2020/09/24 by Karsten Wüllems, Wüllems, Karsten, Tim W. Nattkemper +1
Biochemistry, Genetics and Molecular Biology · Chemistry · Environmental Science · #Applications (stat.AP) #Computation (stat.CO) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Biological sciences #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Isotope Analysis in Ecology #Mass Spectrometry Techniques and Applications #Metabolomics and Mass Spectrometry Studies #Quantitative Methods (q-bio.QM) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2009.14677
openalex publication_date 2020/09/24 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
The computational analysis of Mass Spectrometry Imaging (MSI) data aims at\nthe identification of interesting mass co-localizations and the visualization\nof their lateral distribution in the sample, usually a tissue cross section.\nBut as the morphological structure of tissues and the different kinds of mass\nco-localization naturally show a huge diversity, the selection and tuning of\nthe computational method is a time-consuming effort. In this work we address\nthe special problem of computationally grouping mass channel images according\nto their similarities in their lateral distribution patterns. Such an analysis\nis driven by the idea, that groups of molecules that feature a similar\ndistribution pattern may have a functional relation. But the selection of the\nsimilarity function and other parameters is often done by a time-consuming and\nunsatsifactory trial and error. We propose a new flexible workflow scheme\ncalled SoRC (sum of ranked cluster indices) for automating this tuning step and\nmaking it much more efficient. We test SoRC using three different data sets\nacquired from the lab for three different kinds of samples (barley seed, mouse\nbladder tissue, human PXE skin). We show, that SORC can be applied to score and\nvisualize the results obtained with the applied methods in short time without\ntoo much effort. In our application example, the SoRC results for the three\ndata sets reveal that a) some well-known similarity functions are suited to\nachieve good results for all three data sets and b) for the MSI data featuring\na higher degree of irregularity improved results can be achieved by applying\nnon-standard similarity functions. The SoRC scores computed with our approach\nindicate that an automated testing and scoring of different methods for mass\nchannel image grouping can improve the final outcome of a study by finally\nselecting the methods of the highest scores.\n