An in vivo database model for pharmacological and physiological dosage for experimental animals

https://doi.org/10.1016/j.compbiomed.2009.04.001Get rights and content

Abstract

We developed a database compiling in vivo doses of compounds for various activities in certain animal species. The related database covers almost 100 years of experiments. The conceptual scheme of the database was created using concepts of the entity-relationship modeling principles. Using published references, dosages and their effects on laboratory animals were entered as data. As the next stage of our work, we have started to examine the available literature to information about the experimental dosages of various drugs used in other studies.

The database provides various interfaces, including graphical-user interfaces and interfaces for Internet access. The database will be useful as a knowledge infrastructure for researchers who have to perform dose-scan experiments for a specific pharmacological activity. The basic benefit of that knowledge infrastructure is that it will enable virtual pharmacological experiments that will be considerably less expensive than conventional laboratory experiments, because the number of animals used and the number of dose-scan experiments will be greatly reduced. The developed database will also be helpful for new drug development studies and for responding to queries about animal types, drugs used, drug interactions, and results.

Introduction

Scientific studies produce data that are meant to help us understand mechanisms, causes, and effects, and all such studies have certain associated costs. Most such data are not used immediately, but the results are generally stored or published and used later for additional research. Data are important, and they should be updated to improve their accuracy and reliability, and they should be easily retrievable when needed. In the experimental sciences and in applied research, there are usually complicating factors that affect the accessibility and accuracy of the data, e.g., it may require a long time to obtain the desired data, the experimental process is time-consuming, the costs of materials to conduct the research and data acquisition may be high, and there may be a dearth of data in the area of interest to the researchers. In addition, data may be unprocessed, and they must be manipulated and analyzed to produce useful information and understanding. Sometimes it is necessary to perform extensive processing of the available data to produce defensible results and meaningful information or knowledge. While information ages rapidly, knowledge has a longer life-span [1]. From this point of view, the conversion of data to knowledge is one of the basic goals of all research and innovative undertakings. But, this conversion is not an easy process. At this point, computers can serve as a tool to facilitate the process. When we consider the huge amount of data, we must ensure that there are database applications that can store, maintain, distribute, and share the data [2], [3], [4]. But the most important advantage of using an appropriately designed database application for large amounts of data that new information and knowledge can be extracted from the database without the need for conducting additional experiments. The key to the success of these kinds of databases is the robust data models on which they are based [5].

Pharmacology education and research cover a wide range of experiments. One such experiment is the study of in vivo medicine dosage applications for different experimental animals. The animals are subjected to varying dosages of many different medicines. Of course, these experiments sometimes produce conflicting results that are difficult to interpret. In the experiments, many medicines, different dosage levels, different animals, and various exposure routes are studied to determine the effects on the animals. The basic exposure data obtained in this manner should be registered and published so that others will know what has been done so as to avoid duplication of efforts. Using databases in new drug research and investigations is very important, and this importance will increase in the future [6]. Also, to enhance and augment classical laboratory research, many computer-aided applications will be developed for the field of pharmacological research [7].

Although some database applications already exist for biomedical and pharmacological research [8], [9], [10], [11], [12], [13], they are addressed on the Internet simply as a list of references, and none of them is related specifically to a database for drug dosage in pharmacological experiments. Some of these databases were created to address ethical, legal, security, and computational issues [14]. Registered and published pharmacological experiments [15], [16] are too much and almost 100 years old. When we design an appropriate database for those experimental data, it is possible to get information or knowledge about all experiments. Also, by using a database, it is possible to share and maintain these data for all relevant educational institutions and other pharmacological institutions over the world. In this way, extensive savings can be realized in pharmacological experiments when the data produced are added to a database. To accomplish this purpose, the Department of Pharmacology and the Department of Computer Engineering at Anadolu University in Eskisehir, Turkey, undertook a cooperative study and developed a database of pharmacological experiments [17]. The database includes dosage data used with experimental animals for the last 100 years. This study was completed in 14 months. The results of this interdisciplinary study will be addressed in the following sections of this article.

Section snippets

Methods

The methods used in this study to design the database are ordinary in related literature. The design pattern of the database was selected to be relational, because when we searched for data from pharmacological and physiological dosage of experimental animals, it was apparent that there was a very high correlation between the data that were used and recorded.

We applied three-layer architecture to design the relational database [4]. In the first layer, we examined all the data about

Results

When we finished the data entry for the pharmacological and physiological dosage for experimental animals’ database, the final data in the related tables were:

  • Drugs table: It holds 212 records. As an example, if we make a query to the database, like: “select * from Drugs where Drug Code like ‘1.1.’+‘%’”. Some results of this query are shown in Table 1.

  • Alter_Drug table: It holds 201 records. As an example, if we make a query to the database, like: “select * from Alter_Drug where Drug_Code like

Discussion and conclusions

Tables of the developed database contain experimental data. Those data represent various facts about pharmacology and experimental animals. Also those data are an infrastructure for much related information or knowledge. When those data are transformed into information or knowledge, there will be an opportunity for innovation that is the main purpose of research and experimentation. Transformation processes are applied to the data, since transformation is unavoidable when the amount of data is

Conflict of interest statement

None declared.

Ali Gunes was born in Kayseri, Turkey, in 1952. He received a degree from the University of Middle East Technical University and Ph.D. from Anadolu University, in 1982. Since 1982, he has been with the Anadolu University, where he is Professor in Department of Computer Engineering. His research interests focus on software development and biomedical informatics.

References (18)

  • R.L. Ackoff

    From data to wisdom

    Journal of Applied Systems Analysis

    (1989)
  • R. Ramakrishnan

    Database Management Systems

    (1998)
  • P. Alzemi et al.

    Database Systems

    (1999)
  • R. Elwasri et al.

    Fundamentals of Database Systems

    (2000)
  • T.J.M. Bench-Capon, Why Database and Expert Systems?, in: Eighth International Workshop on Database and Expert Systems...
  • E.H. Ohlstein et al.

    Drug discovery in the next millennium

    Annual Review of Pharmacology and Toxicology

    (2000)
  • W.G. Richards

    Virtual screening using grid computing: the screensaver project

    Nature Reviews Drug Discovery

    (2002)
  • Merck, 〈http://www.merck.com〉, Accessed on...
  • The GDB Human Genome Database, 〈http://gdbwww.gdb.org〉, Accessed on...
There are more references available in the full text version of this article.

Cited by (0)

Ali Gunes was born in Kayseri, Turkey, in 1952. He received a degree from the University of Middle East Technical University and Ph.D. from Anadolu University, in 1982. Since 1982, he has been with the Anadolu University, where he is Professor in Department of Computer Engineering. His research interests focus on software development and biomedical informatics.

Yusuf Ozturk was born in Ankara, Turkey, in 1958. He received a degree from the Department of Pharmacology of University of Ankara and Ph.D. from Ankara University, in 1985. Since 1985, he has been with the Anadolu University, where he is Professor in Department of Pharmacology. His research interests focus on diabetes mellitus, diabetic complications and drug impacts.

Ahmad Babanli was born in Baku, Azerbaijan, in 1941. He received a degree from the University of Petroleum and Chemical, Azerbaijan and Ph.D. from Moscow Power University, in 1970. Since 2000, he has been with the Anadolu University, where he is Associate Professor in Department of Computer Engineering. His research interests focus on database design and biomedical informatics.

View full text