The KEGG database has been developed by focusing on conservation and variation of genes and genomes among different organisms. The reference datasets of KEGG pathway maps, BRITE hierarchies and KEGG modules have been developed with the concept of functional orthologs (KOs), so that any genome can be mapped to these datasets for biological interpretation of higher-level functions. However, this approach is inadequate for understanding more detailed features caused by variation of genes and genomes within a species, especially for linking variation of human genes and genomes to pathogenicity and drug responses.


KEGG NETWORK represents a renewed attempt by KEGG to capture knowledge on diseases and drugs in terms of perturbed molecular networks (see Background of KEGG DISEASE). It accumulates variations of molecular interaction/reaction networks in Homo sapiens distinguishing gene variants and other perturbants. Entries of KEGG NETWORK are called "network elements" and identified by N numbers. Reference network elements are those that are represented in KEGG pathway maps and perturbed network elements are variations caused by gene variants, viruses, drugs, etc. See, for example, network variation map nt06210, where reference network elements are shown in green and gene variants are marked in red.

The first release of the KEGG NETWORK database contains only the cancer related network elements. For example, N00002 is a variation of MAPK (ERK) signaling network caused by BCR-ABL fusion kinase, which is related to chronic myeloid leukemia. The network element is defined by the sequence of human gene IDs including variants designated by 25v1, etc.

Network Variation Maps

The network variation map is computationally drawn for a collection of related network elements, either in terms of signaling pathways or cancer types.

06200   Cancer Network
Cancer Types

See Color codes of cancer types

Cancer Network Viewer

Signaling networks shown above are mapped to the global map of signaling pathways in cancers, Pathways in Cancer, by a new tool tentatively called Cancer Network Viewer. It allows viewing of the relationships from gene variants to network elements to signaling pathways and then to the hallmarks of cancers.
Launch cancer network viewer
This tool will be enhanced to accept users' data, such as for mapping of gene variants identified in personal genome sequence data.

Last updated: December 15, 2017
KEGG GenomeNet Kanehisa Laboratories