提出并实现一种基于局部回归的对齐算法,选择鉴定结果最多的一次实验作为参考,实现色谱保留时问的可逆对齐,将n次实验的对齐建模次数从n(n-1)/2次减少为”n-1次。该算法还实现不同实验间色谱保留时间的非线性对齐,并且能够克服少量从色谱柱中多次流出肽段的影响。研究结果表明:该方法计算时间复杂度远比现有算法的低,对齐精度也能够满足后续计算的需要。
A nonlinear and reversible algorithm based on local regression was proposed and implemented, which could reduce the alignment model form n(n-1)/2 to n-1 for a n repeated experiment design. The nonlinear effect of retention time alignment could be modeled by this method, and the negative effect of the peptides with multiple chromatographic peaks could also be overcome. The results show that the time complexity of this algorithm was reduced and the accuracy of it can satisfy the requirement of the following date processing steps.