Database-independent analysis of probable post-translational modifications of soil proteins across ecosystems

Abstract

The identification rate of measured peptide spectra to proteins barely scratches 1% in best-case scenarios. Hundreds of thousands of valuable spectra are lost as no viable match in the database is found. Here, we apply the delta m/z plot that was previously implemented in MSnbase as tool for quality control to 63 soil samples from three ecosystems with different vegetation (39 forests, 11 grasslands, and 13 shrublands) with the aim to extract probable post-translational modifications (PTM) without the need of a reference database. The validity of the approach was verified with amino acids proposed for their respective 1 Da mass interval and compared to their relative abundance in proteins. We found that the average probable PTM and most known PTMs proposed for the mass intervals are similar across ecosystems. Otherwise, 11 mass intervals changed significantly in relative abundance in the three ecosystems but only for one an annotation could be proposed. Our approach not only highlights the opportunity of the database-independent analysis in soil metaproteomics but paves the way for targeted analysis of the yet unknown PTMs.