Abstract Gene clustering, the process of grouping related genes in the same cluster, is at the foundation of different genomic studies that aim at analyzing the function of genes. Microarray technologies have made it possible to measure gene expression levels for thousands of genes simultaneously. For knowledge to be extracted from the datasets generated by these technologies, the datasets have to be presented to a scientist in a meaningful way. Gene clustering methods serve this purpose. In this paper, a hybrid clustering approach that is based on self‐organizing maps and particle swarm optimization is proposed. In the proposed algorithm, the rate of convergence is improved by adding a conscience factor to the self‐organizing maps algorithm. The robustness of the result is measured by using a resampling technique. The algorithm is implemented on a cluster of workstations. Copyright © 2004 John Wiley & Sons, Ltd.

In the past decade chemical and biological laboratory experiments have generated an explosive amount of data. As a result, a set applications that manipulate these dynamic, heterogeneous and massive amounts of data have emerged. An example of such applications in the pharmaceutical industry is the computational process involved in the early drug discovery of lead drug candidates for a given target disease. The discovery of lead drug candidates requires both consecutive and random data access to the pharmaceutical drug candidate database. This paper focuses on performance enhancement techniques for the pharmaceutical drug candidate database application. In particular, this paper compares static load balancing and dynamic load balancing in the context of the drug candidate database application. This database application is based on multi-queries. Some of these queries are multi-join queries.

Large scientific databases often consist of heterogeneous components that differ in the number of attributes and the number of tuples. In this paper, we show that increased performance is achieved when a complex schema is used to represent different components of a data repository. The driving application used in this study is a pharmaceutical drug repository. The types of access patterns associated with this repository consist of the retrieval of a selected set of attribute values for a consecutive or a random set of tuples. These consecutive and random access patterns, which are typical of a drug discovery experiment, are used to measure the data retrieval throughput for different schemas.

An explosion in the amount of data generated through chemical and biological experimentation has been observed in recent years. This rapid proliferation of vast amounts of data has led to a set of cheminformatics and bioinformatics applications that manipulate dynamic, heterogeneous, and massive data. An example of such applications in the pharmaceutical industry is the computational process involved in the early discovery of lead drug candidates for a given target disease. This computational process includes repeated sequential and random accesses to a drug candidate database. Using the above pharmaceutical application, an experimental study was conducted in this paper that shows that for optimal performance, the degree of parallelism exploited in the application should be adjusted according to the drug candidate database instance size and the machine size. Additionally, different degrees of parallelism should be used depending on whether the access to the drug candidate database is random or sequential.