Computational prediction of retention times of veterinary antibiotics obtained by liquid chromatography-mass spectrometry

CRISTIAN ROJAS; NICOLE SARMIENTO; EMILIA AYORA; REINALDO PIS DIEZ

Journal of the Science of Food and Agriculture

John Wiley and Sons Ltd

Año: 2024

A database of 122 antibiotics was selected in which the tR was measured using a Hypersil GOLD column. An optimalthree-feature model was developed by integrating the unsupervised variable reduction, replacement method variable subsetselection, and multiple linear regression. The negligible differences among the coefficient of determination and the root-mean-square error for the training set (R2 = 0.902 and RMSEC = 0.871) and test set (Q2 = 0.854 and RMSEP = 1.064) indicatea stable and predictive model. In a further step, a more in-depth explanation of the mechanism of action of each descriptor inpredicting the tR is provided, with the construction of the theoretical chemical space for accurate predictions of new antibiotics.