论文部分内容阅读
目的:解决肽段色谱保留时间预测模型在肽段鉴定结果验证中应用的实际问题。方法:本文基于Krokh in等提出的改进模型,利用3次样条平滑的方法估计肽段的实验色谱保留时间;然后利用迭代算法同时完成参数估计和异常值排除;最后利用迭代得到的模型和残差范围进行肽段鉴定结果的过滤。结果:对两批不同物种的实验数据,本文方法与直接进行线性回归的方法相比,可以在保证几乎不损失阳性肽段鉴定结果的情况下排除更多的假阳性结果。结论:本文对肽段色谱保留时间模型在肽段鉴定结果中应用的实际问题提出了系统的解决方法。
OBJECTIVE: To solve the practical problems in peptide peptide chromatographic retention time prediction model in the validation of peptide identification results. Methods: Based on the improved model proposed by Krokhin et al., The cubic spline smoothing method was used to estimate the experimental chromatographic retention time of peptides. Then, the iterative algorithm was used to estimate the parameters and eliminate the abnormal values simultaneously. Finally, the model and residual Poor range of peptide identification results of the filter. Results: The experimental data of two batches of different species can eliminate more false positive results without losing the identification results of positive peptides, compared with the method of linear regression. Conclusion: This paper presents a systematic solution to the practical problems of peptide retention time model in peptide identification results.