Bioinformatics Advance Access originally published online on August 13, 2004
Bioinformatics 2005 21(1):137-140; doi:10.1093/bioinformatics/bth476
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Bioinformatics vol. 21 issue 1 © Oxford University Press 2005; all rights reserved.
DisProt: a database of protein disorder
1 Center for Information Science and Technology, Temple University Philadelphia, PA 19122, USA
2 Center for Computational Biology and Bioinformatics, Indiana University School of Medicine Indianapolis, IN 46202, USA
3 Laboratory of Statistical Genetics, The Rockefeller University New York, NY 10021, USA
4 Concurrent Pharmaceuticals 502 W. Office Center Dr, Fort Washington, PA 19034, USA
5 Department of Biological Sciences, University of Idaho ID 83844, USA
6 School of Molecular Biosciences, Washington State University Pullman, WA 99164, USA
*To whom correspondence should be addressed.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPROTEIN SELECTION CRITERIADATABASE CONTENTDATABASE ACCESSREFERENCES |
|---|
Summary: The Database of Protein Disorder (DisProt) is a curated database that provides structure and function information about proteins that lack a fixed three-dimensional (3D) structure under putatively native conditions, either in their entirety or in part. Starting from the central premise that intrinsic disorder is an important structural class of protein and in order to meet the increasing interest thereof, DisProt is aimed at becoming a central repository of disorder-related information. For each disordered protein, the database includes the name of the protein, various aliases, accession codes, amino acid sequence, location of the disordered region(s), and methods used for structural (disorder) characterization. If applicable, most entries also list the biological function(s) of each disordered region, how each region of disorder is used for function, as well as provide links to PubMed abstracts and major protein databases.
Availability: www.disprot.org
Contact: kedunker{at}iupui.edu
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPROTEIN SELECTION CRITERIADATABASE CONTENTDATABASE ACCESSREFERENCES |
|---|
Intrinsic order and disorder are defined herein at the amino acid residue level. The backbone atoms, hence also the residues, of ordered regions or proteins undergo small-amplitude, thermally-driven motions about their equilibrium positions determined as time-averaged values. In some cases, ordered regions cooperatively switch between two or more specific conformations. In contrast, intrinsically disordered proteins or regions exist as dynamic ensembles in which the atom positions and the backbone Ramachandran angles vary significantly over time with no specific equilibrium values and typically involve non-cooperative conformational changes. Thus, the existence of disorder is determined by a protein's dynamical properties, and not necessarily by the presence or absence of local secondary structure. We define an intrinsically disordered protein as one that contains at least one disordered region.
Despite the fact that intrinsically disordered proteins fail to form fixed three dimensional (3D) structure under physiological conditions, existing instead as ensembles of conformations, they carry out critically important biological functions (Dunker et al., 2002a, Iakoucheva et al., 2002). Several recent reviews (Wright and Dyson, 1999, Uversky, 2002, Dyson and Wright, 2002, Tompa, 2002) attest to the growing interest in these proteins. In addition, whole-cell NMR experiments demonstrate that intrinsic disorder can exist in vivo (Dedmon et al., 2002) and thus does not result merely from the failure to find the correct conditions or ligand for folding to occur. A collection of about 100 intrinsically disordered examples, many of which were characterized by two or more experimental methods, were assigned one of 28 specific functions and grouped into four functional classes: (1) molecular recognition; (2) molecular assembly; (3) protein modification; and (4) entropic chain activities (Dunker et al., 2002a). Intrinsically disordered regions are typically involved in regulation, signaling and control pathways in which interactions with multiple partners and high-specificity/low-affinity interactions are often requisite. In this way, the functional diversity provided by disordered regions complements those of ordered protein regions.
At the structural level, we proposed that intrinsically disordered regions may exist in molten globule-like (collapsed) and random coil-like (extended) forms (Dunker et al., 2001). Another form of disorder, the pre-molten globule, has been proposed (Uversky, 2002), but it is unclear whether it is truly distinct from the random coil-like class. We also suggested that functions of disordered proteins may arise from the specific disorder form, from interconversion of disordered forms, or from transitions between disordered and ordered conformations (Dunker et al., 2001). These function-associated conformational changes may be brought about by alterations in environmental or cellular conditions (e.g. disorder-to-order transition upon binding during signal transduction).
