GTOP: a database of protein structures predicted from genome sequences

The TetR-type regulator AtsR is involved in multidrug response in Corynebacterium glutamicum

Background

The TetR (tetracycline repressor) family is one of the major transcription factor families that regulate expression of genes involved in bacterial antimicrobial resistance systems. NCgl0886 protein, designated as AtsR, is a member of the TetR family identified in Corynebacterium glutamicum, which is conserved in several species of the genera Corynebacterium, also including the well-known pathogen C. diphtheriae. AtsR is located at no far upstream of the identically oriented ncgl0884 gene, encoding a putative multidrug efflux pump protein, and in the same operon with ncgl0887, encoding a resistance, nodulation and cell division (RND) superfamily drug exporter. However, the role of AtsR is not clearly understood.

Results

Here we showed that dimeric AtsR directly repressed the expression of the ncgl0887-atsR operon, as well as indirectly controlled the ncgl0884 transcription. Antibiotics and toxic compounds induced the expression of ncgl0887-atsR operon. A perfect palindromic motif (5΄-TGCAA-N₂-TTGCA-3΄; 12 bp) was identified in the upstream region of ncgl0887-atsR operon. Electrophoretic mobility shift assays (EMSAs) demonstrated specific binding of AtsR to this motif, and hydrogen peroxide (H₂O₂) blocked binding. H₂O₂ oxidized cysteine residues to form Cys123-Cys187 intermolecular disulfide bonds between two subunits in AtsR dimer, which altered its DNA-binding characteristics and caused its dissociation, thereby leading to derepression of the drug efflux protein. Deletion of ncgl0884 and ncgl0887 increased the susceptibilities of C. glutamicum for several toxic compounds, but overexpression of atsR decreased the drug tolerance of C. glutamicum.

Conclusions

Our study revealed that AtsR was a redox regulator that sensed oxidative stress via thiol modification. The results obtained here will contribute to our understanding of the drug response mechanism not only in C. glutamicum but also in the related bacteria C. diphtheriae.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12934-022-01850-0.

Dual pathways of tRNA hydroxylation ensure efficient translation by expanding decoding capability

In bacterial tRNAs, 5-carboxymethoxyuridine (cmo⁵U) and its derivatives at the first position of the anticodon facilitate non-Watson-Crick base pairing with guanosine and pyrimidines at the third positions of codons, thereby expanding decoding capabilities. However, their biogenesis and physiological roles remained to be investigated. Using reverse genetics and comparative genomics, we identify two factors responsible for 5-hydroxyuridine (ho⁵U) formation, which is the first step of the cmo⁵U synthesis: TrhP (formerly known as YegQ), a peptidase U32 family protein, is involved in prephenate-dependent ho⁵U formation; and TrhO (formerly known as YceA), a rhodanese family protein, catalyzes oxygen-dependent ho⁵U formation and bypasses cmo⁵U biogenesis in a subset of tRNAs under aerobic conditions. E. coli strains lacking both trhP and trhO exhibit a temperature-sensitive phenotype, and decode codons ending in G (GCG and UCG) less efficiently than the wild-type strain. These findings confirm that tRNA hydroxylation ensures efficient decoding during protein synthesis.

Large-scale aggregation analysis of eukaryotic proteins reveals an involvement of intrinsically disordered regions in protein folding

A subset of the proteome is prone to aggregate formation, which is prevented by chaperones in the cell. To investigate whether the basic principle underlying the aggregation process is common in prokaryotes and eukaryotes, we conducted a large-scale aggregation analysis of ~500 cytosolic budding yeast proteins using a chaperone-free reconstituted translation system, and compared the obtained data with that of ~3,000 Escherichia coli proteins reported previously. Although the physicochemical properties affecting the aggregation propensity were generally similar in yeast and E. coli proteins, the susceptibility of aggregation in yeast proteins were positively correlated with the presence of intrinsically disordered regions (IDRs). Notably, the aggregation propensity was not significantly changed by a removal of IDRs in model IDR-containing proteins, suggesting that the properties of ordered regions in these proteins are the dominant factors for aggregate formation. We also found that the proteins with longer IDRs were disfavored by E. coli chaperonin GroEL/ES, whereas both bacterial and yeast Hsp70/40 chaperones have a strong aggregation-prevention effect even for proteins possessing IDRs. These results imply that a key determinant to discriminate the eukaryotic proteomes from the prokaryotic proteomes in terms of protein folding would be the attachment of IDRs.

VaProS: a database-integration approach for protein/genome information retrieval

Life science research now heavily relies on all sorts of databases for genome sequences, transcription, protein three-dimensional (3D) structures, protein-protein interactions, phenotypes and so forth. The knowledge accumulated by all the omics research is so vast that a computer-aided search of data is now a prerequisite for starting a new study. In addition, a combinatory search throughout these databases has a chance to extract new ideas and new hypotheses that can be examined by wet-lab experiments. By virtually integrating the related databases on the Internet, we have built a new web application that facilitates life science researchers for retrieving experts' knowledge stored in the databases and for building a new hypothesis of the research target. This web application, named VaProS, puts stress on the interconnection between the functional information of genome sequences and protein 3D structures, such as structural effect of the gene mutation. In this manuscript, we present the notion of VaProS, the databases and tools that can be accessed without any knowledge of database locations and data formats, and the power of search exemplified in quest of the molecular mechanisms of lysosomal storage disease. VaProS can be freely accessed at http://p4d-info.nig.ac.jp/vapros/ .

Prediction of protein-destabilizing polymorphisms by manual curation with protein structure

The relationship between sequence polymorphisms and human disease has been studied mostly in terms of effects of single nucleotide polymorphisms (SNPs) leading to single amino acid substitutions that change protein structure and function. However, less attention has been paid to more drastic sequence polymorphisms which cause premature termination of a protein’s sequence or large changes, insertions, or deletions in the sequence. We have analyzed a large set (n = 512) of insertions and deletions (indels) and single nucleotide polymorphisms causing premature termination of translation in disease-related genes. Prediction of protein-destabilization effects was performed by graphical presentation of the locations of polymorphisms in the protein structure, using the Genomes TO Protein (GTOP) database, and manual annotation with a set of specific criteria. Protein-destabilization was predicted for 44.4% of the nonsense SNPs, 32.4% of the frameshifting indels, and 9.1% of the non-frameshifting indels. A prediction of nonsense-mediated decay allowed to infer which truncated proteins would actually be translated as defective proteins. These cases included the proteins linked to diseases inherited dominantly, suggesting a relation between these diseases and toxic aggregation. Our approach would be useful in identifying potentially aggregation-inducing polymorphisms that may have pathological effects.

Mutational robustness of gene regulatory networks

Mutational robustness of gene regulatory networks refers to their ability to generate constant biological output upon mutations that change network structure. Such networks contain regulatory interactions (transcription factor – target gene interactions) but often also protein-protein interactions between transcription factors. Using computational modeling, we study factors that influence robustness and we infer several network properties governing it. These include the type of mutation, i.e. whether a regulatory interaction or a protein-protein interaction is mutated, and in the case of mutation of a regulatory interaction, the sign of the interaction (activating vs. repressive). In addition, we analyze the effect of combinations of mutations and we compare networks containing monomeric with those containing dimeric transcription factors. Our results are consistent with available data on biological networks, for example based on evolutionary conservation of network features. As a novel and remarkable property, we predict that networks are more robust against mutations in monomer than in dimer transcription factors, a prediction for which analysis of conservation of DNA binding residues in monomeric vs. dimeric transcription factors provides indirect evidence.

