• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of narLink to Publisher's site
Nucleic Acids Res. Jan 2007; 35(Database issue): D237–D240.
Published online Nov 29, 2006. doi:  10.1093/nar/gkl951
PMCID: PMC1751546

CDD: a conserved domain database for interactive domain family analysis

Abstract

The conserved domain database (CDD) is part of NCBI's Entrez database system and serves as a primary resource for the annotation of conserved domain footprints on protein sequences in Entrez. Entrez's global query interface can be accessed at http://www.ncbi.nlm.nih.gov/Entrez and will search CDD and many other databases. Domain annotation for proteins in Entrez has been pre-computed and is readily available in the form of ‘Conserved Domain’ links. Novel protein sequences can be scanned against CDD using the CD-Search service; this service searches databases of CDD-derived profile models with protein sequence queries using BLAST heuristics, at http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi. Protein query sequences submitted to NCBI's protein BLAST search service are scanned for conserved domain signatures by default. The CDD collection contains models imported from Pfam, SMART and COG, as well as domain models curated at NCBI. NCBI curated models are organized into hierarchies of domains related by common descent. Here we report on the status of the curation effort and present a novel helper application, CDTree, which enables users of the CDD resource to examine curated hierarchies. More importantly, CDD and CDTree used in concert, serve as a powerful tool in protein classification, as they allow users to analyze protein sequences in the context of domain family hierarchies.

INTRODUCTION

The annotation of conserved domain footprints on protein sequences often serves as the first step toward characterizing protein function in silico. Protein domains may be viewed as units in the molecular evolution of proteins and can be organized into an evolutionary classification. The set of protein domains characterized so far appears to describe no more than a few thousand superfamilies, where members of each superfamily are related to each other by common descent. NCBI's conserved domain database (CDD) attempts to collate that set and to organize related domain models in a hierarchical fashion, meant to reflect major ancient gene duplication events and subsequent functional diversification.

Computational annotation of protein function is generally obtained via sequence similarity: once a close neighbor with known function has been identified, its annotation is copied to the sequence with unknown function. This strategy may work very well in functionally homogeneous families and when applied only for very close neighbors or suspected orthologs, but it is doomed to fail often when domain or protein families are sufficiently diverse and when no close neighbors with known function are available.

To this end, the CDD (1) provides a strategy toward a more accurate assessment of such neighbor relationships, similar to approaches termed ‘phylogenomic inference’ (2). CDD acknowledges that protein domain families may be very diverse and that they may contain sets of related subfamilies. Of these, only few may have been characterized experimentally, and within this set function may have diverged considerably. While it may be possible, and certainly efficient, to represent such a set of subfamilies with just a single family model, that model could only provide very generic annotation. In CDD curation, we attempt to detect evidence for duplication and functional divergence in domain families by means of phylogenetic analysis. We record the resulting subfamily structure as a set of explicit models, but limit the analysis to ancient duplication events—several hundred million years in the past, as judged by the taxonomic distribution of protein sequences with particular domain subfamily footprints.

CDD provides a search tool employing reverse position-specific BLAST (RPS–BLAST), where query sequences are compared to databases of position-specific score matrices (PSSMs), and E-values are obtained in much the same way as in the widely used PSI-BLAST application (3). When CDD is scanned with protein query sequences, a region on a query may pick up more than one overlapping footprint from a set of related models. One of those models provides the best score or lowest E-value, but that alone may not be sufficient to indicate that the query sequence is a bona fide member of the corresponding subfamily. Since the CDD collection also contains imported models, which have not been curated at NCBI, search results may present a mixture of curated models (accessions starting with ‘cd..’) and un-curated models (accessions starting with ‘pfam’, ‘smart’ or ‘COG’). By default, overlapping domain hits are sorted by E-value, but curated models are listed first, if their E-values exceed a secondary significance threshold of 1e-05. Default displays are presented in a concise fashion, where domain hits that overlap with the top-ranked domain hits are hidden.

We have started to distribute CDTree, a helper application for the web browser. CDTree allows users to examine the results of simple phylogenetic analysis on the sequences from a curated domain hierarchy, and view their query sequence in the context of such a phylogenetic sequence tree.

