QSPR in forensic analysis – The prediction of retention time of pesticide residues based on the Monte Carlo method

Foodinformatic prediction of the retention time of pesticide residues detected in fruits and vegetables using UHPLC/ESI Q-Orbitrap

“…In addition, this approach enables the elucidation of the molecular mechanisms of retention phenomena in diverse stationary phases along with the design of new phases with required properties as well as to facilitate protein identification in proteomics studies (Kaliszan, 2007). Thus, several QSPR studies were reported in the literature to predict the t R of pesticide residues (Dashtbozorgi et al, 2013;Torrens & Castellano, 2014;Zdravković et al, 2018). Our research group has also been interested in QSPR studies for the prediction of chromatographic retention indices in the field of food science (foodinformatics) (Rojas et al, 2019;Rojas et al, 2018), as well as the in silico modeling of the water solubility of pesticides (Fioressi et al, 2019).…”

Chemometric modeling of odor threshold property of diverse aroma components of wine

“…Previously also, many researchers reported different strategies for selection of descriptors prior to development of nal QSAR/QSPR models. [18][19][20][21][22] Note that, while developing the models, we have kept aside some compounds as hold out samples (test set) which have not been used for the model development. Aer the model has been developed, the predictive quality of the developed model has been evaluated based on the experimental values of the hold out samples.…”

The development of pseudo-SIM/MRM and risk-based screening methods for characterization of human exposome

“…Retention time prediction by random forest regression models based on QSRR is widely accepted for non-targeted analysis (Bonini et al 2020;Cao et al 2015b;Zdravković et al 2018) Overall, the model achieved linear correlations R 2 = 0.86 (p = 2.4e-11) in test data with mean absolute error (MAE) of 1.00 min (Fig. 6.3A).…”

“…From our previous work in retention time prediction, the random forest algorithm is often used for its excellent prediction power in chemical and physical properties like retention time for small molecules based on SMILES structure and molecular descriptors (Yang et al 2020). However, accurate retention time prediction is demanded as a retention time in most references was predicted based on local LCMS conditions (Bonini et al 2020;Cao et al 2015b;Zdravković et al 2018). Tedious models' modification is required for applications.…”

“…Retention time prediction by random forest regression models based on QSRR is widely accepted for non-targeted analysis (Bonini et al 2020;Cao et al 2015b;Zdravković et al 2018). In our previous study, we achieved a comparable prediction accuracy by QSRR-based regression model with features selection by random forest algorithm (Yang et al 2020).…”