• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 2
  • 1
  • 1
  • Tagged with
  • 2
  • 2
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

蛋白質質譜模擬之研究 / A Simulation Study of Proteomic Mass Spectra

林芳華 Unknown Date (has links)
進入後基因時代,蛋白質體學成為很多科學家有興趣的主題。蛋白質鑑定成為重要的一環,而質譜儀在縮氨酸分析及蛋白質鑑定中扮演重要的角色。腫瘤、卵巢癌及攝護腺癌等研究亦已成為質譜儀上的應用。Coombes 等人 (2004) 提出了一個線性的數學質譜儀模型,而且建議這個模型可被應用在建立質譜儀的模擬中。本文中,我們利用虛擬的質譜儀產生虛擬質譜資料並加以研究。虛擬的質譜實驗包括了前、後兩部份。樣本資料要放入虛擬質譜儀之前,可能出現的蛋白質其強度(intensity)必須先隨機地被決定,之後強度必須被轉換(calibration)變成離子化的個數(abundance) ;之後將樣本資料丟入虛擬質譜儀中,每一個蛋白值的飛行時間(time of flight, TOF) 將會被紀錄。另外一個轉換(calibration)是將飛行時間(TOF)轉成質量電荷比(mass-to-charge ratio,m/z)。質譜儀和兩個轉換都會在資料中產生誤差。在本文中,一個完整的模擬過程將會一步一步被介紹。同時,兩個轉換的方法所產生的誤差也會被探討。之後,我們將此模擬方法應用於模擬一組攝護腺癌中。 / Entering the post genomic era, proteomic has become the topic that scientists are interested in. The authentication of protein has been an important item of the topics. Mass spectrometry (MS) has become an important tool for peptide analysis or proteomic authentication. There are many applications of MS such as oncology, ovarian cancer, and prostate cancer. Coombes et al.(2004) proposed a mathematical model of a virtual spectrometer and suggested that the virtual spectrometer can be applied in conducting a MS simulation. In our study, we focus on designing a simulation study of spectrum data from a virtual MS experiment. The virtual experiment includes two stages: pre- and post-virtual spectrometer. Before the sample data are put into the virtual spectrometer, a virtual population of the intensity of all possible proteins should be determined; a virtual sample is randomly drawn; and the generated sample of intensity should be calibrated to abundance, which is the number of molecules ionized and desorbed from the biological sample. The sample data are then put into the virtual spectrometer and the time of flight (TOF) of each ionized molecule is recorded. Another calibration is employed to transfer a TOF to a mass-to-charge ratio (m/z). The spectrometer and the calibration processes produce variation in MS data. In this study, a complete simulation design of mass spectra will be introduced step by step. Moreover, the calibration effects caused from the two calibration procedures will be investigated. A simulation based on a real data set from a prostate cancer study will be also given as an illustration.
2

線性維度縮減應用質譜儀資料之研究

陳柏宇 Unknown Date (has links)
近年來電腦科技進步、資料庫健全發展,使得處理大量資料的需求增加,因而發展出結合生物醫學與資訊統計兩大領域的生物資訊(Bio-informative)。這個新學門的特色在於資料量及資料變數的龐雜,但過多資料經常干擾資訊的篩選,甚至癱瘓資料分析,因此如何適當地縮減資料(Data Reduction)就變得必要。資料縮減常藉由維度縮減(Dimension Reduction)進行,其中常見的線性維度縮減方法首推主成份分析,屬於非監督式學習(Unsupervised Learning)的一種,而線性的監督式學習(Supervised Learning)方法則有SIR(Sliced Inverse Regression)、SAVE(Sliced Average Variance Estimate)及pHd(Principal Hessian Directions)。非監督式學習的主成份分析,主要在找出少數幾個維度而可以解釋代表自變數的變異程度,而監督式學習的SIR、SAVE及pHd則可以在縮減維度時,同時考量自變數跟應變數之間的關係,而找出可以解釋應變數的維度。 本研究為解決蛋白質質譜儀資料高維度的問題,將應用各種線性維度縮減方法,並分別使用CART(Classification and Regression Tree)、KNN(K-Nearest Neighbor)、SVM(Support Vector Machine)、ANN(Artificial Neural Network)四種分類器,比較各維度縮減方法的分錯率高低,以交叉驗證(Cross Validation)比較維度縮減方法的優劣。研究發現在四種維度縮減方法中,PCA及SIR在各種分類器下都有較為穩定的分錯率,表現較為一致,但SAVE及pHd較不理想。我們也發現在不同的分類器下,PCA跟SIR兩者有不同表現,正確率較高的分類器(SVM與ANN)與PCA結合,而正確率較低的分類器(CART與KNN)與SIR結合,會有較佳的結果。另外,我們也嘗試整合分析(Meta Analysis),綜合幾種線性維度縮減方法,而提出邊際訓練效果法(Marginal Training Effect Method)與加權整合法(Meta Weighted Method),其中發現邊際訓練效果法若可以挑選出有效的維度,可以在不同分類器下提高整體模型,而加權整合法則確保在不同分類器下,讓其分類模型具有較為穩定的準確率;並提出相關係數重疊法(Overlap Correlation Method)來解決需要決定維度大小的問題。

Page generated in 0.0126 seconds