KEGG Release Notes

Future releases

• All KEGG pathway maps redrawn by an internally developed tool called KegSketch.

Current release

• A new version of KEGG Atlas viewer is released using the <canvas> element.
• Subcategorization of KEGG DRUG entries is introduced distinguishing mixtures, crude drugs, and TCM (Traditional Chinese Medicine) formulas.

Past releases

• The KEGG Atlas cancer map is released combining 14 existing cancer pathway maps.
• The new GECS (gene expression to chemical structure) software is released, currently, for prediction of N-glycan structures from microarray gene expression profiles.

• The global map of KEGG metabolic pathways is released as the first component of KEGG Atlas.
• Smaller units of existing KEGG pathways are now stored in the KEGG MODULE database, which also contains other functional units of pathways and complexes.
• The KEGG DISEASE database is released containing gene/molecule lists for selected diseases.
• A new page, KEGG COMPOUND, is prepared for pathway-based reorganization of biochemical compounds.

• KEGG EGENES, which is constructed from massive EST libraries, now incorporates singletons as well as contigs.
• The Organism subfield is introduced in KEGG ENZYME for the organism information manually taken from the original references.
• The Structure field is introduced in KEGG GENES containing cross-reference links from GENES to PDB refined by the corresponding organism information.

• Java Web Start is introduced for launching KegTools: BRITE to KegHier, COMPOUND/GLYCAN to KegDraw, and EXPRESSION to KegArray.
• Peptide entries in the COMPOUND database now contain the SEQUENCE field, similar to PK/NRP entries.

• New version of the hierarchical text file viewer "get_htext" is released.
• PK/NRP entries are added to the COMPOUND database with the new SEQUENCE field.
• Server-side viewer is discontinued for the EXPRESSION database, which is now linked to KegArray.
• New directory organization of the FTP site.

• New search box is introduced in the KEGG top page.

• The entire text of all BRITE files can now be searched with keywords.

• Reactions subcategory in the BRITE database.
• RPAIR database updated to reflect most of the KEGG reactions.

• The Japanese language interface to the KEGG system is released.
• KEGG update notes in RSS feeds.

• The mapping to BRITE hierarchies is implemented as an operation to combine two files: a binary relation file and a hierarchical text file.
• Ortholog group tables to be discontinued.
• KEGG pathway modules released.
• New top page for KEGG SSDB.

• DBGET implementation of KEGG BRITE.
• Standalone java application KegHier for accessing and handling BRITE files.

• Automatic conversion of outside database identifiers is implemented. Gene/protein identifiers such as in NCBI RefSeq (Entrez Gene) and UniProt can now be directly used in KEGG GENES.

• A new category of the KEGG PATHWAY database "Drug Development" is made available. It is a collection of KEGG DRUG structure maps illustrating our knowledge on drug development.
• Manually drawn glycan structures are included in the Glycan section of the KEGG PATHWAY database to supplement computationally generated Composite Structure Map (CSM).

• The LIGAND relational database (primary) version of KEGG DRUG is made available in addition to the DBGET (secondary) version.
• The KEGG REACTION data is enhanced with the introduction of the categorization of chemical structure transformation patterns, called RDM patterns, in the RPAIR database.
• A new interface is introduced to DRUG, GLYCAN, REACTION, EXPRESSION, and Auto Annotation (KAAS).

• KEGG automatic annotation server, named KAAS, is released to assist annotation (KO assignment) and KEGG pathway mapping of newly determined genomes.
• Cross-reference lists are made available for conversion of IDs between KEGG and outside databases.
• KEGG DRUG is introduced separating from KEGG COMPOUND to accumulate chemical structures, therapeutic categories, targets, and other information of drugs.

• Different entry points are prepared for accessing KEGG, called Generalized KEGG, Specialized KEGG, and Personalized KEGG in the KEGG2 (table of contents) page.
• One feature of Specialized KEGG is an organism-specific view of KEGG.
• Another feature is customization of KEGG organism menu used to display pathway maps, etc. Cookie must be enabled to make this feature to work.
• By making use of the LinkDB resource in GenomeNet, KEGG pathways are linked to many different outside databases, such as NCBI Entrez and UniProt.
• Automatic annotation and pathway mapping of draft genomes (DGENES) and EST consensus contigs (EGENES) are released, which significantly increased the number of KEGG Organisms.
• KEGG BRITE is officially introduced as the fourth component of KEGG, providing among others classifications of drugs and compounds.
• KegDraw for drowing chemical compound and glycan structures is released.

• A new glycomics resource "KEGG GLYCAN Composite Structure Map" is released. Genes in the genome or the transcriptome can be mapped on a composite structure generated from the GLYCAN database so that organism-specific or expressed glycan structures can be viewed and examined.
• A high-level KEGG pathway map is introduced for an overview of biosynthetic pathways. Such high-level maps will be used in the future for computational inference in the BRITE database.
• A therapeutic category of drugs in the COMPOUND database is released based on the Japanese Pharmacopoeia.
• A preliminary version of EGENES, a collection of plant EST contigs, is released. Currently it contains no annotation, but in a future release KEGG pathways will be generated by automatic KO assignments.

