基于极限学习机(ELM)岭回归的DNA微阵列数据填补  

DNA Microarray Data Filling Based on Extreme Learning Machine Ridge Regression

作　　者：张建军[1] 张天成[1] 隋宇婷 岳德君[1] 

机构地区：[1]东北大学信息科学与工程学院,沈阳110004

出　　处：《小型微型计算机系统》2014年第10期2357-2362,共6页Journal of Chinese Computer Systems

基　　金：国家"九七三"重点基础研究发展计划项目(2012CB316201)资助;国家自然科学基金项目(61272180;61202086;61272179)资助

摘　　要：近年来,生物学领域中DNA微阵列技术发展迅速,利用微阵列数据可以有效地提取生物学信息,其应用十分广泛.但是微阵列试验中存在的许多客观因素,使得基因表达数据常常出现缺失.针对传统的缺失数据填补方法没有考虑微阵列数据中存在相关性的问题,提出一种基于ELM岭回归(B_ELM)的方法来对DNA微阵列数据进行填补.算法在考虑微阵列数据相关性的同时,解决了ELM前馈神经网络中存在的复共线性问题.为了进一步提高数据填补的效果,采用广义交叉验证算法确定最佳岭参数.最后,采用真实数据集,以均方误差作为评估标准对算法进行了验证,并将其与相关算法进行比较.结果表明,该算法在DNA微阵列缺失数据填补方面具有更高的准确率.In recent years, with the rapid development of DNA microarray technology in biology, microarray data has been widely used to extract biological information with high efficiency. But there are many missing values in gene expression data, due to some objective factors in the microarray experiments. The traditional methods of filling the missing data did not consider the correlations among mi- croarray data. We propose an ELM ridge regression method to fill the missing data in DNA microarray. We not only consider the cor- relations among microarray data, but also resolve the multicollinearity problem in the ELM feedforward neural networks. Moreover, in order to improve the effect of the data filling,the generalized cross-validation algorithm is used to determine the optimal ridge parame- ter. Finally, we conduct a series of experiments with the real data sets. By using the mean squared error as the evaluation criteria, we validate the algorithm and compare the results with the related algorithms. The results show that the proposed algorithm for filling the missing data in DNA microarray has a higher recovery rate.

关 键 词：DNA微阵列 缺失数据 ELM岭回归算法(B_ELM) 广义交叉验证 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]