Pathagent: Multi-agent system for updated pathway information integration

Abstract

Nowadays, integration of biological information coming from multiple sources is a common task for the functional interpretation of experimental results during high-throughput data analysis. In concrete, metabolic pathways information is one of the most widely used genome annotation, during post-hoc analyses of differential gene expression experiments. This paper presents a multi-agent system able to retrieve and integrate metabolic functional annotations coming from multiple available databases. Taking advantage of the multi agent architecture design, the developed tool is flexible, scalable and extensible to adopt more databases. Interoperability with external systems is achieved through a freely available RESTful API.