基因體醫學及生技研發之生物資訊核心設施---結構生物資訊(III)(子計畫四)

Abstract

我們的工作分為三個方面: 定期服務、技術開發和合作研究服務。工作比例分別為：定期服務(50%), 技術開發(30%) 並且合作研究服務(20%) 。我們的核心主要提供基於網際網路的服務，一些例子如: GemDOCK（根據進化演算法所發展出的藥物相接伺服器)，蛋白質結構預測伺服器) CELLO （預測subcellular localization伺服器), SSDB （雙硫鍵蛋白質資料庫）, ChamSeq Finder（探測chameleon序列伺服器），SDSE（計算蛋白質結構熵伺服器），KinasePhos(預測蛋白質phosphorylation sites和催化kinases的工具; ProSplicer伺服器; miRNAmap伺服器; dbPTM伺服器。另外, 我們提供PDB 、SCOP 和PredictProtein 本地鏡站服務。 因為我們網際網路的服務是針對所有用戶（不論是國內和或國外）開放的,我們從用戶獲得了許多迴響。所以, 除了我們固定的網站維護外，我們投擲了許多的努力回應這些用戶的請求，在我們的網路服務做了許多的原創性的改進。但是我們並沒有不請求用戶和我們之間需要有任何正式合作協議。這裡可用個例子說明－我們原始的CELLO只預測原核生物的蛋白質subcellular localization(Protein Science, 2004, 13, 1402)。由於用戶的請求，我們加入了CELLO對真核生物的預測。原來CELLO只預測蛋白質序列，後來我們又加入它預測DNA序列的功能。另一方面, 我們原來要做的Biological spelling checker，後來由於效過不彰，便不再做了。 由於我們不斷擴展的萬維網服務和持續增長的用戶, 我們的伺服器的性能（如計算速度和磁盤存儲空間）漸漸落後於我們服務的區塊。另一方面, 訓練有素的bioinformaticists在臺灣仍然短缺, 我們的核心服務大部分還是倚賴我們生資所的研究生的維護。另外網路安全問題仍是個重要的問題。今年又駭客進入, 但是我們迅速修理了漏洞，並且沒有損害造成。 我們不斷更新我們萬維網服務。例如, 我們的最近新建添加: miRNAMap與dbPTM。其中相關技述及演算法將在Nucleic Acid Research發表。另外, 我們將加強GemDOCK的服務，加pharmacophore 的功能(Proteins (2005), 59, 205)。 這個核心設備的目標, (像其它由NRPGM 支持核心服務設施),主要是服務與NRPGM相關的研究員。但是, 因為我們的基本上是網際網路服務，因此我們的用戶分佈於海內外。最近, Sanofi Pasteur(一家9000 位僱員生技公司) 詢問是否能在當地裝置CELLO鏡像站。這表明了我們的bioinformatics 核心為實驗生物學家的確提供了有用的服務。

Our efforts will be dedicated to three categories: the routine service, technology development and the collaborative research service. Our percentage efforts in these three categories are: the routine service (50%), technology development (30%) and the collaborative research service (20%). Our core provides mainly web-based services. Some of the notable web services that are developed in-house are: GemDOCK, the Protein Structure Prediction Server (PS2) server, CELLO, the SSDB, the ChamSeq Finder, SDSE, KinasePhos, ProSplicer, miRNAmap, dbPTM. In addition, we also provide local mirror service of PDB, SCOP and PredictProtein. Since our web-based services are essentially open to all users, domestic and international, we obtained numerous inputs and requests about our web service. Therefore, in addition to our routine web sites and system maintenances, we have thrown significant efforts into adding the requested, customized features to our web services. Due to our expanding web services and the ever increasing users, the computational capability of our servers appear to lag behind in terms of computational speed and disk storage space. On the other hand, the well-trained bioinformaticists are still in short supply in Taiwan, and our core service still in the main depend on the efforts of our graduate students of the bioinformatics program in maintaining and updating our web services. Most of our web services are developed by us and have gone through the peer-reviewed processes before being posted on the web sites. This will ensure the credibility and originality of our web services. The aim of this core facility, like other core services supported by NRPGM, is mainly for the service of researchers participating in NRPGM. However, since our core service is web-based in nature – quite distinct from other "wet-lab" type core service, our service is essentially open to both domestic and abroad. Our performance can be evidenced by the number of web accesses. Recently, we have got a request from Sanofi Pasteur - a 9000 employees biotechnology company - requesting a on-site web implementation of one of our web services. This indicates that our bioinformatics core has indeed provided useful service for experimental biologists.