Protein family clustering for structural genomics

A major goal of structural genomics is the provision of a structural template for a large fraction of protein domains. The magnitude of this task depends on the number and nature of protein sequence families. With a large number of bacterial genomes now fully sequenced, it is possible to obtain improved estimates of the number and diversity of families in that kingdom. We have used an automated clustering procedure to group all sequences in a set of genomes into protein families. Bench-marking shows the clustering method is sensitive at detecting remote family members, and has a low level of false positives. This comprehensive protein family set has been used to address the following questions. (1) What is the structure coverage for currently known families? (2) How will the number of known apparent families grow as more genomes are sequenced? (3) What is a practical strategy for maximizing structure coverage in future? Our study indicates that approximately 20% of known families with three or more members currently have a representative structure. The study indicates also that the number of apparent protein families will be considerably larger than previously thought: We estimate that, by the criteria of this work, there will be about 250,000 protein families when 1000 microbial genomes have been sequenced. However, the vast majority of these families will be small, and it will be possible to obtain structural templates for 70-80% of protein domains with an achievable number of representative structures, by systematically sampling the larger families.     

Defining the fold space of membrane proteins: the CAMPS database

Recent progress in structure determination techniques has led to a significant growth in the number of known membrane protein structures, and the first structural genomics projects focusing on membrane proteins have been initiated, warranting an investigation of appropriate bioinformatics strategies for optimal structural target selection for these molecules. What determines a membrane protein fold? How many membrane structures need to be solved to provide sufficient structural coverage of the membrane protein sequence space? We present the CAMPS database (Computational Analysis of the Membrane Protein Space) containing almost 45,000 proteins with three or more predicted transmembrane helices (TMH) from 120 bacterial species. This large set of membrane proteins was subjected to single‐linkage clustering using only sequence alignments covering at least 40% of the TMH present in a given family. This process yielded 266 sequence clusters with at least 15 members, roughly corresponding to membrane structural folds, sufficiently structurally homogeneous in terms of the variation of TMH number between individual sequences. These clusters were further subdivided into functionally homogeneous subclusters according to the COG (Clusters of Orthologous Groups) system as well as more stringently defined families sharing at least 30% identity. The CAMPS sequence clusters are thus designed to reflect three main levels of interest for structural genomics: fold, function, and modeling distance. We present a library of Hidden Markov Models (HMM) derived from sequence alignments of TMH at these three levels of sequence similarity. Given that 24 out of 266 clusters corresponding to membrane folds already have associated known structures, we estimate that 242 additional new structures, one for each remaining cluster, would provide structural coverage at the fold level of roughly 70% of prokaryotic membrane proteins belonging to the currently most populated families. Proteins 2006. © 2006 Wiley‐Liss, Inc.     

How many protein-protein interactions types exist in nature?

"Protein quaternary structure universe" refers to the ensemble of all protein-protein complexes across all organisms in nature. The number of quaternary folds thus corresponds to the number of ways proteins physically interact with other proteins. This study focuses on answering two basic questions: Whether the number of protein-protein interactions is limited and, if yes, how many different quaternary folds exist in nature. By all-to-all sequence and structure comparisons, we grouped the protein complexes in the protein data bank (PDB) into 3,629 families and 1,761 folds. A statistical model was introduced to obtain the quantitative relation between the numbers of quaternary families and quaternary folds in nature. The total number of possible protein-protein interactions was estimated around 4,000, which indicates that the current protein repository contains only 42% of quaternary folds in nature and a full coverage needs approximately a quarter century of experimental effort. The results have important implications to the protein complex structural modeling and the structure genomics of protein-protein interactions.     

Comprehensive genome analysis of 203 genomes provides structural genomics with new insights into protein family space

We present an analysis of 203 completed genomes in the Gene3D resource (including 17 eukaryotes), which demonstrates that the number of protein families is continually expanding over time and that singleton-sequences appear to be an intrinsic part of the genomes. A significant proportion of the proteomes can be assigned to fewer than 6000 well-characterized domain families with the remaining domain-like regions belonging to a much larger number of small uncharacterized families that are largely species specific. Our comprehensive domain annotation of 203 genomes enables us to provide more accurate estimates of the number of multi-domain proteins found in the three kingdoms of life than previous calculations. We find that 67% of eukaryotic sequences are multi-domain compared with 56% of sequences in prokaryotes. By measuring the domain coverage of genome sequences, we show that the structural genomics initiatives should aim to provide structures for less than a thousand structurally uncharacterized Pfam families to achieve reasonable structural annotation of the genomes. However, in large families, additional structures should be determined as these would reveal more about the evolution of the family and enable a greater understanding of how function evolves.     