• The Oracle version of the LIGAND database is released replacing the ISIS version.
• New in-house programs SIMCOMP and SUBCOMP can now be used for searching similar compound structures against the COMPOUND or REACTION database.
• Another new program e-zyme has been developed to predict reactions between given compound pairs with assignments of possible EC numbers.
• The RPAIR database is released as part of LIGAND, containing the reactant pairs that constitute enzymatic reactions with information about atomic alignments.

• The DBGET entries of the KEGG databases are shown in a nicer view with buttons to link other KEGG resources and help messages.
• The GFIT tables that have been used for genome annotations are incorporated into the SSDB database (Oracle), which makes the gene cluster search much faster than before.

• KGML v0.4 contains a major restructuring of elements.
• A preliminary version of KegArray, a Java application for microarray data analysis, is released for Mac OS X.
• The KEGG pathways are linked again to human diseases genes in the OMIM database.

• New user interfaces are introduced to reflect the basic structure of KEGG. Conceputually, KEGG consists of three graph objects called the protein network (PATHWAY database), the gene universe (GENES/SSDB database), and the chemical universe (LIGAND database).
• The Java applet version of KEGG genome map browser is discontinued. It is replaced by a new version using dynamically generated image maps.
• The Java applet version of KEGG expression browser will also be discontinued.
• The automatic generation of ortholog group tables incorporates coloring of gene clusters.

• KGML v0.3 comes with part of KEGG regulatory pathways.

• The GLYCAN database is released as a new effort to collect complex carbohydrate structures. GLYCAN is part of the DBGET system and the structure search is available together with the Java-based input tool.
• CarbBank is also implemented in the DBGET system.

• The first release of KEGG API is made avilable. It is a SOAP interface suitable for automating access to and use of KEGG.
• KGML (KEGG Markup Language) v0.2 is available with reaction data in metabolic pathways.
• The results of computationally generated ortholog and paralog groups in SSDB can now be viewed by the adjacency matrix viewer.

• An XML version of the KEGG metabolic pathways is made available in the KGML (KEGG Markup Language) v0.1.
• KO (KEGG Orthology) is officially released in the following: KO database in DBGET, KO entries linked to SSDB clustering, and 'all organisms' pathways linked to KO.
• The reference metabolic pathways are now of two types: those linked to enzymes (EC numbers) and those linked to KEGG reactions (R numbers).

• The GENES and SSDB databases are operational on Oracle.
• SSDB now contains TIGRFAMs in addition to Pfam and Prosite.

• REACTION database is made available in DBGET.

• The fifth top category of "Human Diseases" is introduced in the hierarchy of KEGG pathways.
• The top category of "Cellular Processes" is also being reorganized.
• The hierarchy of KEGG pathways is now called KO (KEGG Orthology). The KO identifiers will be introduced shortly.

• Automatic generation of ortholog group tables using the "select organism" option.

• The Chemscape version of the LIGAND database is released, enabling substructure searches against COMPOUND and REACTION databases.

• The Java-based dendrogram viewer is implemented in both the Expression database and the SSDB database.

• The KEGG Expression Database contains an increasig amount of microarray gene expression profile data for Synechocystis PCC6803 and Bacillus subtilis, together with an enhanced version of KEGG expression browser.
• The initial phase of SSEARCH computation for all pairwise genome comparisons has been completed in the SSDB Database. The result of cluster analysis can now be examined with a Java dendrogram viewer.

• A new database named SSDB is implemented. It contains similarity scores of all gene pairs in the GENES database computed by SSEARCH at the amino acid sequence level. It also contains the information about best hits and double-best hits when two genomes are compared.

• A hierarchy of organism groups in the KEGG menu page and associated DBGET search.
• A hierarchy of genes in the list of ortholog identifiers, which is used for automatic matching of genes in the genome and gene products in the pathway.
• The database name GENOMES is changed to GENOME, which now represents both the information about individual organisms and the collection of Java genome maps.

• The cell lineage of H.roretzi is now represented by a graphical image map, which is linked to actual pictures of early-stage embryo.
• Different types of analysis tools for microarray expression data are being integrated into a single interface of KEGG expression database.

• The gene catalog search is incorporated in the DBGET/LinkDB system as the hfind mode.
• A new version of expression browser is implemented.
• The MOTIF search can now be automatically invoked from any GENES database entry.
• The KEGG metabolic pathway section now includes pathways of the University of Minnesota Biocatalysis/Biodegradation Database (UM-BBD).

• The ortholog IDs are introduced as an extension of the EC numbers for matching the pathway information with the genomic information, such as for automatic generation of organism-specific pathways.
• As a result of this extension, KEGG metabolic pathways may now contain gene names.
• The get_htext program to handle hierarchical text data, such as gene catalogs, has been upgraded. It now runs ten times faster and supports more options.

