- 期刊
新穎質譜特徵選擇程序以利血液生物標記開發
Feature Selection for Assisting Biomarker Discovery in Serum with Mass Spectrometry-Based Proteomics
摘要
在後基因體世代，各類物種的基因體解碼已逐漸完成。分子醫學的研究，也逐漸從基因體學的研究轉向蛋白質體學方向。尤其在偵測與疾病相關聯的生物標記方面，因為技術與儀器的大幅改進，運用蛋白質體來辨識疾病並尋找相關連的生物標記，已是目前方興未艾的重要領域之一。在這篇論文中，我們嘗試設計了一套質譜特徵選擇程序用來幫助生物標記的發現，跟以往質譜特徵選擇程序不同的地方在於我們可以從三組質譜資料集合中選出具有高精確度(specificity)與高敏感度(sensitivity)的通用特徵(universal features)。此外，我們也試著取出不同範圍的質荷比，將原本的質譜圖切割成更小的單元來進行特徵選擇與分析。實驗結果顯示了這種分析方法有助於找到更細緻的生物標記，在某些情況下，亦可提高我們的分類效果。
並列摘要
In the post-genome era, molecular medicine is moving beyond genomics to proteomics, especially in detecting and discovering biomarkers for diagnosis. In this paper, we tried to develop new procedures for assisting biomarker discovery. Differentiating our research from previous studies, we are able to select the universal features that perform high sensitivity and specificity in predicting for all of the three datasets. Moreover, we further identify subsets of features in a series of m/z windows. Each subset of features selected performs well for predicting the dataset prepared from the same procedures.
延伸閱讀
- 沈欣毅（2013）。合成並評估新穎血清素轉運體配體做為正子斷層掃描示蹤劑〔碩士論文，國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU.2013.10310
- Lin, T. W. (2018). 應用於個人化生物分子檢測之血液分離技術研發 [master's thesis, National Taiwan University]. Airiti Library. https://doi.org/10.6342/NTU201804357
- 郭訓志、黃仁澤、薛慧敏（2011）。高維度資料特徵選取之探討－應用於分類蛋白質質譜儀資料。Journal of Data Analysis，6(3)，72-83。https://doi.org/10.6338/JDA.201106_6(3).0004
- Wang, K. Y. (2016). Analysis of targeted post-translational modification in serum protein biomarker using mass spectrometry-based strategy [doctoral dissertation, National Taiwan University]. Airiti Library. https://doi.org/10.6342/NTU201602967
- Mackay, E. M., Koppel, J., Das, P., Woo, J., Schriemer, D. C., & Bathe, O. F. (2015). A Hypothesis-Directed Approach to the Targeted Development of a Multiplexed Proteomic Biomarker Assay for Cancer. Cancer Informatics, (2015), 65-70. https://doi.org/10.4137/CIN.S24388