Differences in dinucleotide frequencies of thermophilic genes encoding water soluble and membrane proteins

The occurrence frequencies of the dinucleotides of genes of three thermophilic and three mesophilic species from both archaea and eubacteria were investigated in this study. The genes encoding water soluble proteins were rich in the dinucleotides of purine dimers, whereas the genes encoding membrane proteins were rich in pyrimidine dimers. The dinucleotides of purine dimers are the counterparts of pyrimidine dimers in a double-stranded DNA. The purine/pyrimidine dimers were favored in the thermophiles but not in the mesophiles, based on comparisons of observed and expected frequencies. This finding is in agreement with our previous study which showed that purine/pyrimidine dimers are positive factors that increase the thermal stability of DNA. The dinucleotides AA, AG, and GA are components of the codons of charged residues of Glu, Asp, Lys, and Arg, and the dinucleotides TT, CT, and TC are components of the codons of hydrophobic residues of Leu, Ile, and Phe. This is consistent with the suitabilities of the different amino acid residues for water soluble and membrane proteins. Our analysis provides a picture of how thermophilic species produce water soluble and membrane proteins with distinctive characters: the genes encoding water soluble proteins use DNA sequences rich in purine dimers, and the genes encoding membrane proteins use DNA sequences rich in pyrimidine dimers on the opposite strand.

SLIDER: a generic metaheuristic for the discovery of correlated motifs in protein-protein interaction networks

Correlated motif mining (cmm) is the problem of finding overrepresented pairs of patterns, called motifs, in sequences of interacting proteins. Algorithmic solutions for cmm thereby provide a computational method for predicting binding sites for protein interaction. In this paper, we adopt a motif-driven approach where the support of candidate motif pairs is evaluated in the network. We experimentally establish the superiority of the Chi-square-based support measure over other support measures. Furthermore, we obtain that cmm is an np-hard problem for a large class of support measures (including Chi-square) and reformulate the search for correlated motifs as a combinatorial optimization problem. We then present the generic metaheuristic slider which uses steepest ascent with a neighborhood function based on sliding motifs and employs the Chi-square-based support measure. We show that slider outperforms existing motif-driven cmm methods and scales to large protein-protein interaction networks. The slider-implementation and the data used in the experiments are available on http://bioinformatics.uhasselt.be.

A visualization of 3D proteome universe: mapping of a proteome ensemble into 3D space based on the protein-structure composition

To visualize a bird's-eye view of an ensemble of proteomes for various species, we recently developed a novel method of mapping a proteome ensemble into Three-Dimensional (3D) vector space. In this study, the "proteome" is defined as the entire set of all proteins encoded in a genome sequence, and these proteins were dealt with at the level of the SCOP Fold. First, we represented the proteome of a species s by a 1053-dimensional vector x(s), where its length ∣x(s)∣ represents the overall amount of all the SCOP Folds in the proteome, and its unit vector x(s)/∣x(s)∣ represents the relative composition of the SCOP Folds in the proteome and the size of the dimension, 1053, is the number of all possible Folds in the proteome ensemble given. Second, we mapped the vector x(s) to the 3D vector y(s), based on the two simple principles: (1) ∣y(s)∣=∣x(s)∣, and (2) the angle between y(s) and y(t) maximally correlates with the angle between x(s) and x(t). We applied to the mapping of a proteome ensemble for 456 species, which were retrieved from the Genomes TO Protein structures and functions (GTOP) database. As a result, we succeeded in the mapping in that the properties of the 1053-dimensional vectors were quantitatively conserved in the 3D vectors. Particularly, the angles between vectors before and after the mapping highly correlated with each other (correlation coefficients were 0.95-0.96). This new mapping method will allow researchers to intuitively interpret the visual information presented in the maps in a highly effective manner.

A novel RNA-recognition-motif protein is required for premeiotic G1/S-phase transition in rice (Oryza sativa L.)

The molecular mechanism for meiotic entry remains largely elusive in flowering plants. Only Arabidopsis SWI1/DYAD and maize AM1, both of which are the coiled-coil protein, are known to be required for the initiation of plant meiosis. The mechanism underlying the synchrony of male meiosis, characteristic to flowering plants, has also been unclear in the plant kingdom. In other eukaryotes, RNA-recognition-motif (RRM) proteins are known to play essential roles in germ-cell development and meiosis progression. Rice MEL2 protein discovered in this study shows partial similarity with human proline-rich RRM protein, deleted in Azoospermia-Associated Protein1 (DAZAP1), though MEL2 also possesses ankyrin repeats and a RING finger motif. Expression analyses of several cell-cycle markers revealed that, in mel2 mutant anthers, most germ cells failed to enter premeiotic S-phase and meiosis, and a part escaped from the defect and underwent meiosis with a significant delay or continued mitotic cycles. Immunofluorescent detection revealed that T7 peptide-tagged MEL2 localized at cytoplasmic perinuclear region of germ cells during premeiotic interphase in transgenic rice plants. This study is the first report of the plant RRM protein, which is required for regulating the premeiotic G1/S-phase transition of male and female germ cells and also establishing synchrony of male meiosis. This study will contribute to elucidation of similarities and diversities in reproduction system between plants and other species.

Meiosis is a pivotal event to produce haploid spores and gametes in all sexually reproducing species and is a fundamentally different type of cell cycle from mitosis. Thus, the molecular mechanisms to switch the cell cycle from mitosis to meiosis have been studied by many researchers. In yeast and metazoans, RNA-binding proteins are known to play important roles in the post-transcriptional regulation of genes implicated in the meiotic entry and meiosis. In contrast, in the plant kingdom, the mechanisms to control the meiotic entry have largely remained elusive. In this study, we discover a novel RNA-recognition-motif (RRM) protein in rice (Oryza sativa L.), designated MEL2, and demonstrate that MEL2 is required for the faithful transition of germ cells from mitosis to meiotic cell cycle. Rice MEL2 shows partial similarity with human DAZAP1, which is an RRM protein and relates to Azoospermia syndrome in human, while there are critical structural differences between germline-specific RRM proteins of mammals and plants. Our findings will lead the molecular-biological studies of plant meiotic entry to the next steps and will enable a comparison of the systems of meiotic entry between animals and plants.

Synonymous codon usage influences the local protein structure observed

Translation of mRNA into protein is a unidirectional information flow process. Analysing the input (mRNA) and output (protein) of translation, we find that local protein structure information is encoded in the mRNA nucleotide sequence. The Coding Sequence and Structure (CSandS) database developed in this work provides a detailed mapping between over 4000 solved protein structures and their mRNA. CSandS facilitates a comprehensive analysis of codon usage over many organisms. In assigning translation speed, we find that relative codon usage is less informative than tRNA concentration. For all speed measures, no evidence was found that domain boundaries are enriched with slow codons. In fact, genes seemingly avoid slow codons around structurally defined domain boundaries. Translation speed, however, does decrease at the transition into secondary structure. Codons are identified that have structural preferences significantly different from the amino acid they encode. However, each organism has its own set of ‘significant codons’. Our results support the premise that codons encode more information than merely amino acids and give insight into the role of translation in protein folding.

Sub-AQUA: real-value quality assessment of protein structure models

Computational protein tertiary structure prediction has made significant progress over the past years. However, most of the existing structure prediction methods are not equipped with functionality to predict accuracy of constructed models. Knowing the accuracy of a structure model is crucial for its practical use since the accuracy determines potential applications of the model. Here we have developed quality assessment methods, which predict real value of the global and local quality of protein structure models. The global quality of a model is defined as the root mean square deviation (RMSD) and the LGA score to its native structure. The local quality is defined as the distance between the corresponding Calpha positions of a model and its native structure when they are superimposed. Three regression methods are employed to combine different types of quality assessment measures of models, including alignment-level scores, residue-position level scores, atomic-detailed structure level scores and composite scores. The regression models were tested on a large benchmark data set of template-based protein structure models of various qualities. In predicting RMSD and the LGA score, a combination of two terms, length-normalized SPAD, a score that assesses alignment stability by considering suboptimal alignments, and Verify3D normalized by the square of the model length shows a significant performance, achieving 97.1 and 83.6% accuracy in identifying models with an RMSD of <2 and 6 A, respectively. For predicting the local quality of models, we find that a two-step approach, in which the global RMSD predicted in the first step is further combined with the other terms, can dramatically increase the accuracy. Finally, the developed regression equations are applied to assess the quality of structure models of whole E. coli proteome.