Implications of structural genomics target selection strategies: Pfam5000, whole genome, and random approaches

Structural genomics is an international effort to determine the three-dimensional shapes of all important biological macromolecules, with a primary focus on proteins. Target proteins should be selected according to a strategy that is medically and biologically relevant, of good value, and tractable. As an option to consider, we present the "Pfam5000" strategy, which involves selecting the 5000 most important families from the Pfam database as sources for targets. We compare the Pfam5000 strategy to several other proposed strategies that would require similar numbers of targets. These strategies include complete solution of several small to moderately sized bacterial proteomes, partial coverage of the human proteome, and random selection of approximately 5000 targets from sequenced genomes. We measure the impact that successful implementation of these strategies would have upon structural interpretation of the proteins in Swiss-Prot, TrEMBL, and 131 complete proteomes (including 10 of eukaryotes) from the Proteome Analysis database at the European Bioinformatics Institute (EBI). Solving the structures of proteins from the 5000 largest Pfam families would allow accurate fold assignment for approximately 68% of all prokaryotic proteins (covering 59% of residues) and 61% of eukaryotic proteins (40% of residues). More fine-grained coverage that would allow accurate modeling of these proteins would require an order of magnitude more targets. The Pfam5000 strategy may be modified in several ways, for example, to focus on larger families, bacterial sequences, or eukaryotic sequences; as long as secondary consideration is given to large families within Pfam, coverage results vary only slightly. In contrast, focusing structural genomics on a single tractable genome would have only a limited impact in structural knowledge of other proteomes: A significant fraction (about 30-40% of the proteins and 40-60% of the residues) of each proteome is classified in small families, which may have little overlap with other species of interest. Random selection of targets from one or more genomes is similar to the Pfam5000 strategy in that proteins from larger families are more likely to be chosen, but substantial effort would be spent on small families.     

Coverage of protein sequence space by current structural genomics targets

By its purest definition the ultimate goal of structural genomics (SG) is the determination of the structures of all proteins encoded by genomes. Most of these will be obtained by homology modeling using the structures of a set of target proteins for experimental determination. Thanks to the open exchange of SG target information, we are able to analyze the sequences of the current target list to evaluate the extent of its coverage of protein sequence space. The presence of homologous sequences currently either in the Protein Data Bank (PDB) or among SG targets has been determined for each of the protein sequences in several organisms. In this way we are able to evaluate the coverage by existing or targeted structural data for the non-membranous parts of entire proteomes. For small bacterial proteomes such as that of H. influenzae almost all proteins have homologous sequences among SG targets or in the PDB. There is significantly lower coverage for more complex organisms, such as C. elegans. We have mapped the SG target list onto the ProtoMap clustering of protein sequences. Clusters occupied by SG targets represent over 150,000 protein sequences, which is approximately 44% of the total protein sequences classified by ProtoMap. The mapping of SG targets also enables an evaluation of the degree of overlap within the target list. An SG target typically occupies a ProtoMap cluster with more than six other homologous targets.     

Pfam 10 years on: 10,000 families and still growing

Classifications of proteins into groups of related sequences are in some respects like a periodic table for biology, allowing us to understand the underlying molecular biology of any organism. Pfam is a large collection of protein domains and families. Its scientific goal is to provide a complete and accurate classification of protein families and domains. The next release of the database will contain over 10,000 entries, which leads us to reflect on how far we are from completing this work. Currently Pfam matches 72% of known protein sequences, but for proteins with known structure Pfam matches 95%, which we believe represents the likely upper bound. Based on our analysis a further 28,000 families would be required to achieve this level of coverage for the current sequence database. We also show that as more sequences are added to the sequence databases the fraction of sequences that Pfam matches is reduced, suggesting that continued addition of new families is essential to maintain its relevance.     

Estimating the number of protein folds and families from complete genome data

Using the data on proteins encoded in complete genomes, combined with a rigorous theory of the sampling process, we estimate the total number of protein folds and families, as well as the number of folds and families in each genome. The total number of folds in globular, water- soluble proteins is estimated at about 1000, with structural information currently available for about one-third of the number. The sequenced genomes of unicellular organisms encode from approximately 25%, for the minimal genomes of the Mycoplasmas, to 70-80% for larger genomes, such as Escherichia coli and yeast, of the total number of folds. The number of protein families with significant sequence conservation was estimated to be between 4000 and 7000, with structures available for about 20% of these.     

