Amara has been a scientist and co-founder of the Institute for the Future in Palo Alto. His "law" states:

We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.

This observation however assumes that we have sufficient time to estimate the power of a new technology, learn it's limits and it's applications. When I look at the acceleration of advances in new 'omics technologies in combination with the increase in the number of such technologies versus the number of scientists that study and use them, I am concerned that we may loose the critical mass necessary to go through the cycle that is described by Roy Amara's law.

I can clearly recognize the stages of the hype cycle when applying it to the microarray technology of Affymetrix - I think we are somewhere between the "Slope of Enlightenment" and the "Plateau of Productivity". Next to Affymetrix many other companies tried to establish a competing technology. Luckily for Affymetrix, the competition came later and Affymetrix had quite a head start.

If I now look at next-gen sequencing, I see a number of competing companies that all started somewhat at the same time and with different technology properties that have an impact on the most suitable application for a given technology. While we are somewhere between the "Peak of Inflated Expectations" and the "Trough of Disillusionment" there are a number of companies who have already announced newer versions of their next-gen sequencing machines as well as a number of companies that are eager to introduce the third generation of sequencing technology. I hope that the different platforms will stabilize soon so that we can enter the "Slope of Enlightenment" and do not get stuck in the "Trough of Disillusionment"...

Posted in Molecular Profiling

Road trains instead of truck tailing

Saturday, March 13th, 2010 by hinrich

Road train

Two years ago I had been wondering about how creative people might get in their attempts to reduce fuel costs. At that time fuel prices had dramatically gone up. So, I jokingly noted that hopefully people will not start truck tailing...

The other day I stumbled across a technology article on the web site of the BBC: 'Road trains' get ready to roll. I was surprised to learn that this thought is actually not so stupid. According to the article, EU researchers attempt to develop a system that would allow a small "platoon" of vehicles to travel together behind a lead vehicle. The professional driver of this lead vehicle would ensure the safety of the journey and following vehicles could join or leave the "train" as they need.

The project will run for three years and at the end the researchers intend to conduct tests in a number of European countries. Not on public roads though...

Posted in Green