訊號傳遞在細胞生理現象中扮演重要的角色。細胞是藉由接收外部不同的訊號,再依照訊號所代表的意義,調節其化學分子做出切合需要的反應。 在假設訊號傳遞路徑中所有的組成分子已知的情況下,本研究整合基因微陣列及蛋白質交互作用這兩種不同型態的資料的方法,發展出一個計分函數,來預測路徑組成分子的順序關係。實驗結果顯示,與單獨使用基因微陣列或蛋白質交互作用的資料相較,整合這兩種不同型態的資料的方法,在驗證路徑分子組成的順序時,普遍有較好的效果,在所有可能排列路徑中,排名大部分都在前百分之七。 MAPK路徑中,採用的方法得出了一個不錯的預測,因此推測此的方法也可以用來預測別的路徑,包括那些我們不是非常瞭解的路徑。另外我們也應用此方法來預測3及4個子單元的蛋白質複合體結合順序,並建立了一個網頁介面供使用者做預測服務。網址:http://210.70.83.81/pop/
Signal transduction plays an important role in the control of most fundamental cellular processes by which cells convert an external signal into a response. In this study we present a computational approach that integrates different types of information to predict the order of the signal transduction pathway components assuming all the pathway components are known. Our method is built on a score function that integrates protein-protein interaction data and microarray gene expression data.Compared to the individual dataset, either protein interactions or gene transcript abundance measurements, the integrated approach leads to better identification of the order of the pathway components. Our method can lead to a good prediction for the well-known yeast MAPK signaling pathways. Therefore, we conjecture that this approach may be applicable to many other pathways including less well understood ones. In addition,we use our approach to predict the combination order of protein complexes with three or four subunits.And we set up a web interface (http://210.70.83.81/pop/) to show the results.