Zhang Weiyang, Cai Jing, Gao Wenqing, Han Renlu, Wang Haixing, Wu Yanfei, Wu Jiawei, Wu Yong, Wang Chenlu, Tang Keqi, Yu Jiancheng
Faculty of Electrical Engineering and Computer Sciences, Ningbo University, Ningbo, China.
Zhejiang Engineering Research Center of Advanced Mass Spectrometry and Clinical Application, Institute of Mass Spectrometry, Ningbo University, Ningbo, China.
Rapid Commun Mass Spectrom. 2023 Sep 30;37(18):e9603. doi: 10.1002/rcm.9603.
In the field of separation science, ion mobility spectrometry (IMS) plays an important role as an analytical tool. However, the lack of sufficient structural resolution is a common problem in qualitative and quantitative analysis using IMS. A method is needed to solve the problem of overlapping peaks caused by insufficient resolution.
The method uses multiple strategies to more effectively use population information to balance exploration and exploitation capabilities, prevent local optimization, accurately resolve overlapping peaks, quickly obtain optimal spectral peak model coefficients, and accurately identify compounds.
Multistrategy JAYA algorithm's (MSJAYA) performance is compared with improved particle swarm optimization (IPSO), dynamic inertia weight particle swarm optimization (DIWPSO), and multiobjective dynamic teaching-learning-based optimization (MDTLBO). The analysis shows that MSJAYA's maximum separation error is within 0.6%, a level of accuracy not guaranteed by the other algorithms. In addition, the separation error fluctuates within a much smaller range, demonstrating MSJAYA's superior robustness.
Compared with other overlapping peak separation algorithms, MSJAYA is more applicable because no special parameters are used. The method allows fast deconvolution analysis of strong overlapping peaks with multiple components, which greatly improves the resolution of IMS.