HUMAN FAMSD-BASE: high quality protein structure model database for the human genome using the FAMSD homology modeling method

Almost all proteins express their biological functions through the structural conformation of their specific amino acid sequences. Therefore, acquiring the three-dimensional structures of proteins is very important to elucidate the role of a particular protein. We had built protein structure model databases, which is called RIKEN FAMSBASE (http://famshelp.gsc.riken.jp/famsbase/). The RIKEN FAMSBASE is a genome-wide protein structure model database that contains a large number of protein models from many organisms. The HUMAN FAMSBASE that is one part of the RIKEN FAMSBASE contains many protein models for human genes, which are significant in the pharmaceutical and medicinal fields. We have now implemented an update of the human protein modeling database consisting of 242918 constructed models against the number of 20743 human protein sequences with an improved modeling method called Full Automatic protein Modeling System Developed (FAMSD). The results of our benchmark test of the FAMSD method indicated that it has an excellent capability to pack amino acid side-chains with correct torsion angles in addition to the main-chain, while avoiding the formation of atom-atom collisions that are not found in experimental structures. This new protein structure model database for human genes, which is named HUMAN FAMSD-BASE, is open to the public as a component part of the RIKEN FAMSBASE at http://mammalia.gsc.riken.jp/human_famsd/. A significant improvement of the HUMAN FAMSD-BASE in comparison with the preceding HUMAN FAMSBASE was verified in the benchmark test of this paper. The HUMAN FAMSD-BASE will have an important impact on the progress of biological science.

DDBJ launches a new archive database with analytical tools for next-generation sequence data

The DNA Data Bank of Japan (DDBJ) (http://www.ddbj.nig.ac.jp) has collected and released 1 701 110 entries/1 116 138 614 bases between July 2008 and June 2009. A few highlighted data releases from DDBJ were the complete genome sequence of an endosymbiont within protist cells in the termite gut and Cap Analysis Gene Expression tags for human and mouse deposited from the Functional Annotation of the Mammalian cDNA consortium. In this period, we started a novel user announcement service using Really Simple Syndication (RSS) to deliver a list of data released from DDBJ on a daily basis. Comprehensive visualization of a DDBJ release data was attempted by using a word cloud program. Moreover, a new archive for sequencing data from next-generation sequencers, the ‘DDBJ Read Archive’ (DRA), was launched. Concurrently, for read data registered in DRA, a semi-automatic annotation tool called the ‘DDBJ Read Annotation Pipeline’ was released as a preliminary step. The pipeline consists of two parts: basic analysis for reference genome mapping and de novo assembly and high-level analysis of structural and functional annotations. These new services will aid users’ research and provide easier access to DDBJ databases.

Phosphorylated intrinsically disordered region of FACT masks its nucleosomal DNA binding elements

FACT is a heterodimer of SPT16 and SSRP1, which each contain several conserved regions in the primary structure. The interaction of FACT with nucleosomes induces chromatin remodeling through the combinatorial action of its distinct functional protein regions. However, there is little mechanistic insight into how these regions cooperatively contribute to FACT functions, particularly regarding the recognition of nucleosomal DNA. Here, we report the identification of novel phosphorylation sites of Drosophila melanogaster FACT (dFACT) expressed in Sf9 cells. These sites are densely concentrated in the acidic intrinsically disordered (ID) region of the SSRP1 subunit and control nucleosomal DNA binding by dFACT. This region and the adjacent segment of the HMG domain form weak electrostatic intramolecular interactions, which is reinforced by the phosphorylation, thereby blocking DNA binding competitively. Importantly, this control mechanism appears to support rapid chromatin transactions during early embryogenesis through the dephosphorylation of some sites in the maternally transmitted dSSRP1.

Development of an accurate classification system of proteins into structured and unstructured regions that uncovers novel structural domains: its application to human transcription factors

BACKGROUND

In addition to structural domains, most eukaryotic proteins possess intrinsically disordered (ID) regions. Although ID regions often play important functional roles, their accurate identification is difficult. As human transcription factors (TFs) constitute a typical group of proteins with long ID regions, we regarded them as a model of all proteins and attempted to accurately classify TFs into structural domains and ID regions. Although an extremely high fraction of ID regions besides DNA binding and/or other domains was detected in human TFs in our previous investigation, 20% of the residues were left unassigned. In this report, we exploit the generally higher sequence divergence in ID regions than in structural regions to completely divide proteins into structural domains and ID regions.

RESULTS

The new dichotomic system first identifies domains of known structures, followed by assignment of structural domains and ID regions with a combination of pre-existing tools and a newly developed program based on sequence divergence, taking un-aligned regions into consideration. The system was found to be highly accurate: its application to a set of proteins with experimentally verified ID regions had an error rate as low as 2%. Application of this system to human TFs (401 proteins) showed that 38% of the residues were in structural domains, while 62% were in ID regions. The preponderance of ID regions makes a sharp contrast to TFs of Escherichia coli (229 proteins), in which only 5% fell in ID regions. The method also revealed that 4.0% and 11.8% of the total length in human and E. coli TFs, respectively, are comprised of structural domains whose structures have not been determined.

CONCLUSION

The present system verifies that sequence divergence including information of unaligned regions is a good indicator of ID regions. The system for the first time estimates the complete fractioning of structured/un-structured regions in human TFs, also revealing structural domains without homology to known structures. These predicted novel structural domains are good targets of structural genomics. When applied to other proteins, the system is expected to uncover more novel structural domains.

Translationally optimal codons associate with structurally sensitive sites in proteins

The mistranslation-induced protein misfolding hypothesis predicts that selection should prefer high-fidelity codons at sites at which translation errors are structurally disruptive and lead to protein misfolding and aggregation. To test this hypothesis, we analyzed the relationship between codon usage bias and protein structure in the genomes of four model organisms, Escherichia coli, yeast, fly, and mouse. Using both the Mantel-Haenszel procedure, which applies to categorical data, and a newly developed association test for continuous variables, we find that translationally optimal codons associate with buried residues and also with residues at sites where mutations lead to large changes in free energy (DeltaDeltaG). In each species, only a subset of all amino acids show this signal, but most amino acids show the signal in at least one species. By repeating the analysis on a reduced data set that excludes interdomain linkers, we show that our results are not caused by an association of rare codons with solvent-accessible linker regions. Finally, we find that our results depend weakly on expression level; the association between optimal codons and buried sites exists at all expression levels, but increases in strength as expression level increases.

Bimodal protein solubility distribution revealed by an aggregation analysis of the entire ensemble of Escherichia coli proteins

Protein folding often competes with intermolecular aggregation, which in most cases irreversibly impairs protein function, as exemplified by the formation of inclusion bodies. Although it has been empirically determined that some proteins tend to aggregate, the relationship between the protein aggregation propensities and the primary sequences remains poorly understood. Here, we individually synthesized the entire ensemble of Escherichia coli proteins by using an in vitro reconstituted translation system and analyzed the aggregation propensities. Because the reconstituted translation system is chaperone-free, we could evaluate the inherent aggregation propensities of thousands of proteins in a translation-coupled manner. A histogram of the solubilities, based on data from 3,173 translated proteins, revealed a clear bimodal distribution, indicating that the aggregation propensities are not evenly distributed across a continuum. Instead, the proteins can be categorized into 2 groups, soluble and aggregation-prone proteins. The aggregation propensity is most prominently correlated with the structural classification of proteins, implying that the prediction of aggregation propensity requires structural information about the protein.

