【摘要】： 我們利用不相關(guān)的人群來研究遺傳關(guān)聯(lián)的問題可能會引起假陽性的結(jié)果,但現(xiàn)在已建立起一些方法,如SPT方法來克服假陽性結(jié)果的出現(xiàn).但在此方法求遺傳背景變量的過程中,往往出現(xiàn)數(shù)據(jù)缺失的現(xiàn)象,并在一定程度上影響遺傳關(guān)聯(lián)結(jié)果的準(zhǔn)確性.在不增加實驗次數(shù)的情況下,缺失值估計是降低缺失數(shù)據(jù)對后續(xù)分析影響的有效方法.如果使用填充值代替缺失值進行數(shù)據(jù)分析,則填充值的準(zhǔn)確性會直接影響分析結(jié)果.因此為了保證數(shù)據(jù)分析和處理的正確性和有效性,確保提供有效數(shù)據(jù),對缺失值進行正確處理是一個非常重要的預(yù)處理過程.針對SPT方法中樣本表達數(shù)據(jù)取值的不連續(xù)的特點,采取了通過度量樣本之間的相關(guān)性來選擇最鄰近樣本,然后利用EM算法來估計缺失值.
[Abstract]:We use unrelated populations to study genetic association problems that may lead to false positive results, but now we have established some methods, such as SPT method, to overcome false positive results. However, in the process of finding genetic background variables by this method, the absence of data often occurs, and the accuracy of genetic association results is affected to some extent. Without increasing the number of experiments, the estimation of missing values is an effective method to reduce the impact of missing data on subsequent analysis. If the missing value is replaced by the fill value, the accuracy of the fill value will directly affect the analysis result. Therefore, in order to ensure the correctness and validity of data analysis and processing and provide effective data, the correct processing of missing values is a very important preprocessing process. Aiming at the discontinuity of sample representation data in SPT method, the nearest neighbor sample is selected by measuring the correlation between samples, and then the missing value is estimated by EM algorithm.
【學(xué)位授予單位】：黑龍江大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2008
【分類號】：R394;O213

【參考文獻】

相關(guān)期刊論文 前2條

1 邱浪波;王廣云;王正志;;基因表達缺失值的加權(quán)回歸估計算法[J];國防科技大學(xué)學(xué)報;2007年01期

2 楊濤,駱嘉偉,王艷,吳君浩;基于馬氏距離的缺失值填充算法[J];計算機應(yīng)用;2005年12期

，

本文編號：2283664