Table 2.

Summary of binning methods with a brief description of the main features of each method

Method	Description	Remarks	References
Equal size binning	Spectrum is divided in simple rectangular regions of the same size. Each bin span the same number of spectral points	Straightforward and fast to apply. Works quite well despite simplicity. The bins size in ppm needs to be specified (0.04, 0.02, 0.01 ppm the most common choices). Peaks can be splitted across multiple bins	Izquierdo-Garcia et al. (2011)
Gaussian binning	A Gaussian kernel weights the signal contribution relative to distance from bin center, and the overlap between bins is controlled by the kernel standard deviation	Overlapping bins are used. Very robust to peak shifts. Two parameters (not easy to tune): standard deviation and step size that make a trade-off between loose of information and robustness	Anderson et al. (2008)
Adaptive-intelligent binning	Iterative algorithm that uses variable bin sizes adaptively inferred from spectra	No arbitrary parameters, reference spectra, a priori knowledge, or data modifications are required. Low-intensity peaks could be troublesome. Noise regions need to be specified	De Meyer et al. (2008)
Dynamic adaptive binning	Bin boundaries are determined by optimizing an objective function using a dynamic programming strategy. The objective function measures the quality of a bin configuration based on the number of peaks per bin	Ability to create bins containing a single peak. Two main parameters and several other parameters for peaks identifications	Anderson et al. (2011)
Adaptive binning using wavelet transform	Wavelet transforms are used to detect peaks in a reference spectrum. Integration is then performed over these peaks in each of the sample spectra. What constitutes a peak is determined by the amount of smoothing implicit in the wavelet transform	Noise regions are excluded. The amount of smoothing depends on the number of levels of the wavelet transform and can be adjusted according to the data resolution and the shifts expected between samples	Davis et al. (2007)
Optimized bucketing algorithm	A bucketing method that optimizes bucket sizes by setting their boundaries at the local minima determined through the average NMR spectrum	A mathematically simple approach. Two parameters need to be chosen, requiring visual inspection of the result	Sousa et al. (2013)