Blinova V G, Dobrynin D A, Finn V K, Kuznetsov S O, Pankratova E S
All-Russia Institute for Scientific and Technical Information (VINITI), Usievicha 20, 125219 Moscow, Russia.
Bioinformatics. 2003 Jul 1;19(10):1201-7. doi: 10.1093/bioinformatics/btg096.
A model for learning potential causes of toxicity from positive and negative examples and predicting toxicity for the dataset used in the Predictive Toxicology Challenge (PTC) is presented. The learning model assumes that the causes of toxicity can be given as substructures common to positive examples that are not substructures of negative examples. This assumption results in the choice of a learning model, called the JSM-method, and a language for representing chemical compounds, called the Fragmentary Code of Substructure Superposition (FCSS). By means of the latter, chemical compounds are represented as sets of substructures which are 'biologically meaningful' from the expert point of view.
The chosen learning model and representation language show comparatively good performance for the PTC dataset: for three sex/species groups the predictions were ROC optimal, for one group the prediction was nearly optimal. The predictions tend to be conservative (few predictions and almost no errors), which can be explained by the specific features of the learning model.
by request to finn@viniti.ru; serge@viniti.ru, http://ki-www2.intellektik.informatik.tu-darmstadt.de/~jsm/QDA.
提出了一种从正例和负例中学习毒性潜在原因并对预测毒理学挑战（PTC）中使用的数据集进行毒性预测的模型。该学习模型假设毒性原因可以表示为正例中常见而负例中不存在的子结构。这一假设导致选择了一种称为JSM方法的学习模型，以及一种用于表示化合物的语言，称为子结构叠加碎片代码（FCSS）。借助后者，化合物被表示为从专家角度来看“具有生物学意义”的子结构集合。
所选的学习模型和表示语言在PTC数据集上表现出相对较好的性能：对于三个性别/物种组，预测是ROC最优的，对于一个组，预测几乎是最优的。预测往往较为保守（预测数量少且几乎没有错误），这可以通过学习模型的特定特征来解释。
可通过向finn@viniti.ru；serge@viniti.ru发送请求获取，网址为http://ki-www2.intellektik.informatik.tu-darmstadt.de/~jsm/QDA 。
