Combining Molecular Modelling with the Use of Artificial Neural Networks as an Approach to Predicting Substituent Constants and Bioactivity

Авторы: Baskin I.I., Keschtova S.V., Palyulin V.A., Zefirov N.S.
Сборник: Molecular Modeling and Prediction of Bioactivity / Ed. Gundertofte K., Jørgensen F.S
Год издания: 2000
Место издания: Kluwer Plenum New York
Первая страница: 468
Последняя страница: 469
DOI: 10.1007/978-1-4615-4141-7_122
Аннотация: Nowadays neither molecular model, no matter how elaborate it may be, is able to encompass all possible interactions, in which a real chemical/biological system is involved, as well as to take them properly into account. In this connection the problem of relating theoretically derived molecular characteristics with experimentally observed properties becomes very important. As a way of solving this problem, we see the use of a technique that would allow one to reveal nonlinear relationships of any complexity between theoretically derived characteristics of molecules and observed experimental properties. As the most promising candidate for that, we consider the use of artificial neural networks, since only this approach allows to find relationships of any complexity between parameters without the need to know in advance or guess its generic form.
Добавил в систему: Баскин Игорь Иосифович

	ИСТИНА	Войти в систему Регистрация
	Интеллектуальная Система Тематического Исследования НАукометрических данных
	Главная Поиск Статистика О проекте Помощь

ИСТИНА