A novel strategy for extracted ion chromatogram extraction to improve peak detection in UPLC-HRMS

Abstract

Extracted ion chromatogram (EIC) extraction is the primary and fundamental step in ultraperformance liquid chromatography-high resolution mass spectrometry (UPLC-HRMS). Unfortunately, research studies on this aspect are far from satisfactory for routine analysis. To address this problem, we provide AiCN-EIC, a novel adaptive ion-connecting network-based strategy for EIC extraction. In this strategy, ions are sorted on the basis of their m/z values and are then connected to form EIC clusters in accordance with their m/z differences. An EIC cluster is constructed by placing ions at the corresponding positions of a cluster in accordance with their scan numbers. A multiscale Gaussian smoothing-based peak detection strategy is introduced after bad EICs are removed. We use two datasets to evaluate the performance of AiCN-EIC. The results indicate that AiCN-EIC can efficiently extract EICs from the results of UPLC-HRMS while exhibiting the advantages of insensitivity to m/z tolerance and threshold value in data acquisition. Quantitative results confirm that the performance of our developed strategy is comparable with that of several state-of-the-art methods. AiCN-EIC has been added into the recently developed Matlab GUI toolbox for UPLC-HRMS-based metabolomics, which can be obtained from http://software.tobaccodb.org/software/aicneic.