pseudoDrift

pseudoDrift normalizes and corrects LC-MS metabolomic data to mitigate signal drift and batch effects for more accurate downstream biological analysis.

Key Features:

• Data Simulation and Outlier Detection: Simulates LC-MS metabolomic data and detects outliers to characterize and quantify technical error.
• Training and Testing Approach: Implements a training and testing methodology that captures and can optionally correct technical errors such as signal drift and batch effects.
• Flexible Correction Strategies: Provides configurable correction strategies allowing tailoring of normalization and correction to study-specific requirements.

Scientific Applications:

• Large-Scale Metabolomic Studies: Normalizes and corrects large-scale LC-MS metabolomic datasets to recover biological signals obscured by technical variation.
• Targeted LC-MS Profiling in Maize: Applied to a targeted LC-MS study profiling 33 phenolic compounds from seedling stem tissue across 602 genetically diverse non-transgenic maize inbred lines to study specialized metabolism dynamics.

Methodology:

Uses simulation-based benchmarking and a training/testing procedure to detect and optionally correct technical errors, with simulations used to validate performance against existing methods.