PR000686 (Project)

Description:The goal of this study was to validate the performance of a custom-written software tool, called Trace, for finding molecular features from ultrasensitive metabolomics experiments using high-resolution mass spectrometry. The software uses a trained neural network model to extract molecular features. As model for validation, we performed MS profiling of single identified cells from early developing embryos of the South African clawed frog (Xenopus laevis) using a custom-built capillary electrophoresis electrospray ionization platform coupled to a quadrupole time-of-flight mass spectrometer. The MS dataset from these measurements was manually curated for molecular features, and the resulting list of molecular features were used to test the robustness and accuracy of Trace at predicting molecular features that were detected from the single cells.