Comparative proteome analysis of psychrophilic versus mesophilic bacterial species: Insights into the molecular basis of cold adaptation of proteins

Background

Cold adapted or psychrophilic organisms grow at low temperatures, where most of other organisms cannot grow. This adaptation requires a vast array of sequence, structural and physiological adjustments. To understand the molecular basis of cold adaptation of proteins, we analyzed proteomes of psychrophilic and mesophilic bacterial species and compared the differences in amino acid composition and substitution patterns to investigate their likely association with growth temperatures.

Results

In psychrophilic bacteria, serine, aspartic acid, threonine and alanine are overrepresented in the coil regions of secondary structures, whilst glutamic acid and leucine are underrepresented in the helical regions. Compared to mesophiles, psychrophiles comprise a significantly higher proportion of amino acids that contribute to higher protein flexibility in the coil regions of proteins, such as those with tiny/small or neutral side chains. Amino acids with aliphatic, basic, aromatic and hydrophilic side chains are underrepresented in the helical regions of proteins of psychrophiles. The patterns of amino acid substitutions between the orthologous proteins of psychrophiles versus mesophiles are significantly different for several amino acids when compared to their substitutions in orthologous proteins of within the mesophiles or psychrophiles.

Conclusion

Current results provide quantitative substitution preferences (or avoidance) of amino acids that lead to the adaptation of proteins to cold temperatures. These finding would help future efforts in selecting mutations for rational design of proteins with enhanced psychrophilic properties.

Detecting clusters of mutations

Positive selection for protein function can lead to multiple mutations within a small stretch of DNA, i.e., to a cluster of mutations. Recently, Wagner proposed a method to detect such mutation clusters. His method, however, did not take into account that residues with high solvent accessibility are inherently more variable than residues with low solvent accessibility. Here, we propose a new algorithm to detect clustered evolution. Our algorithm controls for different substitution probabilities at buried and exposed sites in the tertiary protein structure, and uses random permutations to calculate accurate P values for inferred clusters. We apply the algorithm to genomes of bacteria, fly, and mammals, and find several clusters of mutations in functionally important regions of proteins. Surprisingly, clustered evolution is a relatively rare phenomenon. Only between 2% and 10% of the genes we analyze contain a statistically significant mutation cluster. We also find that not controlling for solvent accessibility leads to an excess of clusters in terminal and solvent-exposed regions of proteins. Our algorithm provides a novel method to identify functionally relevant divergence between groups of species. Moreover, it could also be useful to detect artifacts in automatically assembled genomes.

The GTOP database in 2009: updated content and novel features to expand and deepen insights into protein structures and functions

The Genomes TO Protein Structures and Functions (GTOP) database (http://spock.genes.nig.ac.jp/~genome/gtop.html) freely provides an extensive collection of information on protein structures and functions obtained by application of various computational tools to the amino acid sequences of entirely sequenced genomes. GTOP contains annotations of 3D structures, protein families, functions, and other useful data of a protein of interest in user-friendly ways to give a deep insight into the protein structure. From the initial 1999 version, GTOP has been continually updated to reap the fruits of genome projects and augmented to supply novel information, in particular intrinsically disordered regions. As intrinsically disordered regions constitute a considerable fraction of proteins and often play crucial roles especially in eukaryotes, their assignments give important additional clues to the functionality of proteins. Additionally, we have incorporated the following features into GTOP: a platform independent structural viewer, results of HMM searches against SCOP and Pfam, secondary structure predictions, color display of exon boundaries in eukaryotic proteins, assignments of gene ontology terms, search tools, and master files.

VarySysDB: a human genetic polymorphism database based on all H-InvDB transcripts

Creation of a vast variety of proteins is accomplished by genetic variation and a variety of alternative splicing transcripts. Currently, however, the abundant available data on genetic variation and the transcriptome are stored independently and in a dispersed fashion. In order to provide a research resource regarding the effects of human genetic polymorphism on various transcripts, we developed VarySysDB, a genetic polymorphism database based on 187,156 extensively annotated matured mRNA transcripts from 36,073 loci provided by H-InvDB. VarySysDB offers information encompassing published human genetic polymorphisms for each of these transcripts separately. This allows comparisons of effects derived from a polymorphism on different transcripts. The published information we analyzed includes single nucleotide polymorphisms and deletion-insertion polymorphisms from dbSNP, copy number variations from Database of Genomic Variants, short tandem repeats and single amino acid repeats from H-InvDB and linkage disequilibrium regions from D-HaploDB. The information can be searched and retrieved by features, functions and effects of polymorphisms, as well as by keywords. VarySysDB combines two kinds of viewers, GBrowse and Sequence View, to facilitate understanding of the positional relationship among polymorphisms, genome, transcripts, loci and functional domains. We expect that VarySysDB will yield useful information on polymorphisms affecting gene expression and phenotypes. VarySysDB is available at http://h-invitational.jp/varygene/.

Contact density affects protein evolutionary rate from bacteria to animals

The density of contacts or the fraction of buried sites in a protein structure is thought to be related to a protein's designability, and genes encoding more designable proteins should evolve faster than other genes. Several recent studies have tested this hypothesis but have found conflicting results. Here, we investigate how a gene's evolutionary rate is affected by its protein's contact density, considering the four species Escherichia coli, Saccharomyces cerevisiae, Drosophila melanogaster, and Homo sapiens. We find for all four species that contact density correlates positively with evolutionary rate, and that these correlations do not seem to be confounded by gene expression level. The strength of this signal, however, varies widely among species. We also study the effect of contact density on domain evolution in multidomain proteins and find that a domain's contact density influences the domain's evolutionary rate. Within the same protein, a domain with higher contact density tends to evolve faster than a domain with lower contact density. Our study provides evidence that contact density can increase evolutionary rates, and that it acts similarly on the level of entire proteins and of individual protein domains.

Ma-LMM01 infecting toxic Microcystis aeruginosa illuminates diverse cyanophage genome strategies

Cyanobacteria and their phages are significant microbial components of the freshwater and marine environments. We identified a lytic phage, Ma-LMM01, infecting Microcystis aeruginosa, a cyanobacterium that forms toxic blooms on the surfaces of freshwater lakes. Here, we describe the first sequenced freshwater cyanomyovirus genome of Ma-LMM01. The linear, circularly permuted, and terminally redundant genome has 162,109 bp and contains 184 predicted protein-coding genes and two tRNA genes. The genome exhibits no colinearity with previously sequenced genomes of cyanomyoviruses or other Myoviridae. The majority of the predicted genes have no detectable homologues in the databases. These findings indicate that Ma-LMM01 is a member of a new lineage of the Myoviridae family. The genome lacks homologues for the photosynthetic genes that are prevalent in marine cyanophages. However, it has a homologue of nblA, which is essential for the degradation of the major cyanobacteria light-harvesting complex, the phycobilisomes. The genome codes for a site-specific recombinase and two prophage antirepressors, suggesting that it has the capacity to integrate into the host genome. Ma-LMM01 possesses six genes, including three coding for transposases, that are highly similar to homologues found in cyanobacteria, suggesting that recent gene transfers have occurred between Ma-LMM01 and its host. We propose that the Ma-LMM01 NblA homologue possibly reduces the absorption of excess light energy and confers benefits to the phage living in surface waters. This phage genome study suggests that light is central in the phage-cyanobacterium relationships where the viruses use diverse genetic strategies to control their host's photosynthesis.

