Key words

1 Introduction

Looking back to 1971, when the Protein Data Bank was founded [1], one cannot help feeling that the study of protein structure must have been a lot simpler then. There were only seven experimentally determined protein structures at the time, and the data for each, including the proteins ’ atomic coordinates, were stored in simple, fixed-format text files. Admittedly, accessing and displaying this information was trickier, and computers with graphics capabilities tended to be bulky and expensive things. These days, access and display of the data over the Web are vastly easier, but with this comes the problem, not only in the huge increase in the amount of information, but in the multiplicity of sources from which it can be obtained. New servers and services continually appear, while existing ones are modified and improved. Conversely, other servers are abandoned, switched off or neglected, becoming more and more out of date with time. Thus it has become really difficult to know where to go to get relevant answers most easily. Various lists are available on the Web—for example the Nucleic Acids Research (NAR) list at http://www.oxfordjournals.org/our_journals/nar/database/a. This chapter aims to highlight some of the more useful, and up-to-date (at time of writing), sources of information on protein structure that are currently available.

2 Structures and Structural Data

2.1 Terminology

Firstly, it is important to define what is meant by the term “protein structure .” It is a term that tends to be somewhat loosely used. A preferable term is “model,” as the 3D structures of large molecules such as proteins are models of the atom types, atomic x-, y-, z-coordinates and other parameters that best fit the experimental data. The reason the term “structure” is so commonly used for these models is to distinguish them from “theoretical,” or “homology-built,” models. Nevertheless, it is important to remember that all are models of reality and that only the former type is based on experimental evidence.

Another loosely used term is “database.” Technically, the databases mentioned here are not databases at all, but rather “data resources”—many of which rely on a database for storing and serving up the data. However, the term “database” is becoming common usage for the types of resources described here (e.g., the NAR Database issues), so it is the meaning we adopt here.

2.2 The Protein Data Bank (PDB ) and the wwPDB

The primary repository of 3D structural data on proteins (and other biological macromolecules, including RNA , fragments of DNA , carbohydrates, and different complexes of these molecules) is the Protein Data Bank . As mentioned above, this was founded in 1971 and was located at Brookhaven National Laboratories. In October 1998, the management of the archive was taken over by the Research Collaboratory for Structural Bioinformatics (RCSB), a consortium consisting of Rutgers University, the National Institute of Standards and Technology (NIST) and the San Diego Supercomputer Center [2]. Since 2003 the archive has been managed by an international consortium called the world-wide Protein Data Bank (wwPDB ) whose partners comprise: the RSCB, the Protein Data Bank Europe (PDBe ) at the European Bioinformatics Institute (EBI), the Protein Data Bank Japan (PDBj) at Osaka University, and, more recently, the BioMagResBank (BMRB) at the University of Wisconsin-Madison [3, 4]. Access to the primary data is via the wwPDB’s website: http://www.wwpdb.org. The data come in three different formats: old-style PDB -format files, macro-molecular Crystallographic Information File (mmCIF) format [5], and a XML-style format called PDBML/XML [6]. Due to format limitations, the old-style PDB-format files are no longer available for extremely large structural models (i.e., those having too many atoms, residues or chains than the fixed-format fields allow for). For many of the structures, the wwPDB also make the original experimental data available. Thus, for structural models solved by X-ray crystallography , one can often download the structure factors from which the model was derived, while for structures solved by nuclear magnetic resonance (NMR ) spectroscopy, the original distance and angle restraints can be obtained. As of July 2015, the wwPDB contained over 110,000 structural models, each identified by a unique 4-character reference code, or PDB identifier.

A key task the wwPDB have performed is the remediation of the legacy PDB archive to fix and make consistent the entire PDB data, in particular relating to ligands and literature references [7]. The PDBe and UniProt groups at the EBI have mapped the sequences in the PDB entries onto the appropriate sequences in UniProt [8]. More recently, the focus has been on validation of the structural data, with the establishment of several Validation Task Forces [9–11], and the reporting of quality indices or validation information for each structure.

2.3 Structural Data and Analyses

Rather than download the raw data from the wwPDB for each protein of interest, it is usually more convenient to obtain the required information directly from one of the myriad protein structure databases on the Web. These come in many shapes and sizes, catering for a variety of needs and interests.

