BioGraph: Data Model for Linking and Querying Diverse Biological Metadata.

Studying the association of gene function, diseases, and regulatory gene network reconstruction demands data compatibility. Data from different databases follow distinct schemas and are accessible in heterogenic ways. Although the experiments differ, data may still be related to the same biological entities. Some entities may not be strictly biological, such as geolocations of habitats or paper references, but they provide a broader context for other entities. The same entities from different datasets can share similar properties, which may or may not be found within other datasets. Joint, simultaneous data fetching from multiple data sources is complicated for the end-user or, in many cases, unsupported and inefficient due to differences in data structures and ways of accessing the data. We propose BioGraph-a new model that enables connecting and retrieving information from the linked biological data that originated from diverse datasets. We have tested the model on metadata collected from five diverse public datasets and successfully constructed a knowledge graph containing more than 17 million model objects, of which 2.5 million are individual biological entity objects. The model enables the selection of complex patterns and retrieval of matched results that can be discovered only by joining the data from multiple sources.

Prediction of structural alphabet protein blocks using data mining.

3D protein structures determine proteins' biological functions. The 3D structure of the protein backbone can be approximated using the prototypes of local protein conformations. Sets of these prototypes are called structural alphabets (SAs). Amongst several approaches to the prediction of 3D structures from amino acid sequences, one approach is based on the prediction of SA prototypes for a given amino acid sequence. Protein Blocks (PBs) is the most known SA, and it is composed of 16 prototypes of five consecutive amino acids which were identified as optimal prototypes considering the ability to correctly approximate the local structure and the prediction accuracy of prototypes from an amino acid sequence. We developed models for PBs prediction from sequence information using different data mining approaches and machine learning algorithms. Besides the amino acid sequences, the results of the following tools were used to train the models: the Spider3 predictor of protein structure properties, several predictors of the protein's intrinsically disordered regions, and a tool for finding repeats in amino acid sequences. The highest accuracy of the constructed models is 80%, which is a significant improvement compared to the previous best available prediction, whose accuracy was 61%. Analyzing the models constructed by applying different algorithms, it was noticed that the significance of input attributes differs among the models constructed by algorithms. Using the information about amino acids belonging to intrinsically disordered regions and repeats improves the precision of prediction for some PBs using the CART classification algorithm, while this is not the case with the C5.0 classification algorithm. Improved prediction approaches can have interesting applications in protein structural model approaches or computational protein design.

Entropy-driven translocation of disordered proteins through the Gram-positive bacterial cell wall.

In Gram-positive bacteria, a thick cross-linked cell wall separates the membrane from the extracellular space. Some surface-exposed proteins, such as the Listeria monocytogenes actin nucleation-promoting factor ActA, remain associated with the bacterial membrane but somehow thread through tens of nanometres of cell wall to expose their amino terminus to the exterior. Here, we report that entropy enables the translocation of disordered transmembrane proteins through the Gram-positive cell wall. We build a physical model, which predicts that the entropic constraint imposed by a thin periplasm is sufficient to drive the translocation of an intrinsically disordered protein such as ActA across a porous barrier similar to a peptidoglycan cell wall. We experimentally validate our model and show that ActA translocation depends on the cell-envelope dimensions and disordered-protein length, and that translocation is reversible. We also show that disordered regions of eukaryotic proteins can translocate Gram-positive cell walls via entropy. We propose that entropic forces are sufficient to drive the translocation of specific proteins to the outer surface.

Structural disorder of plasmid-encoded proteins in Bacteria and Archaea.

BACKGROUND: In the last decade and a half it has been firmly established that a large number of proteins do not adopt a well-defined (ordered) structure under physiological conditions. Such intrinsically disordered proteins (IDPs) and intrinsically disordered (protein) regions (IDRs) are involved in essential cell processes through two basic mechanisms: the entropic chain mechanism which is responsible for rapid fluctuations among many alternative conformations, and molecular recognition via short recognition elements that bind to other molecules. IDPs possess a high adaptive potential and there is special interest in investigating their involvement in organism evolution. RESULTS: We analyzed 2554 Bacterial and 139 Archaeal proteomes, with a total of 8,455,194 proteins for disorder content and its implications for adaptation of organisms, using three disorder predictors and three measures. Along with other findings, we revealed that for all three predictors and all three measures (1) Bacteria exhibit significantly more disorder than Archaea; (2) plasmid-encoded proteins contain considerably more IDRs than proteins encoded on chromosomes (or whole genomes) in both prokaryote superkingdoms; (3) plasmid proteins are significantly more disordered than chromosomal proteins only in the group of proteins with no COG category assigned; (4) antitoxin proteins in comparison to other proteins, are the most disordered (almost double) in both Bacterial and Archaeal proteomes; (5) plasmidal proteins are more disordered than chromosomal proteins in Bacterial antitoxins and toxin-unclassified proteins, but have almost the same disorder content in toxin proteins. CONCLUSION: Our results suggest that while disorder content depends on genome and proteome characteristics, it is more influenced by functional engagements than by gene location (on chromosome or plasmid).