Intrinsically Disordered Regions of Human Plasma Membrane Proteins Preferentially Occur in the Cytoplasmic Segment

A systematic survey of intrinsically disordered (ID) regions was carried out in 2109 human plasma membrane proteins with full assignment of the transmembrane topology with respect to the lipid bilayer. ID regions with 30 consecutive residues or more were detected in 41.0% of the human proteins, a much higher percentage than the corresponding figure (4.7%) for inner membrane proteins of Escherichia coli. The domain organization of each of the membrane protein in terms of transmembrane helices, structural domains, ID, and unassigned regions as well as the distinction of inside or outside of the cell was determined. Long ID regions constitute 13.3 and 3.5% of the human plasma membrane proteins on the inside and outside of the cell, respectively, showing that they preferentially occur on the cytoplasmic side. We interpret this phenomenon as a reflection of the general scarcity of ID regions on the extracellular side and their relative abundance on the cytoplasmic side in multicellular eukaryotic organisms.

A Novel Sulfatide, GlcCer-I6 Sulfate, from the Ascidian Ciona intestinalis

A novel sulfated glycosphingolipid, SGL-1, was isolated from the ascidian Ciona intestinalis, prepared from chloroform/methanol extracts and fractionated successively on DEAE Sephadex-A25, Florisil and Iatrobeads column chromatographies. Chemical structural analysis was performed using methylation analysis, gas-liquid chromatography, combined gas-liquid chromatography-mass spectrometry, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry and (1)H-NMR spectroscopy. This chemical structure is presented as GlcCer I(6)-Sulfate. The ceramide moiety was specified by t16:0, t17:0, br,t17:0, t18:0 and br,t18:0 as sphingoids, and 2-hydroxy, saturated fatty acids as represented by docosanoic and tetracosanoic acids.

Gene Cluster Analysis Method Identifies Horizontally Transferred Genes with High Reliability and Indicates that They Provide the Main Mechanism of Operon Gain in 8 Species of γ-Proteobacteria

The formation mechanism of operons remains unresolved: operons may form by rearrangements within a genome or by acquisition of genes from other species, that is, horizontal gene transfer (HGT). One hindrance to its elucidation is the unavailability of a method to accurately identify HGT, although it is generally considered to occur. It is critically important first to select horizontally transferred (HT) genes reliably and then to determine the extent to which HGT is involved in operon formation. For this purpose, we considered indels in terms of gene clusters instead of individual genes and chose candidates of HT genes in 8 species of Escherichia, Shigella, and Salmonella based on the minimization of indels. To select a benchmark set of positively HT genes against which we can evaluate the candidate set, we devised another procedure using intergenetic alignments. Comparison with the benchmark set demonstrated the absence of a significant number of false positives in the candidate set, showing the high reliability of the method. Analyses of Escherichia coli K-12 operons revealed that although approximately 20 operons were probably gained from the last common ancestor of the 8 gamma-proteobacteria, deletion of intervening genes accounts for the formation of no operons, whereas horizontal transfer expanded 2 operons and introduced 4 entire operons. Based on these observations and reasoning, we suggest that the main mechanism of operon gain is HGT rather than intragenomic rearrangements. We propose that genes with related essential functions tend to reside in conserved operons, whereas genes in nonconserved operons mostly confer slight advantage to the organisms and frequently undergo horizontal transfer and decay. HT genes constitute at least 5.5% of the genes in the 8 species and approximately 45% of which originate from other gamma-proteobacteria. Genes involved in viral functions and mobile and extrachromosomal element functions are HT more often than expected. This finding indicates frequent mediation of HGT by bacteriophages. On the other hand, not only informational genes (those involved in transcription, translation, and related processes) but also operational genes (those involved in housekeeping) are HT less frequently than expected.

Detection and characterization of fungal-specific proteins in Saccharomyces cerevisiae

In this study, I searched for fungal-specific proteins in the genome of the budding yeast Saccharomyces cerevisiae, inferred from a comparison of amino acid sequences. I used the GTOP (Genomes to Protein structures and functions) database of the DDBJ (DNA Data Bank of Japan), which consists of 21 genomes from Archaea, 203 genomes from Bacteria, and 50 genomes from Eucarya (including 18 fungal genomes). Among 5,874 proteins of S. cerevisiae, 1,551 have homologs only in Eucarya, and 504 of the 1,551 have homologs only in fungi. To find fungal-specific proteins, homologs of the homologs have been searched repeatedly. As a result, 132 of the 504 are characterized as fungal-specific proteins. The genes encoding the 132 fungal-specific proteins are not included in the list of essential genes for viability in the S. cerevisiae genome deletion project. Among the 132 proteins, 99 are S. cerevisiae-specific, and no protein that is distributed among 10 or more of the 18 fungal species exists. In addition, most of the fungal-specific proteins are very small and functionally unknown. My results show that the fungal-specific proteins have short evolutionary histories, suggesting that S. cerevisiae produces novel proteins and that ancestral fungi also produced small proteins most of which have disappeared or have been combined with other proteins during fungal evolution.

Coverage of whole proteome by structural genomics observed through protein homology modeling database

We have been developing FAMSBASE, a protein homology-modeling database of whole ORFs predicted from genome sequences. The latest update of FAMSBASE ( http://daisy.nagahama-i-bio.ac.jp/Famsbase/ ), which is based on the protein three-dimensional (3D) structures released by November 2003, contains modeled 3D structures for 368,724 open reading frames (ORFs) derived from genomes of 276 species, namely 17 archaebacterial, 130 eubacterial, 18 eukaryotic and 111 phage genomes. Those 276 genomes are predicted to have 734,193 ORFs in total and the current FAMSBASE contains protein 3D structure of approximately 50% of the ORF products. However, cases that a modeled 3D structure covers the whole part of an ORF product are rare. When portion of an ORF with 3D structure is compared in three kingdoms of life, in archaebacteria and eubacteria, approximately 60% of the ORFs have modeled 3D structures covering almost the entire amino acid sequences, however, the percentage falls to about 30% in eukaryotes. When annual differences in the number of ORFs with modeled 3D structure are calculated, the fraction of modeled 3D structures of soluble protein for archaebacteria is increased by 5%, and that for eubacteria by 7% in the last 3 years. Assuming that this rate would be maintained and that determination of 3D structures for predicted disordered regions is unattainable, whole soluble protein model structures of prokaryotes without the putative disordered regions will be in hand within 15 years. For eukaryotic proteins, they will be in hand within 25 years. The 3D structures we will have at those times are not the 3D structure of the entire proteins encoded in single ORFs, but the 3D structures of separate structural domains. Measuring or predicting spatial arrangements of structural domains in an ORF will then be a coming issue of structural genomics.

Human transcription factors contain a high fraction of intrinsically disordered regions essential for transcriptional regulation

Human transcriptional regulation factors, such as activators, repressors, and enhancer-binding factors are quite different from their prokaryotic counterparts in two respects: the average sequence in human is more than twice as long as that in prokaryotes, while the fraction of sequence aligned to domains of known structure is 31% in human transcription factors (TFs), less than half of that in bacterial TFs (72%). Intrinsically disordered (ID) regions were identified by a disorder-prediction program, and were found to be in good agreement with available experimental data. Analysis of 401 human TFs with experimental evidence from the Swiss-Prot database showed that as high as 49% of the entire sequence of human TFs is occupied by ID regions. More than half of the human TFs consist of a small DNA binding domain (DBD) and long ID regions frequently sandwiching unassigned regions. The remaining TFs have structural domains in addition to DBDs and ID regions. Experimental studies, particularly those with NMR, revealed that the transactivation domains in unbound TFs are usually unstructured, but become structured upon binding to their partners. The sequences of human and mouse TF orthologues are 90.5% identical despite a high incidence of ID regions, probably reflecting important functional roles played by ID regions. In general ID regions occupy a high fraction in TFs of eukaryotes, but not in prokaryotes. Implications of this dichotomy are discussed in connection with their functional roles in transcriptional regulation and evolution.