At the simplest level are the sites that provide “atlas” pages—one for every PDB entry—each containing general information obtained from the relevant PDB file. There are usually graphical representations of the structural model together with links that provide interactive 3D visualizations using Java-based, or other, viewers. Each of the founding members of the wwPDB have their own atlas pages: the RCSB, the PDBe , and PDBj. In addition, there are several other sites that have much to commend them, and some of these are mentioned below.

Beyond the atlases, there are a host of other types of sites and servers. These include those providing information on specific structural motifs, focus on selected protein families, classify protein folds, compare protein structure s, provide homology-built models for proteins for which no structure has been determined, and so on. This chapter cherry-picks a few of the more interesting and useful sites to visit.

3 Atlases

Table 1 lists the seven best-known and useful of the atlas sites. All have been developed independently and, not unexpectedly, all have much in common as the information comes from the same source: the PDB entry. The protein name, authors, key reference, experimental methods, atomic coordinates, and so on are obviously identical on all sites. Also common are certain derived data, including quality assessment of each structural model, and information about the protein’s likely “biological unit.”

Table 1 Protein structure atlases
Full size table

Quality assessment is a crucial issue as not all models are equally reliable, and much has been written on this topic over the years [9, 12–16]. The main problem is that the results of any experiment contain errors, but for structural models it is difficult to estimate the extent of those errors. For X-ray models, a rough guide of quality is provided by the resolution at which the structure was solved and its R-factor, but for NMR models there is no such ready measure. Some atlases do provide indications of which models are more reliable, as described shortly.

The second important issue is knowing what a given protein ’s biological unit is. This is not always obvious from the PDB entry. The problem is that the deposited coordinates from an X-ray crystal structure determination correspond to the molecule(s) in the asymmetric unit. This may give a false impression of how the protein operates in vivo. For example, what may look like a monomer from the PDB entry, is, in real life, a dimer, or a trimer, etc. Conversely, the PDB entry might give the coordinates of a dimer, yet the biological unit happens to be a monomer—the dimeric structure being the result of packing in the crystal. For any structural analysis it is crucial to know what the true biological unit is. For some proteins the biological unit has been determined experimentally, and so is known with great confidence. In others it has to be deduced computationally by analysis of the packing of the individual chains in the crystal. Some interfaces are more substantial than others and hence likely to represent genuine biological interactions rather than happenstance crystal contacts. Most of the atlases provide information on the known, or predicted, biological unit. The most commonly used prediction method is Protein Interfaces, Surfaces and Assemblies (PISA) [17].

Beyond these general similarities, the atlases differ in sufficient respects to make them complement one another; they differ in what additional information they pull in, the links they make to external resources, and the analyses of the 3D structures they provide. Consequently, the atlas of choice can be either a matter of personal preference or depend on the type of information required at the time.

Here we focus only those aspects that make each one unique, useful or interesting. We start with the atlases provided by the founding members of the wwPDB , and then discuss some of the others.

3.1 The RCSB PDB

The RCSB’s website [18] has been revamped several times and is an extremely rich source of information about each PDB entry. It used to be a little overwhelming for novices, but recently a great deal of effort has gone into simplifying the design as well as adding new information—such as the relationship of structures to their corresponding genes and to associated diseases and therapeutic drugs. A specific aim of the website has been to “bring a structural view of biology and medicine to a general audience.”

3.1.1 Summary Page

Figure 1 shows the summary page for PDB entry 1ayy, a glycosylasparaginase. The top box shows the primary citation for this entry, being the published description of the experiment that resulted in the structural model and any analysis the authors might have performed on it, including relating the structure to the protein ’s biological function. To the right is a thumbnail image of the protein and links for viewing it in one of three molecular graphics viewers. The “More Images” link shows the asymmetric unit and the biological unit, as described above (although in many cases they are identical). The latter is either as defined by the depositors or as predicted by the PISA algorithm.