ASSESSING DOMAIN FAMILY MEMBERSHIP

Figures 1 and and22 demonstrate how one might use a web browser, the curated CDD resource and the CDTree application to obtain confidence for the transfer of annotation from a domain model to a particular protein sequence. A cartoon depicting domain model footprints on a protein sequence can be obtained by following the ‘Conserved Domains’ link from an Entrez protein search results page, or by submitting a live CD-search (4) request (Figure 1a). A particular domain annotation is examined in detail by clicking on the corresponding item in the graphical display (Figure 1b). This generates a CD summary page (Figure 1c) which lists descriptive information and a multiple sequence alignment including the user query sequence (not shown). The summary page contains a description of the particular family. If the domain model has been curated at NCBI (CDD accessions starting with ‘cd..’), it also lists conserved features that have been recorded by NCBI curators, displays a sequence cluster tree diagram for the particular family and indicates its position in a hierarchy of related domains. A button labeled ‘Interactive Display with CDTree’ (Figure 1d) launches CDTree.

Figure 1
(a) Pre-computed domain annotation is retrieved for a protein sequence in the Entrez database, gi|47123187 from Xenopus laevis. The graphical annotation can be obtained by following the ‘Conserved Domains’ link on the Entrez document summary ...
Figure 2
CDTree default display, as launched from a web browser, showing a curated domain hierarchy with an embedded user query sequence. A protein sequence found in NCBI's Entrez database was inspected for the presence of conserved domains. The user has followed ...

CDTree is a helper application for the web browser and must be downloaded and installed on the user's computer. Instructions for installing CDTree are found at http://www.ncbi.nlm.nih.gov/Structure/cdtree/cdtree.shtml. CDTree functions as a viewer for curated protein domain hierarchies; it retrieves data for display via the web browser. CDTree is a combined domain hierarchy viewer and editor. It uses a separate program, Cn3D (5), to view 3D structure and to display and edit multiple alignments of protein structure and sequence. Cn3D is distributed, installed and configured along with CDTree. CDTree requires a recent version of Cn3D, version v4.2, which is contained in the CDTree installation package. The installation package also contains a stand-alone application, ‘fa2cd’, which can be used to convert FASTA-formatted multiple sequence alignments into models stored in the ‘CD’ format, so that they can be imported into CDTree. CDTree also allows the de novo buildup of alignment models starting from single protein sequences. More details can be found on the CDTree home page (see Table 1). A manuscript detailing CDTree and its various uses is in preparation.

When the user launches CDTree from a CD summary page, CDTree displays the contents of the curated domain or domain hierarchy, and serves as a viewer for the evidence supporting a particular subfamily structure. By default, CDTree displays a sequence tree view, a taxonomy view and a hierarchy overview. The sequence tree has been pre-computed and is contained in the data sent by the server. When a query sequence has been submitted to CD-Search and a matching CDD model has been identified, launching CDTree from its CD summary page will cause the user's query sequence to be added to the model and CDTree will recalculate a sequence tree. This tree will now contain the query sequence as well (Figure 2c).

Potentially, this allows users to distinguish between several cases: (i) Query sequences may be bona fide members of clusters which curators have explicitly declared subfamilies, and which may carry specific functional annotation, as in the case illustrated in Figures 1 and and2;2; (ii) query sequences may be bona fide members of subfamilies which do not (yet) carry specific functional annotation; (iii) query sequences may be members of clusters, which curators have not declared subfamilies, as they may not have seen enough evidence for a subfamily at the time of curation; (iv) query sequences may be outliers and not cluster with any particular group of sequences in the curated hierarchy. Only the first scenario listed above may allow the transfer of specific functional annotation from the model to the query. In all the other cases, annotation transfer from the hierarchy's generic ‘parent’ model may be more appropriate, assuming that the ‘parent’ model provides annotation that is valid for all or the majority of members in a superfamily.

CDD CONTENTS AND AVAILABILITY

CDD can be searched directly as part of NCBI's Entrez database and query system (6), where it is listed as the ‘Domains’ database. Entries in CDD are cross-linked reciprocally to NCBI taxonomy, citations in PubMed®, and to protein sequences in Entrez. Links to protein sequences reflect the results of pre-computed RPS–BLAST searches and are updated on a daily basis and stored in the CDART database (7).

