Because one phenomenon of life is caused by a great variety of genes and molecules, understanding of one phenomenon of life needs exhaustive collection of gene expressions. The collection, called gene expression profile, is recently getting possibility by microarray, which is one of a method to investigate whether a gene is expressed or not. Data Mining, the field of computer science for discovery of useful summary of data, is in the spotlight as analysis method of the collected gene expressions. In this paper, we will introduce the techniques of Data Mining, and explain how you can apply the techniques to analysis of gene expression profiles.
Nitric oxide(NO)is an important molecule with many diverse biological functions as a signaling and cytotoxic molecule in the cardiovascular, nervous, and immuno systems. NO is synthesized from L-Arg by a family of enzymes termed nitric-oxide synthase(NOS). NOS is a fusion protein composed of an oxygenase domain with a cytochrome P450-like heme active site and a reductase domain similar to NADPH-cytochrome P450 reductase. Electron transfer from NADPH via the reductase domain to the heme active site is prerequisite for the activation of molecular oxygen with this enzyme. We summarized current concept of the electron transfer system of NOS based mainly on the eNOS crystal structure and our recent findings.
Recent structure determination of bovine rhodopsin(Rh), a prototypical G protein-coupled receptor(GPCR), by X-ray crystallography has become possible with some noteworthy findings on the selective separation of protein-lipid complex from the natural membrane source and on the phase behavior in crystallization. The structure model of Rh contains much wealth of information explaining how GPCRs can be constrained in their inactive state, and how a number of conserved residues play their roles in the structural frame of GPCRs. A possible mechanism is also presented for the photoactivation of Rh.
Totally in vitro protein evolution systems are powerful tools to explore the protein sequence space widely, uniformly, and rapidly. The systems require "genotype-phenotype linkage" that has been accomplished so far in three methods; ribosome display, in vitro virus display, and STABLE display systems; by physically linking protein and its gene(mRNA or DNA). Useful applications are;(i)evolutionary design of novel and improved proteins,(ii)experimental simulation of protein evolution, and(iii)high-throughput screening of protein-protein interactions for functional proteomics.