A.B.O.S. a novel data analysis tool evaluating complex data out of matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry

A.B.O.S. (A Better Omic System; version 1.1.0; Ars Nova AG, Esslingen, Germany) a simple, interactive software for the analysis of omics data has recently been developed to classify and evaluate biological properties or processes in complex datasets. The tool utilises self-learning algorithms that exploit group-specific properties from large datasets and applies a combination of multivariate analysis techniques such as principal component analysis (PCA), weighting the different variables/parameters according to their discriminatory power. Unlike PCA, however, the program can handle data that are not normally distributed and accounts for the presence of outliers and missing data. The software carries out predictive identifications based on pre-assigned learning groups that can be either detected automatically or defined manually. By combining all parameters shared by the members of each learning groups, the software calculates two ideal reference groups and classifies unknown elements based on their relative distance to these groups. Along with the proposed classification of samples, it also identifies the most important parameters that allow differentiating between classes.

Martin Christner, Dirk Dressler , Mark Andrian , Claudia Reule , Orlando Petrini PLOS