A major hindrance in the study of intrinsically disordered proteins is the absence of an organized database for these proteins. The lack of a curated, systematized database impedes the dissemination of knowledge regarding disorder examples. An additional complication is the diversity of language used to describe similar phenomena. For example, while the absence of a fixed 3D structure is used herein as the definition of disorder, and thus molten globule-like proteins and molten globule-like regions would be included, others prefer to view disordered proteins and regions as operationally identified as missing coordinates in the Protein Data Bank (PDB) (Linding et al., 2003b), as essentially devoid of helix or sheet (Liu et al., 2002), as dynamically flexible ensembles (Ward et al., 2004), or as high B-factor coils (Linding et al., 2003a), (Linding et al., 2003b). Thus, a coordinated, widely used database could help to unify the terminology in this field. One notable attempt to provide such a resource of protein disorder is the ProDDO database (Sim et al., 2001); however, it is not curated, its contents are limited to PDB entries, and it does not provide structural or functional annotation of disorder. For these reasons, we embarked on the construction of the Database of Protein Disorder, known as ‘DisProt’, in which the available structural/functional information was obtained through exhaustive searches of the relevant literature and biological databases Dunker et al., 2002a.
|
PROTEIN SELECTION CRITERIA |
|---|
|
TOPABSTRACTINTRODUCTIONPROTEIN SELECTION CRITERIADATABASE CONTENTDATABASE ACCESSREFERENCES |
|---|
DisProt contains both proteins that lack a fixed 3D structure along their entire lengths as well as proteins that have local regions lacking a fixed 3D structure. The criteria for the inclusion of a protein in DisProt are discussed in the following paragraphs.
Several methods have been used to identify proteins wholly lacking fixed 3D structures, including several variants of NMR spectroscopy, circular dichroism (CD) spectroscopy, small-angle X-ray scattering (SAXS) and estimation of hydrodynamic radius. The details of each method are described elsewhere (Rose, 2002, Daughdrill et al., 2004). NMR spectroscopy provides the most direct estimation of the internal motions that characterize disordered ensembles and can be used to identify each residue lacking a fixed structure if peak assignments are made. However, for many proteins a global characterization is made without making individual NMR peak assignments. CD, SAXS and hydrodynamic radius measurements do not provide residue-by-residue information. For proteins characterized by these methods, part of a given protein might fold into a structured domain that is missed because it represents a small fraction of the total sequence. Thus, inclusion of such proteins in the database characterized by these methods introduces the possibility that regions of ordered structure are misclassified as regions that lack fixed structures. While proteins characterized by a single method are included, an effort is being made to identify proteins characterized by multiple methods. Such well-characterized proteins can serve as useful prototypes for proteins in particular classes.
One of the main methods for characterizing disorder is by missing electron density in X-ray structures, which leads to missing coordinates in the corresponding PDB files. For example, about two-thirds of the structured chains in the PDB contain localized regions of missing electron density (Obradovic et al., 2003); these regions vary in length from single residues to long segments containing more than 100 consecutive residues. For inclusion in this database, a given residue needs to lack electron density for all of its backbone atoms, not just the side chain atoms. For long regions of missing coordinates in X-ray structures there is the possibility that they are actually ordered domains that move as rigid bodies and thus fail to scatter X-rays coherently. Such regions have been called ‘wobbly domains’ (Dunker et al., 2001, Daughdrill et al., 2004). Although such unobserved regions would be misclassified as disordered in the present database, similarities in amino acid compositions between disordered regions characterized by different methods indicate that wobbly domains are the exception rather than the rule (Dunker et al., 2002b). Over time, an effort will be made to identify such misclassified, wobbly-domain regions and remove them from the database. For example, protease digestion typically releases wobbly domains as large, protease resistant fragments, while disordered regions tend to be cut at multiple loci and fail to yield such large fragments. Thus, a simple procedure based on protease digestion could be used to test whether a long unobserved region is a wobbly domain or not. In addition, NMR could be used if the entire protein is not too large.
