Author/Editor		Kompare, Boris
Title		Activity prediction of drugs using artificial intelligence tools
Type		članek
Source		In: Lavrač N, editor. CADAM-95. Zbornik Računalniška analiza medicinskih podatkov; 1995 nov 27-28; Bled. Ljubljana: Inštitut Jožef Štefan,
Publication year		1995
Volume		str. 164-79
Language		eng
Abstract		In the paper construction of QSAR (Quantitative Structure-Activity Relationship) models is demonstrated. The QSAR models are used to predict the unknown (difficult to measure, time-consuming, or expensively determinable) activity of a substance of concern, if its structural and-or other easily (cheaply, time-efficiently) measurable properties are known. Several models constructed with artificial intelligence (AI) tools are shown and commented. The experience of the use of the AI tools and of the interpretability and the use of the constructed models are discussed, too. The example used deals with biological degradability of chemicals in aerobic aquatic environment. But the shown methodology can be of course used in pharmacology during production or testing of drugs, as well as for prediction of substance's activity in human body, or for analysis of medical data.
Descriptors		STRUCTURE-ACTIVITY RELATIONSHIP ARTIFICIAL INTELLIGENCE DRUGS BIODEGRADATION COMPUTER SIMULATION