Vol.3 No.1 2010

8/110

Research paper : A bioinformatics strategy to produce a cyclically developing project structure (M. Suwa et al.)−5−Synthesiology - English edition Vol.3 No.1 (2010) 3.3 Output of the ProjectWhen all GPCRs were identified from the human genome, we obtained 827 level A, 1300 level B, 1517 level C, and 2109 level D sequences. While there were higher possibilities of false-positives (where wrong sequences may be predicted) in sets with higher numbers, they also had higher chances of including new GPCRs. Interestingly, it was found that the majority of the GPCRs were concentrated on chromosome 11, they were dominated by olfactory receptors, and the chemokine receptors were concentrated on chromosome 3. This finding was possible for the first time through this comprehensive gene identification. We applied for patent in 2002 for several hundred sequences that were determined to be new sequences. A certain pharmaceutical company requested disclosure and we received income from this disclosure. Hence, we were able to produce results of the Product Realization Research. The GPCR sequence with additional structural and functional information using the computational method was organized as a database and publicized in 2003 (SEVENS[7] http://sevens.cbrc.jp/1.20/, the very first version). At this point, the core technology was completed to a point, and the first cycle of the Project that started from scratch came to a milestone. 4 Project that undergoes cyclic development 4.1 Hop: Core technology development of the whole ProjectThe Project that started in 2000 completed one cycle consisting of analysis of elemental technology, systemization, and product realization, and still continues after publication on the website. If the “first cycle of Full Research” as described in the previous section was the “hop” of the triple jump, the leap continued in “step” and “jump.” Following is the description of the development into joint research, and further technological development through joint research. 4.2 Step: Feedback to core technology through collaboration with industry and academiaIn 2002, we experimentally confirmed the expression of several sequences in human tissue for the new GPCR in SEVENS jointly with companies, and applied for patent for particularly important sequences. The fact that the expressions were confirmed for genes predicted by the computational method demonstrated the adequacy of our policy. However, we also had issues. As a method for confirming the gene expression, we used the polymerase chain reaction (PCR) where minute nucleic acid sequence samples could be rapidly multiplied in a short time. However, it is desirable that the sequence used in PCR analysis have full length with the correct terminals on both ends. However, we found that there were many cases where the terminals were lacking as a result of failure of start (or stop) exon identification in the predicted genes. Most of these were long genes composed of several exons, and since they extended into a wide region, the parameter set in section 3.1 (1,000 bases) was not sufficient as the parameter for the extension of the gene region. Therefore, we investigated the gene existence region to cover a wider area than the commonly assumed area, and it was surprisingly found that it was necessary to extend an arbitrary exon upstream and downstream to 140,000 bases. Although the subject of SEVENS pipeline was GPCR, it is applicable to other types of protein if the parameters at each phase are changed. We attempted this in the joint research with a venture research center of the University of Tokyo in 2002. In chronic inflammatory diseases, such as rheumatoid arthritis and multiple sclerosis, the immunocytes aggregate excessively at the site of the chronic inflammation and destroy the tissue. This is because the migration of the immunocytes is triggered when the protein called chemokine binds with its GPCR (CCR2). There was a competition for the search of a molecule that inhibits the binding of chemokine (antagonist). However, to avoid the side effects that were expected to occur when the antagonist intercepted the chemokine receptors of different subtypes with structures similar to CCR2 that may be active during organ formation, cell multiplication and differentiation, there was demand to look for a molecule that controlled CCR2 through a different route from the antagonist. The experimental research showed that a new gene that gathered specifically at the C-terminal of the CCR2 (FROUNT) could be the candidate. We found that this was a long protein composed of 600 residues, where multiple helixes appeared repeatedly. Also, as a result of searching the genome for the characteristic of having several short and weak motifs, we found that there were only two regions that completely matched the new gene, but there were several that showed matches at a weak score. This study was published in Nature Immunology[8]. The technologies that were re-investigated in the two joint researches were reflected in the SEVENS pipeline. 4.3 Jump: The development of the new function prediction programJoint research with a pharmaceutical company started in 2004. Here, a computation system to efficiently and comprehensively screen the ligands that regulate the activation of G protein selectively was build, and we applied this to the ligand screening of the orphan receptors whose binding ligands were unknown. First, we selected 108 novel human GPCRs from the level A data set of SEVENS. These were orphan receptors. Next, the ligands to be used in screening were comprehensively identified from the human genome based on known peptide ligands after optimizing the gene identification pipeline for

※このページを正しく表示するにはFlashPlayer9以上が必要です