“…The Baies method relates to probability parametric methods of classification; the prototype is described in [4]. The logarithm of the probability that a compound C belongs to the class of active compounds a when it contains where P(B ij C ∈ a) is the a priori probability of encountering the descriptor B ij in the class of active a compounds.…”
Section: Methods Of Predictionmentioning
confidence: 99%
“…Formation of a training sample from 246 known carcinogenic compounds and 247 compounds that definitely do not exhibit carcinogenic properties is described in detail in [4]. Non-parametric methods of classification were employed for prediction in the present study, therefore the models of the generalised patterns of classes of active/inactive compounds were constructed not in the form of catalogues of descriptors, as in [4,9], but in the form of matrices containing full QL descriptions of all the active and inactive structures.…”
Section: Forming the Training Samplementioning
confidence: 99%
“…The QL descriptions of the sulphenamide accelerators were generated both for the ordinary structural formulae (see Table 1), and for the variants with internal bonds (see Figure 1) -six descriptions in all. We used an extended version of the QL language [9], which differs from the basic version [4] in that the carbon-containing descriptors are more detailed. Altogether, 11 types of descriptors of 1-4 ranks are defined in QL.…”
Section: Description Of the Structure Of The Compounds On The Basis Omentioning
confidence: 99%
“…The present work employs computer-based techniques for making an a priori assessment of the possible carcinogenicity of three sulphenamide curing accelerators (called sulphenamide accelerators hereinafter), the names and structural formulae of which are given in Table 1. The techniques are based on relations obtained by the methods of pattern recognition theory, linking absence/presence of carcinogenicity to the structure of the compounds, represented as a set of structural descriptors in the QL language [4]. The method proved to be highly effective: the accuracy of predicting carcinogenic risk reached 94.7%.…”
“…The Baies method relates to probability parametric methods of classification; the prototype is described in [4]. The logarithm of the probability that a compound C belongs to the class of active compounds a when it contains where P(B ij C ∈ a) is the a priori probability of encountering the descriptor B ij in the class of active a compounds.…”
Section: Methods Of Predictionmentioning
confidence: 99%
“…Formation of a training sample from 246 known carcinogenic compounds and 247 compounds that definitely do not exhibit carcinogenic properties is described in detail in [4]. Non-parametric methods of classification were employed for prediction in the present study, therefore the models of the generalised patterns of classes of active/inactive compounds were constructed not in the form of catalogues of descriptors, as in [4,9], but in the form of matrices containing full QL descriptions of all the active and inactive structures.…”
Section: Forming the Training Samplementioning
confidence: 99%
“…The QL descriptions of the sulphenamide accelerators were generated both for the ordinary structural formulae (see Table 1), and for the variants with internal bonds (see Figure 1) -six descriptions in all. We used an extended version of the QL language [9], which differs from the basic version [4] in that the carbon-containing descriptors are more detailed. Altogether, 11 types of descriptors of 1-4 ranks are defined in QL.…”
Section: Description Of the Structure Of The Compounds On The Basis Omentioning
confidence: 99%
“…The present work employs computer-based techniques for making an a priori assessment of the possible carcinogenicity of three sulphenamide curing accelerators (called sulphenamide accelerators hereinafter), the names and structural formulae of which are given in Table 1. The techniques are based on relations obtained by the methods of pattern recognition theory, linking absence/presence of carcinogenicity to the structure of the compounds, represented as a set of structural descriptors in the QL language [4]. The method proved to be highly effective: the accuracy of predicting carcinogenic risk reached 94.7%.…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.