Comparing function and structure between entire proteomes

More than 30 organisms have been sequenced entirely. Here, we applied a variety of simple bioinformatics tools to analyze 29 proteomes for representatives from all three kingdoms: eukaryotes, prokaryotes, and archaebacteria. We confirmed that eukaryotes have relatively more long proteins than prokaryotes and archaes, and that the overall amino acid composition is similar among the three. We predicted that approximately 15%-30% of all proteins contained transmembrane helices. We could not find a correlation between the content of membrane proteins and the complexity of the organism. In particular, we did not find significantly higher percentages of helical membrane proteins in eukaryotes than in prokaryotes or archae. However, we found more proteins with seven transmembrane helices in eukaryotes and more with six and 12 transmembrane helices in prokaryotes. We found twice as many coiled-coil proteins in eukaryotes (10%) as in prokaryotes and archaes (4%-5%), and we predicted approximately 15%-25% of all proteins to be secreted by most eukaryotes and prokaryotes. Every tenth protein had no known homolog in current databases, and 30%-40% of the proteins fell into structural families with >100 members. A classification by cellular function verified that eukaryotes have a higher proportion of proteins for communication with the environment. Finally, we found at least one homolog of experimentally known structure for approximately 20%-45% of all proteins; the regions with structural homology covered 20%-30% of all residues. These numbers may or may not suggest that there are 1200-2600 folds in the universe of protein structures. All predictions are available at http://cubic.bioc.columbia.edu/genomes.     

Structural genomics is the largest contributor of novel structural leverage

The Protein Structural Initiative (PSI) at the US National Institutes of Health (NIH) is funding four large-scale centers for structural genomics (SG). These centers systematically target many large families without structural coverage, as well as very large families with inadequate structural coverage. Here, we report a few simple metrics that demonstrate how successfully these efforts optimize structural coverage: while the PSI-2 (2005-now) contributed more than 8% of all structures deposited into the PDB, it contributed over 20% of all novel structures (i.e. structures for protein sequences with no structural representative in the PDB on the date of deposition). The structural coverage of the protein universe represented by today’s UniProt (v12.8) has increased linearly from 1992 to 2008; structural genomics has contributed significantly to the maintenance of this growth rate. Success in increasing novel leverage (defined in Liu et al. in Nat Biotechnol 25:849–851, 2007) has resulted from systematic targeting of large families. PSI’s per structure contribution to novel leverage was over 4-fold higher than that for non-PSI structural biology efforts during the past 8 years. If the success of the PSI continues, it may just take another ~15 years to cover most sequences in the current UniProt database.     

Families of membranous proteins can be characterized by the amino acid composition of their transmembrane domains

SUMMARY: In eukaryotes, membranous proteins account for 20-30% of the proteome. Most of these proteins contain one or more transmembrane (TM) domains. These are short segments that transverse the bilayer lipid membrane. Various properties of the TM domains, such as their number, their topology and their arrangement within the membrane, are closely related to the protein's cellular functions. The properties of the TM domains also determine the cellular targeting and localization of these proteins. It is not known, however, whether the information encoded by TM domains suffices for the purpose of classifying proteins into their functional families. This is the question we address here. We introduce an algorithm that creates a profile of each functional family of membranous proteins based only on the amino acid composition of their TM domains. This is complemented by a classifier program for each such family (to determine whether a given protein belongs to it or not). We find that in most instances TM domains contain enough information to allow an accurate discrimination of approximately 80% sensitivity and approximately 90% specificity among unrelated polytopic functional families with the same number of TM domains. SUPPLEMENTARY INFORMATION: Available at www.protonet.cs.huji.ac.il/TM/     

Completeness in structural genomics

Structural genomics has the goal of obtaining useful, three-dimensional models of all proteins by a combination of experimental structure determination and comparative model building. We evaluate different strategies for optimizing information return on effort. The strategy that maximizes structural coverage requires about seven times fewer structure determinations compared with the strategy in which targets are selected at random. With a choice of reasonable model quality and the goal of 90% coverage, we extrapolate the estimate of the total effort of structural genomics. It would take approximately 16,000 carefully selected structure determinations to construct useful atomic models for the vast majority of all proteins. In practice, unless there is global coordination of target selection, the total effort will likely increase by a factor of three. The task can be accomplished within a decade provided that selection of targets is highly coordinated and significant funding is available.     