Several NMR techniques can indicate regions of disorder within otherwise ordered protein structures. When modeling the 3D structure to fit the NMR data, some regions give multiple solutions to the NMR restraints and so appear as ensembles of possible structures. The standard view is that such regions are ordered, but the data are simply insufficient to identify the particular 3D structure. The alternative possibility is that such regions actually lack particular 3D structures because they are disordered. Such ambiguities can be sorted out by additional NMR relaxation experiments that directly measure the motions of the residues in such regions (Bracken, 2001).
|
DATABASE CONTENT |
|---|
|
TOPABSTRACTINTRODUCTIONPROTEIN SELECTION CRITERIADATABASE CONTENTDATABASE ACCESSREFERENCES |
|---|
DisProt was designed as a companion to major online protein repositories; hence, whenever possible, DisProt provides links to PDB, SWISS-PROT and TrEMBL, GenBank, and PIR databases. The release 1.2 of the DisProt knowledge base consists of 154 proteins with 190 disordered regions, 164 of which are longer than 30 consecutive residues. Currently, DisProt is essentially non-redundant: no two proteins have sequence identity >50%, and only four pairs and one triple of the proteins have sequence identity between 30 and 50%. The non-redundant nature of the database, however, was unintentional. Further releases of DisProt will include similar sequences, but an effort will be made to provide a representative, non-redundant subset to the community. Although relatively small, the set of DisProt proteins provides a considerable coverage of protein sequence space. For example, a BLAST search (with BLOSUM62, gap penalties of 11/1, and E-value threshold of 10–2) revealed that 6720 or 5.1% of 132 648 SWISS-PROT sequences were covered by our database. Since low-complexity regions were not filtered by this procedure due to the positive correlation between disorder and low-complexity sequences, we observe that the 5.1% coverage can serve as an upper limit estimate.
The basic unit in DisProt is a protein chain that has been characterized as having one or more disordered regions. For each listed protein, the database provides the name and various aliases, accession codes and links to other databases, amino acid sequence, and the location(s) of the disordered region(s). Additional fields were allocated for the following information: (1) detection method(s), e.g. X-ray diffraction, NMR, circular dichroism; (2) function of the disordered region, e.g. containing a site for phosphorylation, binding to a specific partner; (3) structural transitions if known, i.e. whether a disorder-to-order or an order-to-disorder transition has been associated with the function of the disordered region; and (4) references, specifically describing the structural characterization or functions of the disordered region including direct links to PubMed citations.
|
DATABASE ACCESS |
|---|
|
TOPABSTRACTINTRODUCTIONPROTEIN SELECTION CRITERIADATABASE CONTENTDATABASE ACCESSREFERENCES |
|---|
DisProt provides a browsable list of current proteins, as well as lists of proteins classified by their functions, possible structural transitions and the method(s) by which they were characterized. Every protein in our database can be viewed in HTML format directly at http://www.disprot.org/protein.php?id=disprot_id. In addition to the HTML representation, we provide the data in XML format for each disordered protein. Finally, a reduced view of the data is made available in FASTA format.
DisProt allows standard and advanced search. The standard search mode provides full-text search of the entire data base, while the advanced mode provides an interface for a fast, sequence-based search of DisProt, an option rarely found in similar databases. Through this option, a user enters a query sequence and is provided with a list of DisProt proteins with significant similarity to the query sequence.
Communication and feedback
A major function of DisProt is to serve as a central repository for submission information related to intrinsically disordered protein. Through a ‘Comment’ feature we encourage feedback from the bioscience community in terms of verifying the information we provide, as well as obtaining relevant data either from the ongoing research or from the previous research, both published and unpublished. It will be the responsibility of biological experts from our group to verify the submitted information and to update the database accordingly.
Implementation
DisProt was built using PostgreSQL relational database management system running on a Suse Linux platform. Its web interface is based on an Apache web server scripting with PHP language.
Supplementary data
Very often, gaps in the ATOM list of PDB records are indicative of protein disorder. We have parsed the August 5, 2003 release of PDB and extracted all protein chains with gaps of three or more residues in the ATOM list of X-ray characterized proteins. We provided this data for download through DisProt (missingXray.080503.txt file lists chains with missing residues in FASTA format). This data is expected to facilitate future growth of DisProt. Please note that the residue numbering herein sometimes does not match that of the corresponding PDB entry, so comparisons with the PDB entry should be carried out using sequence matching, not using the residue numbers.
