PG-Metrics: A chemometric-based approach for classifying bacterial peptidoglycan data sets and uncovering their subjacent chemical variability

New approaches and techniques for bacterial cell wall analysis

Peptidoglycan (PG) has remained for decades in the spotlight of the never-ending battle against pathogenic bacteria as this essential bacterial structure is one of the most successful targets for antibiotics. Most of our current understanding about the composition, architecture, and dynamics of the PG relies on techniques which have experienced great technological and methodological improvements in the past years. Here we summarize recent advances in these methods with the intention to furnish a valuable resource for both PG experts and newcomers.

Introducing a novel procedure for peak alignment in one-dimensional 1H-NMR spectroscopy: a prerequisite for chemometric analyses of wine samples

Shifted peak positions in 1H-nuclear magnetic resonance (1H-NMR) spectroscopy of wine samples are inevitably occurring mainly due to variations in the sample matrix, which consists of ethanol, glycerol, carbohydrates, acids, phenolic compounds, minerals, and aroma compounds. Slight variations in pH during sample preparation or fluctuations in instrumental factors may contribute to shifted peak positions that need to be corrected before subjecting the NMR data to chemometric techniques to ensure samples are compared on the correct chemical shift scale. In the current work, a novel procedure for correcting 1H-NMR spectroscopy peak positions was developed by mapping of the raw NMR spectra on a common chemical shift axis using a simple interpolation approach. The mapping allowed a substantial correction of peak positions and subsequently reduced the computational burden in further spectral processing. Fine-tuning of the peak alignments was carried out efficiently by interval-wisely applying the correlation optimized warping (COW) algorithm. Our preceding mapping approach enabled the use of substantially simpler alignment parameters of the COW algorithm, thereby accelerating the whole peak alignment process. The developed procedure may also be suitable for facilitating NMR analyses of other sample types, such as agricultural, clinical or pharmaceutical samples in targeted or untargeted analytical approaches.