Satoshi Fujishima

Predictive Activity Profiling of Drugs by Topological-Fragment-Spectra-Based Support Vector Machines

Aiming at the prediction of pleiotropic effects of drugs, we have investigated the multilabel classification of drugs that have one or more of 100 different kinds of activity labels. Structural feature representation of each drug molecule was based on the topological fragment spectra method, which was proposed in our previous work. Support vector machine (SVM) was used for the classification and the prediction of their activity classes. Multilabel classification was carried out by a set of the SVM classifiers. The collective SVM classifiers were trained with a training set of 59,180 compounds and validated by another set (validation set) of 29,590 compounds. For a test set that consists of 9,864 compounds, the classifiers correctly classified 80.8% of the drugs into their own active classes. The SVM classifiers also successfully performed predictions of the activity spectra for multilabel compounds.

MolSpace: a computer desktop tool for visualization of massive molecular data

The authors have developed a software tool, MolSpace, to visualize massive molecular datasets. MolSpace can project a set of massive multivariate data onto a visual space (two- or three-dimensional space) by means of principal component analysis. MolSpace allows users not only to draw a scatter diagram of the data but also to display their two- or three-dimensional molecular structures as the objects in that space. With a probe (a molecular object) the user can navigate vast data spaces, thus facilitating understanding of the data structure. In addition, partial space searching is also available that is based on similarity searching techniques. It is possible to interrogate a three-dimensional structure of a chemical compound that corresponds to each object on the space in real time. The detail of the system is discussed with an illustrative example.

Classification of dopamine antagonists using TFS-based artificial neural network.

In the former work, the authors proposed the Topological Fragment Spectral (TFS) method as a tool for the description of the topological structure profile of a molecule. This paper describes the TFS-based artificial neural network (TFS/ANN) approach for the classification and the prediction of pharmacological active classes of chemicals. Dopamine antagonists of 1227 that interact with different types of receptors (D1, D2, D3, and D4) were used for the training. The TFS/ANN successfully classified 89% of the drugs into their own active classes. Then, the trained model was used for predicting the class of unknown compounds. For the prediction set of 137 drugs that were not included in the training set, the TFS/ANN model predicted 111 (81%) drugs of them into their own active classes correctly.

Classification of pharmacological activity of drugs using support vector machine

In the present work, we investigated an applicability of Support Vector Machine (SVM) to classify of pharmacological activities of drugs. The numerical description of each drugs chemical structure was based on the Topological Fragment Spectra (TFS) proposed by the authors. 1,227 Dopamine antagonists that interact with different types of receptors (Dl, D2, D3 and D4) were used for training SVM. For a prediction set of 137 drugs not included in the training set, the obtained SVM classified 123 (89.8 %) drugs into their own activity classes correctly. The comparison between using SVM and artificial neural network will also be discussed.

Pharmacophore knowledge refinement method in the chemical structure space

Studies on the structure-activity relationship of drugs essentially require a relational learning scheme in order to extract meaningful chemical subgraphs; however, most relational learning systems suffer from a vast search space. On the other hand, some propositional logic mining methods use the presence or absence of chemical fragments as features, but rules so obtained give only crude knowledge about part of the pharmacophore structure. This paper proposes a knowledge refinement method in the chemical structure space for the latter approach. A simple hill-climbing approach was shown to be very useful if the seed fragment contains the essential characteristic of the pharmacophore. An application to the analysis of dopamine D1 agonists is discussed as an illustrative example.

Extended study of the classification of dopamine receptor agonists and antagonists using a TFS-based support vector machine

We previously investigated the classification and prediction of dopamine D1 receptor agonists and antagonists using a topological fragment spectra (TFS)-based support vector machine (SVM), in which the dataset contained noise compounds that had no D1 receptor activity. This work extended the dataset to seven activity classes (dopamine D1, D2, and auto-receptor agonists, and D1, D2, D3, and D4 antagonists) and increased the noise ratio to ten times that of active compounds. In total, this study used 16,008 compounds for training and 1,779 compounds for prediction. The TFS-based SVM gave good, stable results for both classification and prediction, even in the case that included ten times the noise data. The resulting model correctly predicted 97.6% of the prediction set of 1,779 compounds.