針對癌症研究中的高通量資料建立整合性分析平台

Abstract

科學家們為了瞭解基因在癌細胞內的運作，經常使用高通量技術來進行癌症研究，但是高通量技術一次檢測的基因位點動輒上萬，甚至是全基因組，使得高通量技術的資料不但龐大又複雜，而且通常會以不同型態的資料相互比較，例如基因表現與基因甲基化程度，進而推測基因的調控機制，本研究偏重於分析生物晶片的資料，不過目前仍以分析illumina及Affymetrix生物晶片為主，我們根據基因表現、DNA甲基化及miRNA資料進行分析，還包含基因表現與DNA甲基化、miRNA表現兩兩之間的比較分析，結合R語言中Bioconductor工具及其他相關的公開資料庫或工具，經由資料前處理、正規化，最後得到差異表現的基因，並利用Gene Ontology與KEGG對這些基因做詮釋，觀察它們在細胞中的位置、參與的作用和調控路徑，試著找出基因與基因間的交互作用，藉由整合上述的流程歸納出一個分析平台，可以分析單一型態資料，亦可以併入不同型態資料做相關性比較，而且適用於各式生物晶片。

As the scientists trying to find out the effects of genes in the cancer biology, they usually use the high‐throughput technology to analyze the cancer cells. But the quantities of data from the high‐ throughput technologies are large and the formats are complicated. And even more, sometimes they need to compare the different types of high‐throughput data makes it much more difficult to analyze. Therefore, we build this integrated analysis platform and focus on microarray analyzing including gene expression, DNA methylation and miRNA expression data. Also, this platform includes the comparison of gene expression and DNA methylation data or gene expression and miRNA data. By using Bioconductor, the tool in R language, and other open source database and tool, we can process the preprocessing, normalizing and filtering the differential expression gene. Finally, Gene Ontology and KEGG can be used for annotation of genes in order to reveal the characters of genes in the tumor.