Future developments
We will continue to expand the database in terms of the number of protein disorder records. We will also work towards improving the content of DisProt records. We are already in the process of collecting information for a new functional narrative field in the protein records. This field will contain not only a summarizing description of the function of the disordered region, but also an indication of the biological context associated with the particular function. Various functional classification schemes Dunker et al., 2002a, Tompa, 2002 will be included in future releases of the database. In addition, to minimize the problems associated with different views on disordered proteins, we will make an effort to clearly indicate whether a given protein or region has molten globule-like, pre-molten globule-like or random coil-like properties.
|
Acknowledgments |
|---|
Support from NIH R01 LM007688-01 to A.K.D. and Z.O. is gratefully acknowledged. A.K.D. thanks the Indiana Genomics Initiative (INGEN) for its generous support; INGEN is funded in part by the Lilly Endowment, Inc.
Received on February 5, 2004; revised on July 2, 2004; accepted on August 3, 2004
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPROTEIN SELECTION CRITERIADATABASE CONTENTDATABASE ACCESSREFERENCES |
|---|
Bracken, C. (2001) NMR spin relaxation methods for characterization of disorder and folding in proteins. J. Mol. Graph. Model., 19, 3–12[CrossRef][Medline].
Daughdrill, G.W., Pielak, G.J., Uversky, V.N., Cortese, M.S., Dunker, A.K. In Buchner, J and Kiefhaber, T.1 (Eds.). Protein Folding Handbook, (2004) , Weinheim Wiley-VCH.
Dedmon, M.M., Patel, C.N., Young, G.B., Pielak, G.J. (2002) FlgM gains structure in living cells. Proc. Natl Acad. Sci., USA, 99, , pp. 12681–12684
Dunker, A.K., Lawson, J.D., Brown, C.J., Williams, R.M., Romero, P., Oh, J.S., Oldfield, C.J., Campen, A.M., Ratliff, C.M., Hipps, K.W., et al. (2001) Intrinsically disordered protein. J. Mol. Graph. Model., 19, 26–59[CrossRef][Web of Science][Medline].
Dunker, A.K., Brown, C.J., Lawson, J.D., Iakoucheva, L.M., Obradovic, Z. (2002a) Intrinsic disorder and protein function. Biochemistry, 41, 6573–6582[CrossRef][Medline].
Dunker, A.K., Brown, C.J., Obradovic, Z. (2002b) Identification and functions of usefully disordered proteins. Adv. Protein Chem., 62, 25–49[Web of Science][Medline].
Dyson, H.J. and Wright, P.E. (2002) Coupling of folding and binding for unstructured proteins. Curr. Opin. Struct. Biol., 12, 54–60[CrossRef][Web of Science][Medline].
Iakoucheva, L.M., Brown, C.J., Lawson, J.D., Obradovic, Z., Dunker, A.K. (2002) Intrinsic disorder in cell-signaling and cancer-associated proteins. J. Mol. Biol., 323, 573–584[CrossRef][Web of Science][Medline].
Linding, R., Jensen, L.J., Diella, F., Bork, P., Gibson, T.J., Russell, R.B. (2003a) Protein disorder prediction: implications for structural proteomics. Structure, 11, 1453–1459[Medline].
Linding, R., Russell, R.B., Neduva, V., Gibson, T.J. (2003b) GlobPlot: exploring protein sequences for globularity and disorder. Nucleic Acids Res, 31, 3701–3708
Liu, J., Tan, H., Rost, B. (2002) Loopy proteins appear conserved in evolution. J. Mol. Biol., 322, 53–64[CrossRef][Web of Science][Medline].
Obradovic, Z., Peng, K., Vucetic, S., Radivojac, P., Brown, C.J., Dunker, A.K. (2003) Predicting intrinsic disorder from amino acid sequence. Proteins, 53, (Suppl. 6), 566–572.
(Ed.). Advances in Protein Chemistry, (2002) , NY Academic Press Vol. 62, .
Sim, K.L., Uchida, T., Miyano, S. (2001) ProDDO: a database of disordered proteins from the Protein Data Bank (PDB). Bioinformatics, 17, , pp. 379–380
Tompa, P. (2002) Intrinsically unstructured proteins. Trends Biochem. Sci., 27, 527–33[CrossRef][Web of Science][Medline].
