Agent interaction for bioinformatics data management

Abstract

As genome projects produce increasingly large quantities of sequence data, fast and reliable sequence analysis methods are required. Basic methods for comparing pairs of sequences or detecting patterns are well-developed, and now the key problem in analyzing this genomic data is how to integrate the software and primary databases in a flexible and robust way. The wide range of available programs conform to very different input, output, and processing requirements, typically with little consideration given to issues of integration. Key to addressing these issues appropriately is not to consider them as a result of the biological domain, but instead as an information processing problem that suggests nothing as much as an agentbased approach. In this paper, we introduce GeneWeaver, a multi-agent system for bioinformatics, and describe in detail the agent interactions which allow the integration and management of analysis methods and data. The system does not offer new methods but instead manages existing databases and analysis tools in an effective and flexible way, and facilitates easy and dynamic growth.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 20 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.