中文摘要：

SEQUEST与Mascot为目前蛋白组学分析研究中使用最为广泛的蛋白质库搜索工具。尝试将Mascot与SEQUEST搜索结果进行比较,进而采用不同多变量判别方法对二者的搜索结果进行判别分析,以降低其结果的假阳性率。通过对Mascot与SEQUEST搜索结果进行比较,发现所得结果差异很大;利用多变量判别分析方法对Mascot及SEQUEST搜索结果进行判别分析,可有效提高SEQUEST结果中假阳性结果与正确结果之间的区分能力。对于Mascot搜索结果,采用多变量判别分析方法仍无法显著降低其假阳性结果,利用Decoy库搜索结果进行估计时亦存在导致错误估计的风险。

英文摘要：

Mascot and SEQUEST are two of the most popular protein database search tools for proteomics research currently.In this study,we try to compare Mascot search results to SEQUEST search results,and then use different multivariate discriminant algorithms to analyze both search results to reduce false positives present in those.After the comparison of Mascot and SEQUEST search results,it can be found that there is a big difference between the results obtained by these two tools.For the search results of SEQUEST,multivariate discriminant algorithms can effectively reduce the false positive identifications.However,the discriminant algorithms could not reduce the false positive identifications for Mascot search results.Also,there are the estimate errors for estimating the decoy database search results by Mascot search.