DeepTox: Toxicity prediction using deep learning

In this study, a new method with one pretreatment coupled with LC-Q-TOF/MS detection was developed for nontarget and high-throughput screening of 556 pesticides residues in four kinds of tea. Pesticides were identified by matching the data acquired from tea samples with the created accurate mass LC-Q-TOF/MS database and MS/MS library. There were 336 pesticides, accounting for 60.4% of the total target pesticides, meeting the Recovery-RSD standard at 10 μg/kg, which demonstrates the applicability and accuracy of this method. In addition, there was no need for Reference: standards in this study; MS/MS fragment information was typically matched to the database to enable conclusive confirmation of chemical identity. What's more, full-scan technology has brought nontarget screening into its own, and pesticide residue detection has achieved rapid progress in becoming automatic, digitized, and informationalized. In conclusion, the application of LC-Q-TOF/MS is an effective tool for rapid screening of pesticide residues in tea and will play an important role in food safety supervision for pesticide residue.