An algorithm to calibrate ionic isotopes using data mining strategy in hyphenated chromatographic datasets from herbal samples.

The bottleneck of analytical instrument itself and non-ideal instrumental performance will produce a certain degree of drifts between the measured isotopes and the true values. An AAID-IC algorithm was thereby proposed to keep the isotopic distributions more accurate in hyphenated instruments, e.g. Gas Chromatography (GC)/ Liquid Chromatography (LC) - Mass Spectrometry (MS). During this data mining process, chemical information will be fully used from dozens of data points in retention time (rt) dimension: the target isotopes were firstly re-constructed in mass charge ratio (m/z) dimension; their re-calculation values were then averaged from an interesting rt zone; the calibration functions were followed established based on a well-defined series of calibration ions. It is worth mentioning that natural metabolites in complex samples can be identified as reference materials to amend the target isotopes. Next, the corrected mass axes (m/z values)/isotope abundances were transformed into an ionic isotopic curve using Gaussian box. Taking herbal sample as an example, AAID-IC can better reduce the systematic and random errors of the m/z ions in one run environment, whether it's profile or bar graph from any type of MS and any ionization method employed. Finally, the calibrated values can be utilized to deduce the elemental compositions of molecular (fragment) ions in GC/LC-MS determination.