KEGG is a manually curated database integrating eighteen databases categorized into systems, genomic, chemical, and health information. It provides KEGG mapping tools to understand cellular and organism-level functions from genome sequences and molecular datasets. KEGG mapping reconstructs molecular network systems based on functional orthologs. Since the introduction of the KEGG NETWORK database, various diseases have been associated with network variants, which are perturbed molecular networks caused by human gene variants, viruses, other pathogens, and environmental factors. Network variation maps are created as aligned sets of related networks showing how different viruses inhibit or activate specific cellular signaling pathways. The KEGG pathway maps are now integrated with network variation maps in the NETWORK database, as well as with conserved functional units of KEGG modules and reaction modules in the MODULE database. The KO database for functional orthologs continues to be improved and virus KOs are being expanded for better understanding of virus-cell interactions and for enabling prediction of viral perturbations. KEGG is an integrated database consisting of eighteen original databases in four categories. The databases in the health information category together with two outside databases of drug labels are collectively called KEGG MEDICUS. The original databases are all manually curated except the computationally generated SSDB database. The content covers wide-ranging biological objects, including genes and proteins, chemical substances and reactions, molecular interaction/reaction/relation networks, and human diseases and drugs. Each biological object is given a unique identifier. One important principle of organizing biological objects in the KEGG database is the distinction of reference data (classes) and variation data (instances). The NETWORK database is a human-specific database consisting of network elements, simply called networks, defined as functionally meaningful segments of signaling and other pathways. There are three types of networks: reference networks, disease-related variant (perturbed) networks and drug-target relations. Variant networks are further divided into three types of perturbants: human gene variants, pathogens and environmental factors. The NETWORK database may also be viewed as a collection of network variation maps identified by nt numbers, displaying aligned sets of both reference and variant networks. The KEGG NETWORK database currently contains disease-associated variant networks for cancers, neurodegenerative diseases, endocrine and metabolic diseases including inborn errors of metabolism, viral infections and some bacterial infections. The most recent addition consists of six neurodegenerative diseases, Alzheimer disease, Parkinson disease, amyotrophic lateral sclerosis, Huntington disease, spinocerebellar ataxia and prion disease, for which variant networks have been defined from improved pathway maps. Common features of neurodegeneration are found to be linked to molecular networks, including accumulation of abnormal protein aggregates, impairment of ubiquitin-proteasome system, endoplasmic reticulum stress, autophagy impairment, mitochondrial dysfunction and axonal transport defect. The network variation map nt06410 for calcium signaling is the most characteristic signaling pathway involved in many ofKEGG is a manually curated database integrating eighteen databases categorized into systems, genomic, chemical, and health information. It provides KEGG mapping tools to understand cellular and organism-level functions from genome sequences and molecular datasets. KEGG mapping reconstructs molecular network systems based on functional orthologs. Since the introduction of the KEGG NETWORK database, various diseases have been associated with network variants, which are perturbed molecular networks caused by human gene variants, viruses, other pathogens, and environmental factors. Network variation maps are created as aligned sets of related networks showing how different viruses inhibit or activate specific cellular signaling pathways. The KEGG pathway maps are now integrated with network variation maps in the NETWORK database, as well as with conserved functional units of KEGG modules and reaction modules in the MODULE database. The KO database for functional orthologs continues to be improved and virus KOs are being expanded for better understanding of virus-cell interactions and for enabling prediction of viral perturbations. KEGG is an integrated database consisting of eighteen original databases in four categories. The databases in the health information category together with two outside databases of drug labels are collectively called KEGG MEDICUS. The original databases are all manually curated except the computationally generated SSDB database. The content covers wide-ranging biological objects, including genes and proteins, chemical substances and reactions, molecular interaction/reaction/relation networks, and human diseases and drugs. Each biological object is given a unique identifier. One important principle of organizing biological objects in the KEGG database is the distinction of reference data (classes) and variation data (instances). The NETWORK database is a human-specific database consisting of network elements, simply called networks, defined as functionally meaningful segments of signaling and other pathways. There are three types of networks: reference networks, disease-related variant (perturbed) networks and drug-target relations. Variant networks are further divided into three types of perturbants: human gene variants, pathogens and environmental factors. The NETWORK database may also be viewed as a collection of network variation maps identified by nt numbers, displaying aligned sets of both reference and variant networks. The KEGG NETWORK database currently contains disease-associated variant networks for cancers, neurodegenerative diseases, endocrine and metabolic diseases including inborn errors of metabolism, viral infections and some bacterial infections. The most recent addition consists of six neurodegenerative diseases, Alzheimer disease, Parkinson disease, amyotrophic lateral sclerosis, Huntington disease, spinocerebellar ataxia and prion disease, for which variant networks have been defined from improved pathway maps. Common features of neurodegeneration are found to be linked to molecular networks, including accumulation of abnormal protein aggregates, impairment of ubiquitin-proteasome system, endoplasmic reticulum stress, autophagy impairment, mitochondrial dysfunction and axonal transport defect. The network variation map nt06410 for calcium signaling is the most characteristic signaling pathway involved in many of