Intrinsically Disordered Loops Inserted into the Structural Domains of Human Proteins

Much attention has been paid recently to proteins with partially or fully disordered structures, which are found to exist mostly in eukaryotes and are involved mainly in pivotal cellular processes such as transcriptional regulation, translation and cellular signal transduction. Long disordered sequences are sometimes inserted within the single structural domains of proteins, forming loops from the molecular surface. Such intrinsically disordered loops (IDLs) either are invisible in X-ray crystallography, or hamper protein crystallization itself due to great flexibility. Perhaps because of this, such long disordered sequences have not been characterized adequately. Here, we propose an informational method that stringently identifies IDLs in the structural domains of proteins using the amino acid sequence alone. A genome-wide survey of human proteins conducted with the method identified 50 IDL-containing proteins, several of which have experimentally determined 3D structures. Similar searches in other entirely sequenced organisms revealed that IDLs are prevalent in eukaryotes, while they are much less so in prokaryotes. As there is a statistically significant coincidence between the boundaries of IDLs and those of exons, we suggest that IDLs were produced mainly by exon addition in eukaryotes. IDLs are almost always located at the surface of proteins and are enriched with hydrophilic residues, and IDL-containing proteins tend to be intracellular. Some of the well-characterized proteins with IDLs illustrate that IDLs play pivotal roles in the switching of intracellular signaling or regulatory functions, suggesting that IDL insertion is an effective way to create functionally different domain variants.

Investigation of protein functions through data-mining on integrated human transcriptome database, H-Invitational database (H-InvDB)

H-Invitational Database (H-InvDB; ) is a human transcriptome database, containing integrative annotation of 41,118 full-length cDNA clones originated from 21,037 loci. H-InvDB is a product of the H-Invitational project, an international collaboration to systematically and functionally validate human genes by analysis of a unique set of high quality full-length cDNA clones using automatic annotation and human curation under unified criteria. Here, 19,574 proteins encoded by these cDNAs were classified into 11,709 function-known and 7865 function-unknown hypothetical proteins by similarity with protein databases and motif prediction (InterProScan). The proportion of "hypothetical proteins" in H-InvDB was as high as 40.4%. In this study, we thus conducted data-mining in H-InvDB with the aim of assigning advanced functional annotations to those hypothetical proteins. First, by data-mining in the H-InvDB version of GTOP, we identified 337 SCOP domains within 7865 H-Inv hypothetical proteins. Second, by data-mining of predicted subcellular localization by SOSUI and TMHMM in H-InvDB, we found 1032 transmembrane proteins within H-Inv hypothetical proteins. These results clearly demonstrate that structural prediction is effective for functional annotation of proteins with unknown functions. All the data in H-InvDB are shown in two main views, the cDNA view and the Locus view, and five auxiliary databases with web-based viewers; DiseaseInfo Viewer, H-ANGEL, Clustering Viewer, G-integra and TOPO Viewer; the data also are provided as flat files and XML files. The data consists of descriptions of their gene structures, novel alternative splicing isoforms, functional RNAs, functional domains, subcellular localizations, metabolic pathways, predictions of protein 3D structure, mapping of SNPs and microsatellite repeat motifs in relation with orphan diseases, gene expression profiling, and comparisons with mouse full-length cDNAs in the context of molecular evolution. This unique integrative platform for conducting in silico data-mining represents a substantial contribution to resources required for the exploration of human biology and pathology.

The CGL2612 protein from Corynebacterium glutamicum is a drug resistance-related transcriptional repressor: structural and functional analysis of a newly identified transcription factor from genomic DNA analysis

The emergence of antibiotic-resistant bacteria often causes serious clinical problems. The TetR family is one of the major transcription factor families that regulate expression of genes involved in bacterial antimicrobial resistance systems. CGL2612 protein is a transcription factor newly identified by genomic DNA analysis on Corynebacterium glutamicum, which belongs to the mycolic acid-containing Actinomycetales, including the well known pathogens Corynebacterium diphtheriae and Mycobacterium tuberculosis. Crystal structure analysis showed that the CGL2612 protein exhibits significant structural similarity to the multidrug resistance (MDR)-related transcription factor QacR from Staphylococcus aureus, despite poor amino acid sequence similarity between these proteins. Binding DNA sequence analysis of CGL2612 protein using the systematic evolution of ligands by the exponential enrichment (systematic evolution of ligands by exponential enrichment, or SELEX) method revealed that this protein is a new member of the TetR family, which regulates expression of the immediately upstream gene, cgl2611, probably encoding a major facilitator superfamily permease. Subsequent functional analyses confirmed a function of the CGL2612 as a transcriptional repressor responsible for the antimicrobial resistance system in C. glutamicum. The strategy used in the present study is one of the most convenient and powerful methods to analyze functionally unknown transcription factors, and the results obtained here will contribute to our understanding of the drug resistance mechanism not only in C. glutamicum but also in the related bacteria, C. diphtheriae and M. tuberculosis.

Tertiary structure predictions on a comprehensive benchmark of medium to large size proteins

We evaluate tertiary structure predictions on medium to large size proteins by TASSER, a new algorithm that assembles protein structures through rearranging the rigid fragments from threading templates guided by a reduced Calpha and side-chain based potential consistent with threading based tertiary restraints. Predictions were generated for 745 proteins 201-300 residues in length that cover the Protein Data Bank (PDB) at the level of 35% sequence identity. With homologous proteins excluded, in 365 cases, the templates identified by our threading program, PROSPECTOR_3, have a root-mean-square deviation (RMSD) to native < 6.5 angstroms, with >70% alignment coverage. After TASSER assembly, in 408 cases the best of the top five full-length models has a RMSD < 6.5 angstroms. Among the 745 targets are 18 membrane proteins, with one-third having a predicted RMSD < 5.5 A. For all representative proteins less than or equal to 300 residues that have corresponding multiple NMR structures in the Protein Data Bank, approximately 20% of the models generated by TASSER are closer to the NMR structure centroid than the farthest individual NMR model. These results suggest that reasonable structure predictions for nonhomologous large size proteins can be automatically generated on a proteomic scale, and the application of this approach to structural as well as functional genomics represent promising applications of TASSER.

Molecular models of protein targets from Mycobacterium tuberculosis

Structural characterization of enzymes that belong to microbial metabolic pathways is very important for structure-based drug design since some of these proteins may be present in the bacterial genome, but absent in humans. Thus, metabolic pathways became potential targets for drug design. The motivation of this work is the fact that Mycobacterium tuberculosis is the cause of the deaths of millions of people in the world, so that the structural characterization of protein targets to propose new drugs has become essential. DBMODELING is a relational database, created to highlight the importance of methods of molecular modeling applied to the Mycobacterium tuberculosis genome with the aim of proposing protein-ligand docking analysis. There are currently more than 300 models for proteins from Mycobacterium tuberculosis genome in the database. The database contains a detailed description of the reaction catalyzed by each enzyme and their atomic coordinates. Information about structures, a tool for animated gif image, a table with a specification of the metabolic pathway, modeled protein, inputs used in modeling, and analysis methods used in this project are available in the database for download. The search tool can be used for researchers to find specific pathways or enzymes.

Proteome-wide functional classification and identification of prokaryotic transmembrane proteins by transmembrane topology similarity comparison

We propose a new method for classifying and identifying transmembrane (TM) protein functions in proteome-scale by applying a single-linkage clustering method based on TM topology similarity, which is calculated simply from comparing the lengths of loop regions. In this study, we focused on 87 prokaryotic TM proteomes consisting of 31 proteobacteria, 22 gram-positive bacteria, 19 other bacteria, and 15 archaea. Prior to performing the clustering, we first categorized individual TM protein sequences as "known," "putative" (similar to "known" sequences), or "unknown" by using the homology search and the sequence similarity comparison against SWISS-PROT to assess the current status of the functional annotation of the TM proteomes based on sequence similarity only. More than three-quarters, that is, 75.7% of the TM protein sequences are functionally "unknown," with only 3.8% and 20.5% of them being classified as "known" and "putative," respectively. Using our clustering approach based on TM topology similarity, we succeeded in increasing the rate of TM protein sequences functionally classified and identified from 24.3% to 60.9%. Obtained clusters correspond well to functional superfamilies or families, and the functional classification and identification are successfully achieved by this approach. For example, in an obtained cluster of TM proteins with six TM segments, 109 sequences out of 119 sequences annotated as "ATP-binding cassette transporter" are properly included and 122 "unknown" sequences are also contained.

