Isotope cluster detection and validation

In [1] we propose an approach for the exhaustive detection and mass-specific validation of isotope clusters. In this approach we perform an targeted peak picking for the detection of isotopoloue features and we perform a massspecific validation of putative isotope clusters. The approach is implemented in the following workflow using functions from xcms [2] and CAMERA [3]. First, we perform peak picking without any prior knowledge which we denote as traditional peak picking. Here, we use the centWave algorithm [4] which applies a heuristic for the detection of ROIs (ROI identification step). Given the set of detected ROIs, chromatographic peaks are extracted using continuous wavelet transformation (ROI examination step). This step results in a peak table with one row for each detected feature and one column for each feature property such as m/z, retention time, integrated peak area, and signal-to-noise ratio. Second, we perform the proposed targeted peak picking as described in [1]. Here, a set of isotope ROIs is predicted on basis of the previously extracted peak table. Given the set of predicted isotope ROIs, chromatographic peaks are extracted using continuous wavelet transformation (ROI examination step). Notably, this ROI examination step is identical to the ROI examination step in the traditional peak picking step with the exception that we use relaxed peak picking parameters this time. This step results in an extended peak table which is enriched with features corresponding to isotope peaks. Third, we extract pseudospectra from the extended peak table [3]. This step aims at the extraction of compound spectra on basis of the retention times, but multiple coeluting compounds are potentially assigned to the same spectrum which is the reason for the usage of the term pseudospectrum. In case of multiple raw data files a retention time correction (xmcs function retcor) can be advisable prior to the extraction of pseudospectra. This step results