Fig. 1
figure 1

Part of the RCSB atlas page for PDB entry 1ayy, a glycosylasparaginase determined by X-ray crystallography at 2.32 Å

Full size image

The Molecular Description box provides a schematic diagram of the protein ’s sequence and structural domains, together with its secondary structure , and which parts of the protein the structure corresponds to. An expanded view can be obtained by clicking on “Protein Feature View,” as shown in Fig. 2. Often structural models are not of the whole protein but merely cover one or two domains or, in some cases, are mere fragments of the protein. The diagram makes it clear what the coverage is. The little plus symbol at the bottom opens up a window showing other known structures of the same protein—which is particularly useful in identifying structures that may be more complete, or solved at a higher resolution . The sequence domains are as defined by Pfam [19], while the structural domain definitions come from SCOP [20].

Fig. 2
figure 2

The Protein Feature View of PDB entry 1ayy on the RCSB PDB server. The diagram shows the protein ’s sequence (Pfam ) and structural (SCOP ) domains, its hydrophathy, secondary structure , and structural coverage

Full size image

The large box at the bottom of Fig. 1 is a “validation report slider” providing an at-a-glance assessment of the structure’s likely quality (only available for X-ray models). The graphic indicates how the structure compares on a number of quality-related parameters against all other structures in the database as well as structures solved at the same resolution . The parameters include the R free, an atom-atom “clash score,” number of Ramachandran plot outliers as computed by the MolProbity structure validation program [21], and the real-space R-value Z-score as computed by the Uppsala Electron-Density Server [22]. An almost identical schematic is provided by the PDBe website (see Fig. 3). A link above the schematic provides the full validation report for the structure in question.

Fig. 3
figure 3

Validation schematics for PDB entry 1sqt, as shown on the PDBe website. Above the thumbnail images of the protein on the left are two “quality sliders.” The top one shows how well the overall model quality compares against all other structures in the PDB, and the second how well the model fits the experimental data from which it was derived. The red end of the slider indicates a poor model/fit, while the blue indicates the model is a good one. The right-hand set of sliders show the quality of the model as judged by four different global quality criteria: the R free, an atom-atom clash score computed by MolProbity , number of Ramachandran plot outliers, and the real-space R-value Z-score as computed by the Uppsala Electron-Density Server. The black vertical box on each slider corresponds to the percentile rank of the given score with respect to the scores of previously deposited PDB entries, while the white vertical box shows the rank with respect to entries solved at a similar resolution

Full size image

3.1.2 Other Information

Besides the summary information, further structural details are presented on additional pages titled: 3D View, Sequence, Annotations, Seq. Similarity, 3D Similarity, Literature, Biology & Chemistry, Methods, and Links.

For ligands there is the 3D Java-based Ligand Explorer [23] which allows you to select and view different types of protein –ligand interactions. There is also a schematic 2D PoseView [24] diagram of the protein–ligand interactions.

The advanced search option allows for quite complex queries and subqueries on the data, telling you how many hits each set of conditions returns as you refine your search.

3.1.3 Molecule of the Month

One particularly eye-catching feature of the RCSB site is the “Molecule of the Month” pages written by David S. Goodsell of The Scripps Research Institute and illustrated with his beautiful plots [25]. Each month the structure and function of a different protein or protein family is described, with specific references to the PDB entries that have contributed to the understanding of how the proteins achieve their biological roles. The collection of short articles, which are suitable for specialists and non-specialists alike, dates back to the year 2000 and now numbers over 180 entries, providing a nice reference and educational resource. Additionally, and particularly useful as teaching materials, are the accompanying videos, posters, lesson plans and curricula provided by the PDB-101 educational portal.

3.2 The PDBe

The website of the Protein Data Bank Europe (PDBe ) [26] has many similarities to the RCSB’s. The atlas pages for each entry show the usual summary information describing the structure and the experimental details used to obtain it. Additional pages relate to Structure analysis, Function and Biology, Ligands and Environments, and Experiments and Validation. The Molecular Details link shows the protein ’s sequence features, a diagram of its secondary structure topology and a 3D JSmol view (Fig. 4). These are connected such that clicking on one diagram highlights the corresponding residues in the others.