• Introduced the GENOMES database for a collection of completely sequenced genomes.
• Enhancements of the expression map browsers including the new tool.

• A new type of hierarchical text data is introduced for browsing cell lineage maps.
• FASTA search against GENES database can now be performed both at the amino acid and nucleotide sequence levels.

• A preliminary version of expression map browser is made available for analyzing microarray gene expression data.
• The table of contents page is reorganized.
• The regulatory pathways page is reorganized.

• In the ortholog group tables, the column-wise extraction of gene definitions is now possible in addition to the row-wise extraction.
• An entry in the GENES database can now be retrieved in the DBGET/LinkDB system either by the gene accession (such as b1882 or YAL041W) or by the primary gene name (such as cheY or CDC24).

• A new version of genome map browser has been implemented. The browser now has two windows: an HTML window and a Java window. The HTML window contains the overall map in Java, which integrates both linear and circular chromosomes, and also incorporates plasmids and extra chromosomal elements. The Java window is an enlarged window around the position specified in the overall map, which now displays five lines by default.

• It has become possible to color genome maps and ortholog group tables, in addition to pathway maps, for distinguishing different groups of genes such as in the gene expression profile experiments.

• An effort to classify chemical compounds is initated.
• The objects in the KEGG pathways can now be marked by any combination of colors, which should be useful in coloring genes according to the expression profiles obtained by microarray or DNA chip experiments.

• The KEGG mirror server package has been released.
• The genome map browser has a new interface that runs much faster than the previous version.

• A new capability to compare two genome maps has been implemented. It is a 'dot matrix' of similar genes that can be used to identify clusters of genes that share sequence similarities.
• The GENES amino acid sequences and the whole GENOME nucleotide sequences can now be searched by both FASTA and BLAST either from the KEGG table of contents page or the KEGG tools page.

• A new category of ortholog/paralog group tables is added to the KEGG pathway section.
• A prediction tool is made available for gene function assignment by searching sequence similarities and by making use of the information of a functional unit in the ortholog/paralog tables.
• Started providing a suite of genome map comparison tools, including the search for homologous gene clusters.

• Genome map search - A new capability to search genome maps for gene positions is released.
• Pathway map search - The capabilities to search pathway maps for enzymes, compounds, and genes are combined in a single user interface under the WWW version, which is now consistent with the e-mail version.
• In the result of path computation program PathComp, the user can now examine a series of compound structures to select plausible pathways.

• Started organizing molecular assemblies which are tentatively classified under the regulatory pathway section.
• Started organizing different types of molecular catalogs, especially for human proteins.

• The compound data has been fully integrated. A circle on the pathway diagram is now a clickable and searchable object.
• All image maps are now handled at client side, which enables easy recognition of clickable area by the cursor.
• The gene catalog for each organism now contains the classification of regulatory pathways.
• The pathway search routines by EC numbers, compound numbers, and gene names can now be used as an e-mail server. The address is: pathway@genome.ad.jp.

• A set of gene names of a given organism can be used as a query to search matching names on the KEGG pathways.
• A set of amino acid sequences can be used as a query to search similar sequences on the KEGG pathways, which will be especially useful in pathway reconstruction.

• KEGG Release 3.1 contains a new collection of regulatory pathways, in addition to the previous collection of metabolic pathways.
• The KEGG TOC page is non-framed.

• This is the first official CD printing. The previous ones were made in-house.

• The KEGG TOC page is framed.
• New Java launchers are installed.

• The Internet version relies less on Java making it more widely available.
• The CD version experiments a number of advanced features of Java and Javascript which may only be appreicated by Netscape 3.01 or later.

• The weekly updated CD version of KEGG may be downloaded by anonymous FTP.

• The Java option of KEGG is now fully implemented.

• The first official release of KEGG contains five organisms, Haemophilus, Mycoplasma, Methanococcus, Synechocystis, and Yeast.
• A new program, ec_point, is released for searching enzymes on the metabolic pathways.
• Genome browsers, ZoomDial for circular genomes and ZoomLine for linear genomes, are released.

• The clickable chromosome maps are available for accessing GDB and OMIM data.
• The enzyme classification according to PROSITE patterns is added.
• The metabolic pathway data are being verified with different sources and modified accordingly.

• The pathway data for most of the known metabolic pathways have been entered and properly classified. They may now be accessed through a clickable diagram.
• The genes tables are available for 13 species and human disease genes.
• It is now possible to identify enzymes that appear in OMIM, PDB, and PIR databases. Click the title box in each pathway diagram.
• To assist enzyme search against the pathway diagrams, a new window is added for EC number search.

• A bare backbone of KEGG is available consisting of gene classification tables for selected organisms, three types of enzyme classification tables, and a selected number of metabolic pathway diagrams.
• The basic capabilities of hierarchical text handling and graphics diagram handling are available.
• The search capabilities include key word search on a hierarchical text and enzyme (boxed object) search on pathway diagrams.
• A repertoire of enzyme genes in each category may be identified on the corresponding pathway diagram. Click the link marked MAP in the gene classification table.