Alternative Splice Variants Encoding Unstable Protein Domains Exist in the Human Brain

Alternative splicing has been recognized as a major mechanism by which protein diversity is increased without significantly increasing genome size in animals and has crucial medical implications, as many alternative splice variants are known to cause diseases. Despite the importance of knowing what structural changes alternative splicing introduces to the encoded proteins for the consideration of its significance, the problem has not been adequately explored. Therefore, we systematically examined the structures of the proteins encoded by the alternative splice variants in the HUGE protein database derived from long (>4 kb) human brain cDNAs. Limiting our analyses to reliable alternative splice junctions, we found alternative splice junctions to have a slight tendency to avoid the interior of SCOP domains and a strong statistically significant tendency to coincide with SCOP domain boundaries. These findings reflect the occurrence of some alternative splicing events that utilize protein structural units as a cassette. However, 50 cases were identified in which SCOP domains are disrupted in the middle by alternative splicing. In six of the cases, insertions are introduced at the molecular surface, presumably affecting protein functions, while in 11 of the cases alternatively spliced variants were found to encode pairs of stable and unstable proteins. The mRNAs encoding such unstable proteins are much less abundant than those encoding stable proteins and tend not to have corresponding mRNAs in non-primate species. We propose that most unstable proteins encoded by alternative splice variants lack normal functions and are an evolutionary dead-end.

Alteration of sugar donor specificities of plant glycosyltransferases by a single point mutation

In comparison with the amino acid sequences of seven species of glucosyltransferases and six species of galactosyltransferases, glutamine and histidine are highly conserved as the last amino acid residue of a glycosyltransferase-specific conserved region (UDPGT) in glucosyltransferases and galactosyltransferases, respectively. Consequently, the sugar donor specificities of glycosyltransferases are successfully altered by a single amino acid point mutation. UDP-galactose:anthocyanin galactosyltransferase (ACGaT), isolated from Aralia cordata, acquired glucosyltransferase activity in addition to the inherent galactosyltransferase activity by replacing histidine with glutamine. In contrast, UDP-glucose:flavonoid glucosyltransferase (UBGT), isolated from Scutellaria baicalensis, did not acquire galactosyltransferase activity by replacing glutamine with histidine, and exhibited a remarkable decrease in glucosyltransferase activity.

Microbial genomes have over 72% structure assignment by the threading algorithm PROSPECTOR_Q

The genome scale threading of five complete microbial genomes is revisited using our state-of-the-art threading algorithm, PROSPECTOR_Q. Considering that structure assignment to an ORF could be useful for predicting biochemical function as well as for analyzing pathways, it is important to assess the current status of genome scale threading. The fraction of ORFs to which we could assign protein structures with a reasonably good confidence level to each genome sequences is over 72%, which is significantly higher than earlier studies. Using the assigned structures, we have predicted the function of several ORFs through "single-function" template structures, obtained from an analysis of the relationship between protein fold and function. The fold distribution of the genomes and the effect of the number of homologous sequences on structure assignment are also discussed.

Automated structure prediction of weakly homologous proteins on a genomic scale

We have developed TASSER, a hierarchical approach to protein structure prediction that consists of template identification by threading, followed by tertiary structure assembly via the rearrangement of continuous template fragments guided by an optimized C(alpha) and side-chain-based potential driven by threading-based, predicted tertiary restraints. TASSER was applied to a comprehensive benchmark set of 1,489 medium-sized proteins in the Protein Data Bank. With homologues excluded, in 927 cases, the templates identified by our threading algorithm PROSPECTOR_3 have a rms deviation from native <6.5 A with approximately 80% alignment coverage. After template reassembly, this number increases to 1,172. This shows significant and systematic improvement of the final models with respect to the initial template alignments. Furthermore, significant improvements in loop modeling are demonstrated. We then apply TASSER to the 1,360 medium-sized ORFs in the Escherichia coli genome; approximately 920 can be predicted with high accuracy based on confidence criteria established in the Protein Data Bank benchmark. These results from our unprecedented comprehensive folding benchmark on all protein categories provide a reliable basis for the application of TASSER to structural genomics, especially to proteins of low sequence identity to solved protein structures.

Automated structure prediction of weakly homologous proteins on a genomic scale

We have developed TASSER, a hierarchical approach to protein structure prediction that consists of template identification by threading, followed by tertiary structure assembly via the rearrangement of continuous template fragments guided by an optimized C(alpha) and side-chain-based potential driven by threading-based, predicted tertiary restraints. TASSER was applied to a comprehensive benchmark set of 1,489 medium-sized proteins in the Protein Data Bank. With homologues excluded, in 927 cases, the templates identified by our threading algorithm PROSPECTOR_3 have a rms deviation from native <6.5 A with approximately 80% alignment coverage. After template reassembly, this number increases to 1,172. This shows significant and systematic improvement of the final models with respect to the initial template alignments. Furthermore, significant improvements in loop modeling are demonstrated. We then apply TASSER to the 1,360 medium-sized ORFs in the Escherichia coli genome; approximately 920 can be predicted with high accuracy based on confidence criteria established in the Protein Data Bank benchmark. These results from our unprecedented comprehensive folding benchmark on all protein categories provide a reliable basis for the application of TASSER to structural genomics, especially to proteins of low sequence identity to solved protein structures.

Gopinath G, Graudens E, Hahn Y, Han M, Han ZG, Hanada K, Hanaoka H, Harada E, Hashimoto K, Hinz U, Hirai M, Hishiki T, Hopkinson I, Imbeaud S, Inoko H, Kanapin A, Kaneko Y, Kasukawa T, Kelso J, Kersey P, Kikuno R, Kimura K, Korn B, Kuryshev V, Makalowska I, Makino T, Mano S, Mariage-Samson R, Mashima J, Matsuda H, Mewes HW, Minoshima S, Nagai K, Nagasaki H, Nagata N, Nigam R, Ogasawara O, Ohara O, Ohtsubo M, Okada N, Okido T, Oota S, Ota M, Ota T, Otsuki T, Piatier-Tonneau D, Poustka A, Ren SX, Saitou N, Sakai K, Sakamoto S, Sakate R, Schupp I, Servant F, Sherry S, Shiba R, Shimizu N, Shimoyama M, Simpson AJ, Soares B, Steward C, Suwa M, Suzuki M, Takahashi A, Tamiya G, Tanaka H, Taylor T, Terwilliger JD, Unneberg P, Veeramachaneni V, Watanabe S, Wilming L, Yasuda N, Yoo HS, Stodolsky M, Makalowski W, Go M, Nakai K, Takagi T, Kanehisa M, Sakaki Y, Quackenbush J, Okazaki Y, Hayashizaki Y, Hide W, Chakraborty R, Nishikawa K, Sugawara H, Tateno Y, Chen Z, Oishi M, Tonellato P, Apweiler R, Okubo K, Wagner L, Wiemann S, Strausberg RL, Isogai T, Auffray C, Nomura N, Gojobori T, Sugano S. Integrative annotation of 21,037 human genes validated by full-length cDNA clones

