« Tuesday's | Main | Why Stop Israel from Blockading Lebanon? »
August 30, 2006
NMPDR - The National Microbial Pathogen Data Resource
I've decided I should write about the web site Bruce is working on for the government. When I say "government", I mean either the Department of Energy, the National Science Foundation, or the Department of Homeland Security. Bruce doesn't understand how any of the financing works; as far as he's concerned, he's trying to develop a database system where the client is God. God is not a good client. When Bruce was working for Harris Bank, he could go to one of the managers and ask how to generate credit card numbers for new customers. Bruce has repeatedly asked God to explain how to properly line up DNA strands from different types of bacteria and he's still waiting for a response.
The web site is called the National Microbial Pathogen Data Resource, and it is the front end for a database describing the DNA of hundreds of life forms. To give you an idea of the complexity involved here: the database contains over 42 billion characters of data, of which 24 billion are DNA sequences.
The P in NMPDR stands for pathogen, which means special attention is paid to the life forms that cause human disease. Though the work performed in organizing the data is so complex that no one person can understand all the pieces, the motivation behind the work is blindingly simple.
Cells use DNA to create protein molecules. Most normal inorganic molecules are small things that are represented in chemistry classes by little balls as shown on the left. Protein molecules, on the other hand, are huge things that look like an explosion in a confetti factory.
The complicated surface of the protein causes other molecules to attach themselves to it, creating a three-dimensional molecular jigsaw puzzle. The molecules thus brought together participate in the chemical reactions that sustain life.
Drugs are alien chemicals that mess up protein surfaces so they can't do what they're supposed to do. For example, Clostridium botulinum generates proteins that destroy the chemicals used to transmit instructions to your muscles. A chemical that inhibits those proteins would be a useful drug for treating botulism.
Proteins are created by areas on DNA called genes. The NMPDR organizes genes into groups called subsystems. The theory is that once you identify the genes in a subsystem for one species, you can find similar genes in other species that do the same thing. In the case of Clostridium botulinum, the subsystem that produces the dangerous proteins is called Streptolysin S Biosynthesis and Transport. One of the genes that participates is number 69. The NMPDR page for that gene contains information about neighboring genes, diagrams of the relevant DNA strand, and the components of the protein the gene creates. A drug targeting expert could take that information, create a computer model of the protein, and then apply the model to different known drugs to see if one of them will make the protein inoperable.
The fundamental instrument of Western civilization's continued good health is antibiotics: drugs that kill almost every known bacteria. Bacteria are learning to fight back, so we need new tools. This is not a prediction, it's a fact: this NMPDR page lists the genes used by some organisms to fight the antibiotic tetracycline.
The discovery of antibiotics was mostly an accident. More recently, the production of monoclonal antibodies has enabled drug manufacturers to produce chemicals that mimic the immune response of higher animals, but trial and error is still a big part of the process.
Imagine, then, a computer program that not only tells you which bacterial proteins a drug will target, but also which human proteins are affected. In one fell swoop you know what the drug cures and what kind of side effects to expect.
That sort of thing is still science fiction, but actual people are working on making it a reality.
Respectfully submitted,
Ferdinand T. Cat
# At Wed 12:43 PM | Permalink | Trackback URI | Comments (0) | More NMPDR | Tags: antibiotics bioinformatics DNA evolution medicine NMPDR pathogens proteins RNA
Leave a comment
Leave a comment