Metabolomics approach to understand the resource partitioning in Chlorella during growth

We use untargeted high-resolution mass spectrometry to understand the metabolic differences at exponential and stationary growth stages using 22 Chlorella strains collected from South East Asia. Using this data, we demonstrate the use of a filtering procedure based on the SVD on the untargeted metabolite profile data, compared over two growth stages and run in four batches, and to remove structure in data related to day of sample assay. This approach preserves signal of demonstrable biological origin (strain-related variation in mass feature intensity) after minimizing the influence of batch effects. Our approach will be broadly applicable in metabolomics analysis to identify and remove batch effects. See MTBLS129 for associated study.