Functional coverage of the human genome by existing structures, structural genomics targets, and homology models

The bias in protein structure and function space resulting from experimental limitations and targeting of particular functional classes of proteins by structural biologists has long been recognized, but never continuously quantified. Using the Enzyme Commission and the Gene Ontology classifications as a reference frame, and integrating structure data from the Protein Data Bank (PDB), target sequences from the structural genomics projects, structure homology derived from the SUPERFAMILY database, and genome annotations from Ensembl and NCBI, we provide a quantified view, both at the domain and whole-protein levels, of the current and projected coverage of protein structure and function space relative to the human genome. Protein structures currently provide at least one domain that covers 37% of the functional classes identified in the genome; whole structure coverage exists for 25% of the genome. If all the structural genomics targets were solved (twice the current number of structures in the PDB), it is estimated that structures of one domain would cover 69% of the functional classes identified and complete structure coverage would be 44%. Homology models from existing experimental structures extend the 37% coverage to 56% of the genome as single domains and 25% to 31% for complete structures. Coverage from homology models is not evenly distributed by protein family, reflecting differing degrees of sequence and structure divergence within families. While these data provide coverage, conversely, they also systematically highlight functional classes of proteins for which structures should be determined. Current key functional families without structure representation are highlighted here; updated information on the "most wanted list" that should be solved is available on a weekly basis from http://function.rcsb.org:8080/pdb/function_distribution/index.html.     

Consensus predictions of membrane protein topology

We have explored the possibility that consensus predictions of membrane protein topology might provide a means to estimate the reliability of a predicted topology. Using five current topology prediction methods and a test set of 60 Escherichia coli inner membrane proteins with experimentally determined topologies, we find that prediction performance varies strongly with the number of methods that agree, and that the topology of nearly half of all E. coli inner membrane proteins can be predicted with high reliability (>90% correct predictions) by a simple majority-vote approach.     

Protein function space: viewing the limits or limited by our view?

Given that the number of protein functions on earth is finite, the rapid expansion of biological knowledge and the concomitant exponential increase in the number of protein sequences should, at some point, enable the estimation of the limits of protein function space. The functional coverage of protein sequences can be investigated using computational methods, especially given the massive amount of data being generated by large-scale environmental sequencing (metagenomics). In completely sequenced genomes, the fraction of proteins to which at least some functional features can be assigned has recently risen to as much as approximately 85%. Although this fraction is more uncertain in metagenomics surveys, because of environmental complexities and differences in analysis protocols, our global knowledge of protein functions still appears to be considerable. However, when we consider protein families, continued sequencing seems to yield an ever-increasing number of novel families. Until we reconcile these two views, the limits of protein space will remain obscured.     

Camps 2.0: exploring the sequence and structure space of prokaryotic, eukaryotic, and viral membrane proteins

Structural bioinformatics of membrane proteins is still in its infancy, and the picture of their fold space is only beginning to emerge. Because only a handful of three-dimensional structures are available, sequence comparison and structure prediction remain the main tools for investigating sequence-structure relationships in membrane protein families. Here we present a comprehensive analysis of the structural families corresponding to α-helical membrane proteins with at least three transmembrane helices. The new version of our CAMPS database (CAMPS 2.0) covers nearly 1300 eukaryotic, prokaryotic, and viral genomes. Using an advanced classification procedure, which is based on high-order hidden Markov models and considers both sequence similarity as well as the number of transmembrane helices and loop lengths, we identified 1353 structurally homogeneous clusters roughly corresponding to membrane protein folds. Only 53 clusters are associated with experimentally determined three-dimensional structures, and for these clusters CAMPS is in reasonable agreement with structure-based classification approaches such as SCOP and CATH. We therefore estimate that ～1300 structures would need to be determined to provide a sufficient structural coverage of polytopic membrane proteins. CAMPS 2.0 is available at http://webclu.bio.wzw.tum.de/CAMPS2.0/.     

Target space for structural genomics revisited

MOTIVATION: Structural genomics eventually aims at determining structures for all proteins. However, in the beginning experimentalists are likely to focus on globular proteins to achieve a rapid basic coverage of protein sequence space. How many proteins will structural genomics have to target? How many proteins will be excluded since we already have structural information for these or since they are not globular? We have to answer these questions in the context of our target selection for the North-East Structural Genomics Consortium (NESG). RESULTS: We estimated that structural information is available for about 6-38% of all proteins; 6% if we require high accuracy in comparative modelling, 38% if we are satisfied with having a rough idea about the fold. Excluding all regions that are not globular, we found that structural genomics may have to target about 48% of all proteins. This corresponded to a similar percentage of residues of the entire proteomes (52%). We explored a number of different strategies to cluster protein space in order to find the number of families representing these 48% of structurally unknown proteins. For the subset of all entirely sequenced eukaryotes, we found over 18 000 fragment clusters each of which may be a suitable target for structural genomics. AVAILABILITY: All data are available from the authors, most results are summarized at: http://cubic.bioc.columbia.edu/genomes/RES/2002_bioinformatics/     