Fig. 4
figure 4

The “Molecule details” of PDB entry 2oig, a mouse dCTP pyrophosphatase 1, from the PDBe website. The tracks at the top represent the protein ’s sequence and structure domains, its secondary structure and residue-by-residue quality indicators. It is similar to the RCSB’s Protein Feature View in Fig. 2. At bottom left is a topology diagram of the secondary structure elements—here four helices. Clicking on the diagram identifies the residues, and the corresponding residues are highlighted in the diagram above (by a shaded grey box) and in the JSmol 3D image on the right

Full size image

3.2.1 PDBeFold

In addition to the atlas pages, the PDBe website has a number of useful applications. These include PDBeFold which performs fold matching of any one or more protein structure s against one or more others. The server makes use of the secondary structure similarity matching program SSM [27]. You can match a single PDB entry against another, or against all structures in the PDB. You can upload your own PDB-format file, or a list of PDB pairs to compare. The outputs include structure-based alignments with computed rmds values and various scores of significance. The superposed structures can be viewed or their coordinates downloaded.

3.2.2 PDBeMotif

PDBeMotif [28, 29] allows searches for sequence and structural motifs as well as for ligands and specific protein –ligand interactions. Structural motifs can be defined in terms of patterns of secondary structure , φ/ψ and χ angles, and Cα and side-chain positions. Searches are entered either via a simple Web form or using a graphical query generator. The hits from a search can be viewed in three dimensions, aligned by ligand, PROSITE pattern, active site residues or by environment. One can generate various statistics on protein–ligand interactions (e.g., to compare the different distributions of residues binding to ATP and GTP). Of particular use is an option to upload a PDB file and scan its ligands and active sites against the PDBe data.

3.2.3 PDBePISA

PDBePISA is a service for computing the stability of protein –protein or other macromolecular complexes (protein, ligands, and DNA /RNA ). It uses the PISA [17] program and provides an analysis of the surfaces, interfaces, and assemblies to suggest which groupings are likely to be biological assemblies rather than crystal packing ones. The assessment is based on the number, type, and strength of interactions across each interface. The service is especially useful for obtaining the full biological units for large multimeric complexes where the PDB entry consists only of a single protein chain.

3.3 JenaLib

The Jena Library of Biological Macromolecules, JenaLib [30], was one of the earliest sites offering atlas pages for each PDB entry, specializing in hand-curated images of the structures showing functionally informative views. Rather than split information across several pages, JenaLib shows all the information on a single page but has a collapse/expand mechanism for controlling what is shown and what is hidden. In addition to several of the standard 3D viewers the site features its own: the JenLib Jmol viewer. This viewer is an extension of Jmol which has a number of options not found in other viewers, such as highlighting of PROSITE motifs, single amino acid polymorphisms and CATH [31] or SCOP domain structures.

JenaLib has more links to external databases than the other atlas sites and is particularly strong on its many visualizations of each entry—both in terms of its interactive viewing options and its preprepared still images.

A particularly useful feature is a form for generating lists of PDB entries according to a number of criteria. Additionally, there are a number of precomputed lists of structures; for example, all nucleic acid structures without protein , all carbohydrate structures, and so on.

3.4 OCA

OCA ’s main difference from the other atlases is its linkage between proteins and the diseases associated with them. It differs also in that its home page is a search form, with searches possible on gene name, function, disease and membrane orientation (for membrane-spanning proteins).

3.5 PDBsum

The last of the atlases described here is PDBsum [32, 33]. Its original aim was to provide pictorial structural analyses where other sites were presenting tables of numbers, but the other atlases have come to include more schematic diagrams over the years. It still provides some unique features, including an option that allows users to upload their own PDB files and get a set of password-protected PDBsum pages generated for them.

3.5.1 Summary Page

Each entry’s summary page has a thumbnail image of the structure, the usual header information and a clickable schematic diagram showing how much of the full-length protein sequence is actually represented by the 3D structural model. The diagram shows the protein’s secondary structure and annotates it with any Pfam sequence domains and CATH structural domains. Also included is a thumbnail Ramachandran plot of the protein and the primary citation.

