Pharmaceutical Drug Design Using Dynamic Connectionist Ensemble Networks
This article presents a dynamic ensemble neural network model for a pharmaceutical drug design problem. Designing drugs is a current problem in the pharmaceutical research domain. By designing a drug, we mean to choose some variables of drug formulation (
- PDF / 209,166 Bytes
- 11 Pages / 439 x 666 pts Page_size
- 36 Downloads / 165 Views
2 3
Norwegian Center of Excellence, Center of Excellence for Quantifiable Quality of Service, Norwegian University of Science and Technology, O.S. Bragstads plass 2E, Trondheim, Norway [email protected] Department of Computer Science, Babes-Bolyai University, Cluj-Napoca, 3400, Romania [email protected] University Iuliu Hatieganu, Faculty of Medicine, Department of Biostatistics and Medical Informatics, Cluj-Napoca, Romania [email protected]
Summary. This article presents a dynamic ensemble neural network model for a pharmaceutical drug design problem. Designing drugs is a current problem in the pharmaceutical research domain. By designing a drug, we mean to choose some variables of drug formulation (inputs), for obtaining optimal characteristics of drug (outputs). To solve such a problem, we propose a dynamic ensemble neural network model and the performance is compared with several neural network architectures and learning approaches. The idea is to build a dynamic ensemble neural network depicting the dependence between inputs and outputs for the drug design problem. Bootstrap techniques were used to generate more samples of data since the number of experimental data is reduced due to the costs and time durations of experimentations. We obtain in this way a better estimation of some drug parameters. Experiment results indicate that the proposed method is efficient.
13.1 Introduction This article presents a dynamic neural network ensemble for modeling the situations that interfere in the process of designing drugs. By designing a drug, we mean to choose some variables of drug formulation, for obtaining optimal characteristics of drug [2, 3]. Our application is made on a particular class of drugs, namely retard drugs. We approach this problem with a bootstrap simulation which is suitable in some particular situations [1, 4]. The problem comes from the pharmaceutical research activity. It refers to a specific class of drugs that has delayed action called generically retard drugs. The pharmaceutical experimental situation leads to a mathematical optimization problem [12, 14, 15]. The pharmacist researcher must take into account several variables A. Abraham et al.: Pharmaceutical Drug Design Using Dynamic Connectionist Ensemble Networks, Studies in Computational Intelligence (SCI) 123, 221–231 (2008) c Springer-Verlag Berlin Heidelberg 2008 www.springerlink.com
222
A. Abraham et al. Table 13.1. Sample data showing the inputs and outputs Variables of formulation: Inputs ExpNo 1 2 3 4 5 6 7 8 9 10 11
X1 20 40 20 40 20 40 20 40 30 30 30
X2 2 2 8 8 2 2 8 8 5 5 5
X3 3 3 3 3 9 9 9 9 6 6 6
X4 5 0 0 5 5 0 0 5 2.5 2.5 2.5
X5 1 0 1 0 0 1 0 1 0.5 0.5 0.5
Responses: Outputs Y1 84.0 71.9 92.5 88.1 99.2 68.2 99.1 83.9 85.0 81.2 85.0
Y2 973.8 1150.0 1121.4 1200.0 910.0 985.1 1010.0 925.4 1055.8 1030.0 1060.0
Y3 4.2 1.6 4.2 3.7 5.8 4.1 5.3 5.5 3.8 4.1 4.1
Y4 1.043 1.016 1.044 1.038 1.061 1.043 1.056 1.058 1.036 1.042 1.042
Y5 7.85 8.2 8.83 8.87 8.3 7.9 9.05 8.5 8.3 8.37 8.4
Y6 1.165 2.264 0.700 1.205 1.914 2
Data Loading...