The human genome sequence defines our inherent biological potential; the realization of the biology encoded therein requires knowledge of the function of each gene. Currently, our knowledge in this area is still limited. Several lines of investigation have been used to elucidate the structure and function of the genes in the human genome. Even so, gene prediction remains a difficult task, as the varieties of transcripts of a gene may vary to a great extent. We thus performed an exhaustive integrative characterization of 41,118 full-length cDNAs that capture the gene transcripts as complete functional cassettes, providing an unequivocal report of structural and functional diversity at the gene level. Our international collaboration has validated 21,037 human gene candidates by analysis of high-quality full-length cDNA clones through curation using unified criteria. This led to the identification of 5,155 new gene candidates. It also manifested the most reliable way to control the quality of the cDNA clones. We have developed a human gene database, called the H-Invitational Database (H-InvDB; http://www.h-invitational.jp/). It provides the following: integrative annotation of human genes, description of gene structures, details of novel alternative splicing isoforms, non-protein-coding RNAs, functional domains, subcellular localizations, metabolic pathways, predictions of protein three-dimensional structure, mapping of known single nucleotide polymorphisms (SNPs), identification of polymorphic microsatellite repeats within human genes, and comparative results with mouse full-length cDNAs. The H-InvDB analysis has shown that up to 4% of the human genome sequence (National Center for Biotechnology Information build 34 assembly) may contain misassembled or missing regions. We found that 6.5% of the human gene candidates (1,377 loci) did not have a good protein-coding open reading frame, of which 296 loci are strong candidates for non-protein-coding RNA genes. In addition, among 72,027 uniquely mapped SNPs and insertions/deletions localized within human genes, 13,215 nonsynonymous SNPs, 315 nonsense SNPs, and 452 indels occurred in coding regions. Together with 25 polymorphic microsatellite repeats present in coding regions, they may alter protein structure, causing phenotypic effects or resulting in disease. The H-InvDB platform represents a substantial contribution to resources needed for the exploration of human biology and pathology.

3D-GENOMICS: a database to compare structural and functional annotations of proteins between sequenced genomes

The 3D-GENOMICS database (http://www.sbg.bio. ic.ac.uk/3dgenomics/) provides structural annotations for proteins from sequenced genomes. In August 2003 the database included data for 93 proteomes. The annotations stored in the database include homologous sequences from various sequence databases, domains from SCOP and Pfam, patterns from Prosite and other predicted sequence features such as transmembrane regions and coiled coils. In addition to annotations at the sequence level, several precomputed cross- proteome comparative analyses are available based on SCOP domain superfamily composition. Annotations are available to the user via a web interface to the database. Multiple points of entry are available so that a user is able to: (i) directly access annotations for a single protein sequence via keywords or accession codes, (ii) examine a sequence of interest chosen from a summary of annotations for a particular proteome, or (iii) access precomputed frequency-based cross-proteome comparative analyses.

Molecular modeling of the membrane targeting of phospholipase C pleckstrin homology domains

Phospholipases C (PLCs) reversibly associate with membranes to hydrolyze phosphatidylinositol-4, 5-bisphosphate (PI[4,5]P(2)) and comprise four main classes: beta, gamma, delta, and epsilon. Most eukaryotic PLCs contain a single, N-terminal pleckstrin homology (PH) domain, which is thought to play an important role in membrane targeting. The structure of a single PLC PH domain, that from PLCdelta1, has been determined; this PH domain binds PI(4,5)P(2) with high affinity and stereospecificity and has served as a paradigm for PH domain functionality. However, experimental studies demonstrate that PH domains from different PLC classes exhibit diverse modes of membrane interaction, reflecting the dissimilarity in their amino acid sequences. To elucidate the structural basis for their differential membrane-binding specificities, we modeled the three-dimensional structures of all mammalian PLC PH domains by using bioinformatic tools and calculated their biophysical properties by using continuum electrostatic approaches. Our computational analysis accounts for a large body of experimental data, provides predictions for those PH domains with unknown functions, and indicates functional roles for regions other than the canonical lipid-binding site identified in the PLCdelta1-PH structure. In particular, our calculations predict that (1). members from each of the four PLC classes exhibit strikingly different electrostatic profiles than those ordinarily observed for PH domains in general, (2). nonspecific electrostatic interactions contribute to the membrane localization of PLCdelta-, PLCgamma-, and PLCbeta-PH domains, and (3). phosphorylation regulates the interaction of PLCbeta-PH with its effectors through electrostatic repulsion. Our molecular models for PH domains from all of the PLC classes clearly demonstrate how a common structural fold can serve as a scaffold for a wide range of surface features and biophysical properties that support distinctive functional roles.

Unique amino acid composition of proteins in halophilic bacteria

The amino acid compositions of proteins from halophilic archaea were compared with those from non-halophilic mesophiles and thermophiles, in terms of the protein surface and interior, on a genome-wide scale. As we previously reported for proteins from thermophiles, a biased amino acid composition also exists in halophiles, in which an abundance of acidic residues was found on the protein surface as compared to the interior. This general feature did not seem to depend on the individual protein structures, but was applicable to all proteins encoded within the entire genome. Unique protein surface compositions are common in both halophiles and thermophiles. Statistical tests have shown that significant surface compositional differences exist among halophiles, non-halophiles, and thermophiles, while the interior composition within each of the three types of organisms does not significantly differ. Although thermophilic proteins have an almost equal abundance of both acidic and basic residues, a large excess of acidic residues in halophilic proteins seems to be compensated by fewer basic residues. Aspartic acid, lysine, asparagine, alanine, and threonine significantly contributed to the compositional differences of halophiles from meso- and thermophiles. Among them, however, only aspartic acid deviated largely from the expected amount estimated from the dinucleotide composition of the genomic DNA sequence of the halophile, which has an extremely high G+C content (68%). Thus, the other residues with large deviations (Lys, Ala, etc.) from their non-halophilic frequencies could have arisen merely as "dragging effects" caused by the compositional shift of the DNA, which would have changed to increase principally the fraction of aspartic acid alone.

Enlarged FAMSBASE: protein 3D structure models of genome sequences for 41 species

Enlarged FAMSBASE is a relational database of comparative protein structure models for the whole genome of 41 species, presented in the GTOP database. The models are calculated by Full Automatic Modeling System (FAMS). Enlarged FAMSBASE provides a wide range of query keys, such as name of ORF (open reading frame), ORF keywords, Protein Data Bank (PDB) ID, PDB heterogen atoms and sequence similarity. Heterogen atoms in PDB include cofactors, ligands and other factors that interact with proteins, and are a good starting point for analyzing interactions between proteins and other molecules. The data may also work as a template for drug design. The present number of ORFs with protein 3D models in FAMSBASE is 183 805, and the database includes an average of three models for each ORF. FAMSBASE is available at http://famsbase.bio.nagoya-u.ac.jp/famsbase/.

A systematic investigation identifies a significant number of probable pseudogenes in the Escherichia coli genome

Pseudogenes are open reading frames (ORFs) encoding dysfunctional proteins with high homology to known protein-coding genes. Although pseudogenes were reported to exist in the genomes of many eukaryotes and bacteria, no systematic search for pseudogenes in the Escherichia coli genome has been carried out. Genome comparisons of E. coli strains K-12 and O157 revealed that many protein-coding sequences have prematurely terminated orthologs encoding unstable proteins. To systematically screen for pseudogenes, we selected ORFs generated by premature termination of the orthologous protein-coding genes and subsequently excluded those possibly arising from sequence errors. Lastly we eliminated those with close homologs in this and other species, as these shortened ORFs may actually have functions. The process produced 95 and 101 pseudogene candidates in K-12 and O157, respectively. The assigned three-dimensional structures suggest that most of the encoded proteins cannot fold properly and thus are dysfunctional, indicating that they are probably pseudogenes. Therefore, the existence of a significant number of probable pseudogenes in E. coli is predicted, awaiting experimental verification. Most of them were found to be genes with paralogs or horizontally transferred genes or both. We suggest that pseudogenes constitute a small fraction of the genomes of free-living bacteria in general, reflecting the faster elimination than production of pseudogenes.