3.5.2 Quality Assessment

Hovering the mouse over the thumbnail Ramachandran pops up a full-size version. A reliable model will have more points in the core regions (colored red) and, ideally, none in the cream-colored, disallowed regions. Residues in the latter are labeled, so if a model has many labeled residues, it might be an idea to look for an alternative. Clicking on the plot goes to a page showing the summary results from the PROCHECK quality assessment program [34] and from this page you can generate a full PROCHECK report.

3.5.3 Enzyme Reactions

For enzymes, the relevant reaction catalyzed by the enzyme is shown by a reaction diagram where possible. If any of the ligands bound to the protein correspond to any of the reactants, cofactors or products, the corresponding molecule in the diagram is boxed in red. If a ligand is merely similar to one of these, a blue box surrounds the molecule instead and a percentage similarity is quoted.

3.5.4 Figures from Key references

The majority of experimentally determined protein structure s are reported in the scientific literature, often in high profile journals, and each PDB file cites the “key” reference—i.e., the one describing the structure determination, analysis and biological significance of the protein. Like the other atlas sites, PDBsum cites this reference, shows its abstract and provides links to both the PubMed entry and to the online version of the article. Where PDBsum differs is that for many of these references it also gives one or two figures (plus figure legends) taken directly from the key reference itself [35]. This is done with permission from the relevant publishers and is useful for two reasons. Firstly, a carefully selected figure can speak volumes about an important aspect of the protein’s structure or function. And secondly, each paper’s lead author is requested to review which figures have been selected by the automated process and, if need be, suggest better choices. About one in six authors take the trouble to do this. And some even add an additional comment to appear on the entry’s summary page (e.g., PDB entry 1hz0).

3.5.5 Secondary Structure and Topology Diagrams

From the summary page are various additional pages giving schematic diagrams of different aspects of the 3D structure. The “Protein” page shows a diagram of the chain’s secondary structure elements, much like the RCSB’s diagram shown in Fig. 2. Additional features include the annotation of residues that are catalytic—as defined in the Catalytic Site Atlas (CSA) [36]—or are included in the SITE records of the PDB file, or interact with a ligand, DNA /RNA or metal, or belong to a PROSITE pattern [37]. CATH structural domains are marked on the sequence, in contrast to the RCSB’s diagram which uses SCOP . Where there is information on the conservation of each residue in the sequence—obtained from the ConSurf-HSSP site [38]—the secondary structure plot can be redisplayed with the residues colored by their conservation.

Next to the secondary structure plot is a topology diagram either of the whole chain or, where it has been divided into its constituent CATH domains, of each domain (Fig. 5). The diagram shows the connectivity of the secondary structure elements, with the constituent β-strands of each β-sheet laid side-by-side, parallel or antiparallel, to show how each sheet in the chain/domain is formed, and where any helices are found relative to the sheets.

Fig. 5
figure 5

A topology diagram taken from PDBsum for the second domain of chain A in PDB entry 2b6d: a bovine lactoferrin. The diagram illustrates how the β-strands, represented by the block arrows, join up, side-by-side, to form the domain’s central β-sheet. The diagram also shows the relative locations of the α-helices, here represented by cylinders. The small arrows indicate the directionality of the protein chain, from the N- to the C-terminus. The numbers within the secondary structural elements correspond to the residue numbering given in the PDB file

Full size image

3.5.6 Intermolecular Interactions

Some of the other pages are devoted to schematic representations of intermolecular interactions. Thus for each ligand molecule or metal ion in the structure there is a schematic LIGPLOT diagram [39] of the hydrogen bonds and non-bonded interactions between it and the residues of the protein to which it is bound (see Fig. 6). Similarly, any DNA –protein interactions are schematically depicted by a NUCPLOT diagram [40]. Protein–protein interactions at the interface between two or more chains are shown by two plots: the first shows an overview of which chains interact with which (Fig. 7b), while the second shows which residues actually interact across the interface (Fig. 7c).

