Structure–camphor odour relationships using the Generation and Selection of Pertinent Descriptors approach

• Published in: Chemometrics and intelligent laboratory systems 1999-06, Vol.48 (1), p.35-46
• Main Authors: Zakarya, D., Chastrette, M., Tollabi, M., Fkih-Tetouani, S.
• Format: Article
• Language: English
• Subjects: Camphor odour; Discriminant analysis; GSPD methodology; KNN; Neural network
• ISSN: 0169-7439; 1873-3239
• DOI: 10.1016/S0169-7439(99)00003-9

Structure–camphor odour relationships were carried out using GSPD (Generation and Selection of Pertinent Descriptors) or GESDEM (Genération et Sélection de Descripteurs et Elaboration de Motifs). This methodology consists in constructing descriptors from fragments. Their size (called order) is increased stepwise to obtain an optimum order giving a best classification. The set of studied compounds included 99 aliphatic alcohols, for which olfactory properties have been described by Schnabel et al. [K.O. Schnabel, H.D. Belitz, C.V. Ranson, Untersuchungen zur Struktur–Aktivitats–Beziehung bei Geruchsstoffen, Z. Lebensm.-Unters.-Forsch. 187 (1988) 215–223]. Classification of compounds was tested by means of artificial neural network (NN) with back-propagation algorithm, Kth nearest neighbour (KNN) and discriminant analysis (DA). After GSPD, all compounds were well-classified and 93% of them were well-predicted by means of leave-one-out method using NN. A new test sample composed of 42 alcohols (25 camphor and 17 non-camphor) with structures analogue to these of training set was collected from Belstein and trained on weight matrix of network which served in classification phase. The results of prediction were at 90.5% in agreement with those of literature.