Uversky, VN. (2002) Natively unfolded proteins: a point where biology waits for physics. Protein Sci., 11, 739–756[CrossRef][Web of Science][Medline].
Ward, J.J., Sodhi, J.S., McGuffin, L.J., Buxton, B.F., Jones, D.T. (2004) Prediction and functional analysis of native disorder in proteins from the three kingdoms of life. J. Mol. Biol., 337, 635–645[CrossRef][Web of Science][Medline].
Wright, P.E. and Dyson, H.J. (1999) Intrinsically unstructured proteins: re-assessing the protein structure-function paradigm. J. Mol. Biol., 293, 321–331[CrossRef][Web of Science][Medline].
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. J. McGuffin Intrinsic disorder prediction from the analysis of multiple protein fold recognition models Bioinformatics, August 15, 2008; 24(16): 1798 - 1804. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Schlessinger, M. Punta, and B. Rost Natively unstructured regions in proteins identified from contact predictions Bioinformatics, September 15, 2007; 23(18): 2376 - 2384. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Shimizu, S. Hirose, and T. Noguchi POODLE-S: web application for predicting protein disorder by using physicochemical features and reduced amino acid set of a position-specific scoring matrix Bioinformatics, September 1, 2007; 23(17): 2337 - 2338. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hirose, K. Shimizu, S. Kanai, Y. Kuroda, and T. Noguchi POODLE-L: a two-level SVM prediction system for reliably predicting long disordered regions Bioinformatics, August 15, 2007; 23(16): 2046 - 2053. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. V. Balasingham, R. F. Collins, R. Assalkhou, H. Homberset, S. A. Frye, J. P. Derrick, and T. Tonjum Interactions between the Lipoprotein PilP and the Secretin PilQ in Neisseria meningitidis J. Bacteriol., August 1, 2007; 189(15): 5716 - 5727. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Sykacek, R. Clarkson, C. Print, R. Furlong, and G. Micklem Bayesian modelling of shared gene function Bioinformatics, August 1, 2007; 23(15): 1936 - 1944. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z.-Q. Ye, S.-Q. Zhao, G. Gao, X.-Q. Liu, R. E. Langlois, H. Lu, and L. Wei Finding new structural and sequence attributes to predict possible disease association of single amino acid polymorphism (SAP) Bioinformatics, June 15, 2007; 23(12): 1444 - 1450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Min, K. H. Kirsch, Y. Zhao, S. Jeay, A. H. Palamakumbura, P. C. Trackman, and G. E. Sonenshein The Tumor Suppressor Activity of the Lysyl Oxidase Propeptide Reverses the Invasive Phenotype of Her-2/neu-Driven Breast Cancer Cancer Res., February 1, 2007; 67(3): 1105 - 1112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Zimmermann and U. H. E. Hansmann Support vector machines for prediction of dihedral angle regions Bioinformatics, December 15, 2006; 22(24): 3009 - 3015. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. V. Galzitskaya, S. O. Garbuzynskiy, and M. Yu. Lobanov FoldUnfold: web server for the prediction of disordered regions in protein chain Bioinformatics, December 1, 2006; 22(23): 2948 - 2949. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Vullo, O. Bortolami, G. Pollastri, and S. C. E. Tosatto Spritz: a server for the prediction of intrinsically disordered regions in protein sequences using kernel machines. Nucleic Acids Res., July 1, 2006; 34(Web Server issue): W164 - W168. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. R. Romero, S. Zaidi, Y. Y. Fang, V. N. Uversky, P. Radivojac, C. J. Oldfield, M. S. Cortese, M. Sickmeier, T. LeGall, Z. Obradovic, et al. Alternative splicing in concert with protein intrinsic disorder enables increased functional diversity in multicellular organisms PNAS, May 30, 2006; 103(22): 8390 - 8395. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Haynes and L. M. Iakoucheva Serine/arginine-rich splicing factors belong to a class of intrinsically disordered proteins Nucleic Acids Res., January 10, 2006; 34(1): 305 - 312. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Geddie, T. L. O'Loughlin, K. K. Woods, and I. Matsumura Rational Design of p53, an Intrinsically Unstructured Protein, for the Fabrication of Novel Molecular Sensors J. Biol. Chem., October 21, 2005; 280(42): 35641 - 35646. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||