Improving biomedical science and reducing use of laboratory animals

[cor]

Hello scientists,

I am posting here, hoping for ideas and feedback on how we can use the computer to
1) improve biomedical research, especially in cancer, and at the same time
2) reduce the use of laboratory animals.

My professional background is that I recently started as a bioinformatic engineer in a cancer research institute.

Biomedical science seems to be seriously hampered by the use of animal models, as Khalil writes in a recent article in Current view on Oncology:
"Many of the cell lines and animal models used to study cancer mechanisms and test possible therapeutics are not predictive of what will happen in actual in vivo human tumors. The discrepancy between in vitro and the in vivo human has significantly affected the ability of researchers to devise successful cancer therapeutics. “(1)

An example of the problem with the animal model is shown in the case of DNA (plasmid) vaccination. Although successul in animal models in humans it didnot work that well:
* However It soon became evident that DNA vaccines while robust in small animal models were less immungenic in nonhuman primates and humans. (6). and
* Unfortunately, this (DNA plasmid, ed) technology has not proven to be reliable in humans. (2, p. s7)

Replacing animal models for computermodels based on human (f.e. microarray) data, would improve science and at the same time reduce the use of laboratory animals. Besides the effort in building such models, it also seem to require a paradigma shift in biomedical science. Traditional, biomedical (molecular) science is using a reductionistic approach, Reseach is focused on the effects of knocking down or knocking out of a single gene. A more holistic, system approach is needed. The approach advocated by Khalil is to infer (“reverse engineering”) the gene regulatory mechanisms looking at microarray data, using mathematics and statistics. The “inference engine” should however also take in account what is already known in the form of existing computermodels.

What do you think are the prospects and/or requirements for the use of computer models in stead of animal models in biomedical science?

Thanks for your reply in advance!

Cor Lieftink
(1) Khalil, I G; Hill, C (2005) Systems biology for cancer. Current Opinion in Oncology. 17(1):44-48,

(2) Plotkin , S.A, (2005) Vaccines: past, present and future. Nature Medicine 11, S5 - S11
http://www.nature.com/nm/journal/v11...ll/nm1209.html

(3) Schleef, M. (2005) DNA-Pharmaceuticals: Formulation and Delivery in Gene Therapy, DNA Vaccination and Immunotherapy. Chapter 1
http://media.wiley.com/product_data/...27311874-1.pdf

[admin]

Dear Cor Lieftink,

I welcome you to our forum and I do agree in many respects. As a researcher and an admin at Molecular Station, I do believe that many experiments especially with respect to animals are done un-necessarily. I in fact did a studentship (Ph.D) in a lab that worked with animals. Although great care was taken to minimize animal use (due to the reasons you mention and also due to cost reasons), due to the nature of research, sometimes animals were un-necessarily sacrificed.
Also, a great deal of research in animals does not apply to humans due to the great differences in genetic code much of which has not been ascribed functional significance yet.

We have posted animal techniques although we have also posted this link on Non Animal Research Methods which you may have seen before and the Animal Use Alternatives database in our protocols database.

We here at Molecular Station see a great future for bioinformatics and computational biology, which will be able to predict experimental outcomes to a high degree and also eliminate experiments which are redundant (determine which questions have already been answered - this is a problem as many experiments are repeated by separate labs needlessly).

We envision a near future that will have a small role for animal studies, but a growing role of as you say global analysis of relavent samples (from humans) using proteomics and genomics capable of analyzing even single cells, which will allow the analysis of samples from living humans without invasive extraction methods. Although this does require a shift in thinking as you mention, it will take a new generation of researchers to make this a reality, as the older generations cannot be changed in their ways.

Sincerely,
Admin Molecular Station

[hezma]

Yes, I also see a movement to more computer use and bioinformatics. There is much data out there that remains to be analyzed by computers.

Also, as mentioned why waste thousands of animals by separate labs to study 1 protein?
This can be done by a few labs with proteome chips which analyze EVERY protein and isoform in the future. These chips only need small amounts of sample / tissue. Thus, one can have a STANDARDIZED facility which caters the animal tissue (from a few animals treated the same way) and sends them out to the labs for analysis.

This would seriously cut down on the amount of materials, animals, time, and money yet maximize the amount of data / information coming out.

Genomics, Proteomics and Bioinformatics is the wave of the future of science.