RESUMO
Effective representation of molecules is a crucial factor affecting the performance of artificial intelligence models. This study introduces a flexible, fragment-based, multiscale molecular representation framework called t-SMILES (tree-based SMILES) with three code algorithms: TSSA (t-SMILES with shared atom), TSDY (t-SMILES with dummy atom but without ID) and TSID (t-SMILES with ID and dummy atom). It describes molecules using SMILES-type strings obtained by performing a breadth-first search on a full binary tree formed from a fragmented molecular graph. Systematic evaluations using JTVAE, BRICS, MMPA, and Scaffold show the feasibility of constructing a multi-code molecular description system, where various descriptions complement each other, enhancing the overall performance. In addition, it can avoid overfitting and achieve higher novelty scores while maintaining reasonable similarity on labeled low-resource datasets, regardless of whether the model is original, data-augmented, or pre-trained then fine-tuned. Furthermore, it significantly outperforms classical SMILES, DeepSMILES, SELFIES and baseline models in goal-directed tasks. And it surpasses state-of-the-art fragment, graph and SMILES based approaches on ChEMBL, Zinc, and QM9.
RESUMO
BACKGROUND: Saffron has gained people's attention and love for its unique flavor and valuable edible value, but the problem of saffron adulteration in the market is serious. It is urgent for us to find a simple and rapid identification and quantitative estimation of adulteration in saffron. Therefore, excitation-emission matrix (EEM) fluorescence combined with multi-way chemometrics was proposed for the detection and quantification of adulteration in saffron. RESULTS: The fluorescence composition analysis of saffron and saffron adulterants (safflower, marigold and madder) were accomplished by alternating trilinear decomposition (ATLD) algorithm. ATLD and two-dimensional principal component analysis combined with k-nearest neighbor (ATLD-kNN and 2DPCA-kNN) and ATLD combined with data-driven soft independent modeling of class analogies (ATLD-DD-SIMCA) were applied to rapid detection of adulteration in saffron. 2DPCA-kNN and ATLD-DD-SIMCA methods were adopted for the classification of chemical EEM data, first with 100% correct classification rate. The content of adulteration of adulterated saffron was predicted by the N-way partial least squares regression (N-PLS) algorithm. In addition, new samples were correctly classified and the adulteration level in adulterated saffron was estimated semi-quantitatively, which verifies the reliability of these models. CONCLUSION: ATLD-DD-SIMCA and 2DPCA-kNN are recommended methods for the classification of pure saffron and adulterated saffron. The N-PLS algorithm shows potential in prediction of adulteration levels. These methods are expected to solve more complex problems in food authenticity. © 2023 Society of Chemical Industry.