發明
中華民國
101149024
I 472944
實施於計算機的差異表現基因辨認方法以及此方法之電腦可讀取記錄媒體
國立中央大學
2015/02/11
以高通量技術辨認樣本群間的差異表現基因被廣泛運用但效率低落。我們研究了175 數據組,發現倍數改變與方差間一般統計檢定無法適當處理的強烈依賴是基本原因,並提出解決方法。此命名為WABE的分析法提供高再現性的基因清單,高統計力及完全的統計控制。在一以接觸過敏皮膚炎病人與正常人數據的論證中,我們證明在五個受試方法中只有WABE能偵測到可靠的生物資訊。此資訊不但與原來研究結果完全衝突,也透露正常人的基因體對鎳接觸有比過敏病人更劇烈的反應。此發現證明鎳過敏的病因是人體固有防禦機制的官能障礙,對預防藥物與治療方式的發展提供了新的觀點。 The popular application of high-throughput expression profiling to detect genes differentially expressed between sample groups has been inefficient. Here, based on 175 data sets studied, we identified as the fundamental cause strong dependence between fold-change and variance no common statistical tests can adequately handle and presented a method as resolution. The method, named WABE, provides reproducible gene lists with high statistical power and complete statistical control. In a demonstration using data from allergic contact dermatitis patients and non-allergic controls, we showed among five methods only WABE extracted reliable biological information. The information is altogether contradictory to the original study and reveals stronger genomic responses to nickel exposure in the controls than in the patients. The finding identifies as the disease cause dysfunction of an innate system and creates a new perspective for alternative prophylactic and therapeutic strategies.
智權技轉組
03-4227151轉27076
版權所有 © 國家科學及技術委員會 National Science and Technology Council All Rights Reserved.
建議使用IE 11或以上版本瀏覽器,最佳瀏覽解析度為1024x768以上|政府網站資料開放宣告
主辦單位:國家科學及技術委員會 執行單位:台灣經濟研究院 網站維護:台灣經濟研究院