Emergence of protein fold families through rational design

Diverse proteins with similar structures are grouped into families of homologs and analogs, if their sequence similarity is higher or lower, respectively, than 20%–30%. It was suggested that protein homologs and analogs originate from a common ancestor and diverge in their distinct evolutionary time scales, emerging as a consequence of the physical properties of the protein sequence space. Although a number of studies have determined key signatures of protein family organization, the sequence-structure factors that differentiate the two evolution-related protein families remain unknown. Here, we stipulate that subtle structural changes, which appear due to accumulating mutations in the homologous families, lead to distinct packing of the protein core and, thus, novel compositions of core residues. The latter process leads to the formation of distinct families of homologs. We propose that such differentiation results in the formation of analogous families. To test our postulate, we developed a molecular modeling and design toolkit, Medusa, to computationally design protein sequences that correspond to the same fold family. We find that analogous proteins emerge when a backbone structure deviates only 1–2 Å root-mean-square deviation from the original structure. For close homologs, core residues are highly conserved. However, when the overall sequence similarity drops to ~25%–30%, the composition of core residues starts to diverge, thereby forming novel families of protein homologs. This direct observation of the formation of protein homologs within a specific fold family supports our hypothesis. The conservation of amino acids in designed sequences recapitulates that of the naturally occurring sequences, thereby validating our computational design methodology.     

Structural clusters of evolutionary trace residues are statistically significant and common in proteins

Given the massive increase in the number of new sequences and structures, a critical problem is how to integrate these raw data into meaningful biological information. One approach, the Evolutionary Trace, or ET, uses phylogenetic information to rank the residues in a protein sequence by evolutionary importance and then maps those ranked at the top onto a representative structure. If these residues form structural clusters, they can identify functional surfaces such as those involved in molecular recognition. Now that a number of examples have shown that ET can identify binding sites and focus mutational studies on their relevant functional determinants, we ask whether the method can be improved so as to be applicable on a large scale. To address this question, we introduce a new treatment of gaps resulting from insertions and deletions, which streamlines the selection of sequences used as input. We also introduce objective statistics to assess the significance of the total number of clusters and of the size of the largest one. As a result of the novel treatment of gaps, ET performance improves measurably. We find evolutionarily privileged clusters that are significant at the 5% level in 45 out of 46 (98%) proteins drawn from a variety of structural classes and biological functions. In 37 of the 38 proteins for which a protein-ligand complex is available, the dominant cluster contacts the ligand. We conclude that spatial clustering of evolutionarily important residues is a general phenomenon, consistent with the cooperative nature of residues that determine structure and function. In practice, these results suggest that ET can be applied on a large scale to identify functional sites in a significant fraction of the structures in the protein databank (PDB). This approach to combining raw sequences and structure to obtain detailed insights into the molecular basis of function should prove valuable in the context of the Structural Genomics Initiative.     

Strategies for high-throughput comparative modeling: applications to leverage analysis in structural genomics and protein family organization

The technological breakthroughs in structural genomics were designed to facilitate the solution of a sufficient number of structures, so that as many protein sequences as possible can be structurally characterized with the aid of comparative modeling. The leverage of a solved structure is the number and quality of the models that can be produced using the structure as a template for modeling and may be viewed as the "currency" with which the success of a structural genomics endeavor can be measured. Moreover, the models obtained in this way should be valuable to all biologists. To this end, at the Northeast Structural Genomics Consortium (NESG), a modular computational pipeline for automated high-throughput leverage analysis was devised and used to assess the leverage of the 186 unique NESG structures solved during the first phase of the Protein Structure Initiative (January 2000 to July 2005). Here, the results of this analysis are presented. The number of sequences in the nonredundant protein sequence database covered by quality models produced by the pipeline is approximately 39,000, so that the average leverage is approximately 210 models per structure. Interestingly, only 7900 of these models fulfill the stringent modeling criterion of being at least 30% sequence-identical to the corresponding NESG structures. This study shows how high-throughput modeling increases the efficiency of structure determination efforts by providing enhanced coverage of protein structure space. In addition, the approach is useful in refining the boundaries of structural domains within larger protein sequences, subclassifying sequence diverse protein families, and defining structure-based strategies specific to a particular family.