利用赫伯特-黃轉換法做為在質譜儀分析技術的前處理方法; A novel preprocessing method using Hilbert Huang transform for MALDI-TOF and SELDI-TOF mass spectrometry data

NCU Institutional Repository > 生醫理工學院 > 系統生物與生物資訊研究所 > 博碩士論文 > Item 987654321/6573

請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/6573

題名:	利用赫伯特-黃轉換法做為在質譜儀分析技術的前處理方法;A novel preprocessing method using Hilbert Huang transform for MALDI-TOF and SELDI-TOF mass spectrometry data
作者:	陳欣皓;Hsin-Hao Chen
貢獻者:	系統生物與生物資訊研究所
關鍵詞:	質譜儀;赫伯特-黃;mass spectrum;HHT
日期:	2009-07-10
上傳時間:	2009-09-22 10:22:45 (UTC+8)
出版者:	國立中央大學圖書館
摘要:	自人類基因體計畫開始以來，基因序列分析已蓬勃發展。蛋白質體學的研究在近代引起了生物學家的注目，蛋白質體學本身扮演著基因體學與細胞表觀行為溝通的橋樑。在此，我們致力於高通量、更有效率與提高準確的質譜儀分析。表面強化雷射解析電離飛行質譜(SELDI-TOF)與基質輔助雷射脫附游離法飛行質譜(MALDI-TOF)技術是質譜儀技術裡面兩種常用的技術。利用質譜圖所偵測到的波峰，我們可以用來當作正常人與病患間的生物標定物來區分。然而，一個質譜圖裡夾雜許多複雜的訊號，特別是雜訊。因此，對於質譜圖資料的前處理分析就顯得更為重要。我們以赫伯特-黃轉換法(Hilbert-Huang Transformation)為主要概念，發展一個新的質譜圖前處理方法。並進而與目前較為熱門的幾個錢處理方法做比較，包含PROcess、SpecAlign以及MassSpecWavelet。我們著重於在質譜圖中較為顯著且重要的波峰，並觀察這些波峰在上述幾個方法的表現；同時，為了更顯客觀性，我們另外做了一次實驗，分別為未加入樣本所得到的質譜圖、與加入樣本得到的質譜圖，由此來驗證我們確實能有效去除雜訊，特別是去除部分化學物質所產生的雜訊。 Motivation: There are a lot of gene sequence analyses, especially the time after human genome project. The proteomics becomes more and more attractive for biologists. It can bridge the gap between the genome sequence and the cellular behavior. We are concerned about the Mass spectrometry which is high throughput, fast, and accurate. Matrix assisted laser desorption ionization (MALDI) and surface-enhanced laser desorption ionization (SELDI) time of flight (TOF) are two popular technologies in the field of spectrometry. With the peaks detected in spectra, we can compare the normal group with disease. However, the spectrum is complicated and full of noise. Consequently, the preprocessing of the mass data plays an important role during our analysis. Results: We provide a novel algorithm of preprocessing dealing with the MALDI and SELDI spectrum. The algorithm uses the Hilbert-Huang Transform mainly. We compare the performance of several famous algorithms including PROcess, SpecAlign, and MassSpecWavelet with ours called HHT. The main thought of performance is chiefly visual comparison. We pick the significant peaks and observe the results which the algorithm shows in figure. The results show that HHT for preprocessing is more fitness than others. Not only detecting the peaks, but HHT has the advantage of denoising the spectra, especially for the complex data.
顯示於類別:	[系統生物與生物資訊研究所] 博碩士論文

文件中的檔案:

檔案	大小	格式	瀏覽次數

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....