Fig. 6
figure 6

LIGPLOT for PDB entry 2oig, tyrosine kinase c-Src, as given in PDBsum showing the interactions between the bound molecule imatinib (a drug, brand name Gleevec) with the residues of the protein . Hydrogen bonds are represented by dashed lines. Residues that interact with the ligand via non-bonded contacts only are represented by the eyelashes

Full size image
Fig. 7
figure 7

Extracts from the protein –protein interaction diagrams in PDBsum for PDB entry 1cow, bovine mitochondrial F1-ATPase. (a) Thumbnail image of the 3D structural model which contains seven protein chains: three of ATPA1_BOVIN (chains A, B, and C), three of ATPB_BOVIN (chains D, E, and F), and a fragment of ATPG_BOVIN (chain G). (b) Schematic diagram showing the interactions between the chains. The area of each circle is proportional to the surface area of the corresponding protein chain. The extent of the interface region on each chain is represented by a colored wedge whose color corresponds to the color of the other chain and whose size signifies the interface surface area. (c) A schematic diagram showing the residue–residue interactions across one of the interfaces, namely that between chains D and G. Hydrogen bonds and salt bridges are shown as solid lines while non-bonded contacts are represented by dashed lines

Full size image

4 Homology Models and Obsolete Entries

4.1 Homology Modeling Servers

As mentioned above, there were over 110,000 structural models in the wwPDB as of July 2015. However, some were not of proteins and many were duplicates: that is the same protein solved under different conditions, or with different ligands bound, or with one or more point mutations. In terms of unique protein sequences, as defined by the UniProt identifier, this 110,000 corresponded to only about 33,000 unique proteins. (Compare this number with the 620 million protein sequences in the European Nucleotide Archive (ENA) [41]). Moreover, for many of these, the 3D structure represents only a part of the full sequence—a single domain or just a fragment.

So for many proteins there is no corresponding structural model in the PDB . In these cases it is common to build a homology model based on the 3D structural model of a closely related protein (if there is one). The PDB used to accept homology-built models together with the experimentally determined ones but, as of 1 July 2002, moved its holding of theoretical models out of the standard PDB archive to a separate ftp site and then, as of October 15, 2006, stopped accepting any new ones. As of July 2015 there were only 1358 models on the ftp site so, with such a small number, it is unlikely that one’s protein of interest will be among them.

The alternative is to build a homology model oneself, and there are various servers that will perform the process largely, or completely, automatically. The best-known is SWISS-MODEL [42]. This accepts a protein sequence and will return a 3D model if it is able to build one. More advanced users can submit multiple sequence alignment s and manually refine the final model. It is important to remember that any homology-built model will, at best, be imperfect and at worst totally misleading—particularly if one or more of the structural models that act as a template for the model contain errors. So a key part of SWISS-MODEL are the various validation checks applied to each model to provide the user with an idea of its likely quality.

Table 2i shows a list of automated homology modeling Web servers.

Table 2 Homology model servers
Full size table

Aside from building a model yourself, it may be possible to download a ready-built, off-the-shelf one. The SWISS-MODEL Repository [43] contained over three million models in July 2015, each accessible by its UniProt accession number or identifier. Similarly ModBase [44] contains a large number of precomputed models for sequences in the SwissProt and TrEMBL databases—34 million models for 5.7 million proteins in July 2015. Table 2iii gives the URLs and references for these servers.

4.2 Threading Servers

What if there is no sufficiently similar protein of known structure and thus no possibility of building a homology model? In these cases, it is sometimes necessary to resort to desperate measures such as secondary structure prediction and fold recognition, or “threading.” The results from these methods need to be treated with extreme care. Occasionally, these methods approximate the right answer—usually for small, single-domain proteins where they may produce topologically near correct models [45]—and they are improving all the time [45], but perhaps should only be used only as a last resort.

4.3 Obsolete Entries

As experimental methods improve, better data sets are collected or earlier errors are detected, so some structural models in the PDB become obsolete. Many are replaced by improved structural models, whereas others are simply quietly withdrawn. None of these obsolete entries disappear entirely, though. Some of the atlases mentioned above include the obsolete entries together with the current ones. The RCSB website provides a full list at: http://www.rcsb.org/pdb/home/obs.do.

