This paper reports on new developments in KEGG, a database resource for understanding higher-order functions and utilities of the cell or the organism from its genomic information. KEGG is already a highly popular database—the KEGG website is used by 20,000 visitors per day—and this paper presents new concepts and new bioinformatics methods to integrate genomic and chemical information and also to analyze interactions with the environment.

“KEGG is a computer representation of the biological system, such as the cell and the organism, accumulating and integrating knowledge at the molecular, cellular, and organism levels.”

KEGG is a computer representation of the biological system, such as the cell and the organism, accumulating and integrating knowledge at the molecular, cellular, and organism levels. This virtual cell/organism consists of molecular building blocks and wiring diagrams, which can be used for modeling and simulation as well as for browsing and retrieval.

Originally, the wiring-diagrams involved endogenous molecules, both those which are directly encoded in the genome—proteins and RNAs—and those that are indirectly encoded through biosynthesis, biodegradation, and transport pathways—metabolites, glycans, etc.

In this paper we have extended these wiring diagrams to include exogenous molecules, such as those found by chemical genomics experiments. This will help in understanding the interactions between the biological system and the natural environment and would eventually lead to a representation and reconstruction of another higher-level biological system, the biological world.

The Human Genome Project determined the entire DNA sequence of the human genome, a blueprint of life. In reality, however, it is not possible to infer, for example, human behaviors and diseases simply from DNA sequence information alone.

The cell and the organism are biological systems consisting of various molecular building blocks with elaborate wiring-diagrams. A disease should be considered as a perturbed state of the biological system, and a drug is a perturbant bringing the biological system back to the stable state.

KEGG has enabled in silico analysis of the cell and the organism, and this paper presents a conceptual framework and practical methods for integrating environmental information, such as for integrated analysis of both genetic and environmental factors of common diseases.

The KEGG project was initiated in 1995, the last year of the first five-year phase—the mapping phase rather than the sequencing phase—of the Japanese Human Genome Project. I was responsible for the informatics portion of the Japanese HGP and foresaw the needs for new informatics technologies to decipher the genome when it would be sequenced.

Obviously, computer algorithms alone would not solve this problem, and a knowledge-based approach was necessary. Thus, the KEGG resource was started as a reference knowledge base linking genomes to biological systems, and now to the environment as well.