The current version of CDD, version v2.09, contains a total of 12 422 models, of which 2494 have been curated at NCBI. Of these curated models, <300 are solitary domain models, while the rest are organized into hierarchies. The largest hierarchies contain well over 100 individual models each. 5252 models have been obtained from Pfam (version 11) (8), 575 have been obtained from SMART (9) and 4101 have been derived from the COG collection (10). Together, these models cover about 69% of non-identical protein sequences in NCBI's Entrez protein database. The full set of models as imported from Pfam, SMART, COG and KOG(9), are available in separate search databases, although not all of them have been indexed in Entrez, since lineage-specific models with limited taxonomic scope, as well as largely redundant models, have been filtered out.

The size of the CDD model collection, details with respect to versions of external databases mirrored in CDD, and the control of redundancy may change over time, as we attempt to provide a resource that is more comprehensive as well as efficient. Expert curation of CDD is an ongoing effort, and we plan to eventually replace a majority of imported models with hierarchies curated at NCBI. Table 1 lists URLs and FTP site addresses for tools and services mentioned above.

Table 1
URLs and FTP-site addresses for CDD and CDD-related services

Acknowledgments

We thank the authors of the Pfam, SMART and COG resources. Development of CDTree would not have been possible without Paul Thiessen and his work on Cn3D, Lewis Geer, Jane He, Naigong Zhang and Praveen Cherukuri and their work on the CDART resource, the NCBI BLAST group, the NCBI IEB and the NCBI C++ toolkit developers. This work was supported by the Intramural Research Program of the National Library of Medicine at National Institutes of Health/DHHS. Comments, suggestions and questions are welcome and should be directed to: vog.hin.mln.ibcn@ofni. Funding to pay the Open Access publication charges for this article was provided by Intramural Research Program of the National Library of Medicine at National Institutes of Health/DHHS.

Conflict of interest statement. None declared.

REFERENCES

1. Marchler-Bauer A., Anderson J.B., Cherukuri P.F., DeWeese-Scott C., Geer L.Y., Gwadz M., He S., Hurwitz D.I., Jackson J.D., Ke Z., et al. CDD: a Conserved Domain Database for protein classification. Nucleic Acids Res. 2005;33:D192–D196. [PMC free article] [PubMed]
2. Brown D., Sjolander K. Functional classification using phylogenomic inference. PLoS Comput. Biol. 2006;2:e77. [PMC free article] [PubMed]
3. Altschul S.F., Madden T.L., Schaffer A.A., Zhang J., Zhang Z., Miller W., Lipman D.J. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997;25:3389–3402. [PMC free article] [PubMed]
4. Marchler-Bauer A., Bryant S.H. CD-Search: protein domain annotations on the fly. Nucleic Acids Res. 2004;32:327–331. [PMC free article] [PubMed]
5. Wang Y., Geer L.Y., Chappey C., Kans J.A., Bryant S.H. Cn3D: sequence and structure views for Entrez. Trends Biochem. Sci. 2000;25:300–302. [PubMed]
6. Wheeler D.L., Barrett T., Benson D.A., Bryant S.H., Canese K., Chetvernin V., Church D.M., DiCuccio M., Edgar R., Federhen S., et al. Database resources of the National Center of Biotechnology Information. Nucleic Acids Res. 2006;34:D173–D180. [PMC free article] [PubMed]
7. Geer L.Y., Domrachev M., Lipman D.J., Bryant S.H. CDART: protein homology by domain architecture. Genome Res. 2002;12:1619–1623. [PMC free article] [PubMed]
8. Finn R.D., Mistry I., Schuster-Bockler B., Griffiths-Jones S., Hollich V., Lassmann T., Moxon S., Marshall M., Khanna A., Durbin R., et al. Pfam: clans, web tools and services. Nucleic Acids Res. 2006;34:D247–D251. [PMC free article] [PubMed]
9. Letunic I., Copley R.R., Pils B., Pinkert S., Schultz J., Bork P. SMART 5: domains in the context of genomes and networks. Nucleic Acids Res. 2006;34:D257–D260. [PMC free article] [PubMed]
10. Tatusov R.L., Fedorova N.D., Jackson J.D., Jacobs A.R., Kiryutin B., Koonin E.V., Krylov D.M., Mazumder R., Mekhedov S.L., Nikolskaya A.N., et al. The COG database: and updated version includes eukaryotes. BMC Bioinformatics. 2003;4:41. [PMC free article] [PubMed]

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles
  • Substance
    Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...