5 Fold Databases

5.1 Classification Schemes

In 2006, it was estimated that there are around 900 known fold groups [46]. Many proteins comprise more than one structural domain, with each domain being described by its own fold and often able to fold independently of the rest of the protein . There have been a number of efforts to classify protein domains in a hierarchical manner. The two current market leaders in this field are the SCOP and CATH hierarchical classification systems (see Table 3i). In CATH, protein structure s are classified using a combination of automated and manual procedures, with four major levels in the hierarchy: Class, Architecture, Topology (fold family) and Homologous superfamily [31, 47]. In SCOP the classification is more manual, although some automated methods are employed. Comparisons between the two classification schemes have shown there to be much in common, although there are differences, primarily in how the structures are chopped into domains [48].

Table 3 Fold classification and comparison servers
Full size table

However, it appears that protein folds are not the discrete units that these classification schemes might imply, but rather that protein structure space is a continuum [49] and folds can lose core element by a process of “domain atrophy” [50]. Nevertheless, the two databases are very valuable resources because they group domains by their evolutionary relationships even where this is not apparent from any similarities in the sequences.

5.2 Fold Comparison

Often a given structural domain is associated with a specific biological function. However, the so-called superfolds, which are more common than other folds, tend to be responsible for a wide range of functions [51]. There are a large number of Web servers, such as PDBeFold mentioned above, that can identify all proteins sharing a given protein ’s fold. Each server has a different algorithm or a different way of assessing the significance of a match. Table 3ii lists a selection of the more popular servers. A fuller list, together with brief descriptions of the algorithms and a comparison between them, can be found in various comparisons that have been made between them [52, 53].

6 Miscellaneous Databases

6.1 Selection of Data Sets

For any bioinformatics analysis involving 3D structural models it is important to get a valid and representative data set of models of as high a quality as possible. To help in this process there are various servers that allow you to obtain such lists based on various selection criteria. Table 4 lists several such servers.

Table 4 Selection of data sets
Full size table

6.2 Uppsala Electron Density Server (EDS) and PDB _REDO

As has been mentioned a couple of times already, a key aspect of any structural model is how reliably it represents the protein in question. A poor quality model limits what structural or functional conclusions can be drawn from it. For X-ray models, in addition to the geometrical checks mentioned in passing above, the most useful guide to reliability is how well the model agrees with the experimental data on which it was based. The Uppsala Electron Density Server, EDS [22], displays the electron density map s for PDB entries for which the experimental structure factors are available. The server also provides various useful statistics about the models. For example, the plots of the real-space R-factor (RSR) indicate how well each residue fits its electron density; any tall red spikes are regions to be wary of. Other useful plots include: the occupancy-weighted average temperature factor and a Z-score associated with the residue’s RSR for the given resolution . The latter is used in the wwPDB ’s quality slider (see Fig. 3).

The above calculations require the original experimental data. Another use for the data is to rerefine the structural models. As refinement methods and software improve, so it is possible to revisit structural models solved in the past and rerefine them to, possibly, get better models. A server devoted to such improvement is PDB _REDO [54] (http://www.cmbi.ru.nl/pdb_redo). This provides validation measures before and after the new refinement showing the degree of improvement of the model.

6.3 Curiosities

Finally, there are various sites which deal with slightly more offbeat aspects of protein structure . Some are included in Table 5. A couple detect knots in protein folds: Protein Knots and the pKnot Web server. The former lists 44 PDB entries containing knotted proteins , classified according to the type of knot. Another interesting site, which can while away part of an afternoon, is the Database of Macromolecular Movement which holds many movies showing proteins in motion. Also included is a “Morph Server” which will produce 2D and 3D animations by interpolating between two submitted protein conformations—very useful for producing animations for presentations or websites.

Table 5 Miscellaneous servers
Full size table

7 Summary

This chapter describes some of the more generally useful protein structure databases. There are many, many more that are not mentioned. Some are very small and specialized, such as the so-called “hobby” databases, created by a single researcher and lovingly crafted and conscientiously updated—until, that is, the funding runs out, or the researcher moves on to another post and the database is abandoned and neglected. The larger and more widely used databases have better resources to keep them ticking over, but tend to suffer from a great deal of duplication and overlap. This can be seen in the large numbers of PDB atlases and fold comparison servers. Perhaps one day, a single server of each type will emerge combining the finer aspects of all others to make life a lot easier for the end users of the data.