Finding Statistically Significant Repeats in Nucleic Acids and Proteins.

DNA repeats have great importance for biological research and a large number of tools for determining repeats have been developed. Herein we define a method for extracting a statistically significant subset of a determined set of repeats. Our aim was to identify a subset of repeats in the input sequences that are not expected to occur with a number of their appearances in a random sequence of the same length. It is expected that results obtained in such manner would reduce the quantity of processed material and could thereby represent a more important biological signal. With DNA, RNA, and protein sequences serving as input material, we also examined the possibility of statistical filtering of repeats in sequences over an arbitrary alphabet. A new method for selecting statistically significant repeats from a set of determined repeats has been defined. The proposed method was tested on a large number of randomly generated sequences. The application of the method on biological sequences revealed that for some viruses, shorter repeats are more statistically significant than longer ones because of their frequent appearance, whereas for bacteria, the majority of identified repeats are statistically significant.

Software tools for simultaneous data visualization and T cell epitopes and disorder prediction in proteins.

We have developed EpDis and MassPred, extendable open source software tools that support bioinformatic research and enable parallel use of different methods for the prediction of T cell epitopes, disorder and disordered binding regions and hydropathy calculation. These tools offer a semi-automated installation of chosen sets of external predictors and an interface allowing for easy application of the prediction methods, which can be applied either to individual proteins or to datasets of a large number of proteins. In addition to access to prediction methods, the tools also provide visualization of the obtained results, calculation of consensus from results of different methods, as well as import of experimental data and their comparison with results obtained with different predictors. The tools also offer a graphical user interface and the possibility to store data and the results obtained using all of the integrated methods in the relational database or flat file for further analysis. The MassPred part enables a massive parallel application of all integrated predictors to the set of proteins. Both tools can be downloaded from http://bioinfo.matf.bg.ac.rs/home/downloads.wafl?cat=Software. Appendix A includes the technical description of the created tools and a list of supported predictors.

Epitope distribution in ordered and disordered protein regions. Part B - Ordered regions and disordered binding sites are targets of T- and B-cell immunity.

Intrinsically disordered proteins exist in highly flexible conformational states linked to different protein functions. In this work, we have presented evidence that HLA class-I- and class-II-binding T-cell epitopes, experimentally verified in several tumor-associated antigens and nuclear systemic autoantigens, are predominantly located in ordered protein regions or at disorder/order borderlines, defined by the majority of analyzed publicly available disorder predictors. We have also observed the overlapping of secondary structural elements and prevalently hydrophobic regions with T-cell epitopes in Epstein Barr Virus (EBV) nuclear antigen 1 (EBNA-1), cancer/testis antigen MAGE-A4, and Sm-B/B', U1 snRNPA (U1A) and U1-70kDa autoantigens. The results are in accordance with the clustering of the predicted HLA class-I and class-II epitopes in protein parts which encompass the consensus of ordered regions, determined by individual disorder predictors. Some HLA class-II epitopes and linear B-cell epitopes were located near the segments predicted to have elevated crystallographic B factor in EBNA-1, Sm-B/B' and U1 snRNP A proteins, suggesting that protein flexibility could influence the structural availability of epitopes. Naturally processed T-cell epitopes and linear B-cell epitopes could also be found within putative disordered binding sites, determined by "dips" in the prevalently disordered parts of prediction profiles of the majority of disorder predictors, and peaks in ANCHOR-prediction profile. Two minor antigenic regions within EBNA-1, mapped to the residues 58-85 and 398-458, encompassing putative disordered binding sites, contain epitopes connected with anti-Ro 60kDa and anti-Sm B/B' autoimmunity in systemic lupus erythematosus. One of these regions overlaps residues 395-450, identified as the binding site of USP7 (HAUSP), which regulates the EBNA-1 replication function. In Sm-B/B', one of the putative disordered binding sites (residues 114-165) encompasses the T-cell epitope 136-153, while another, residues 200-216, flanks two proline-rich B-cell epitopes (residues 190-198 and 216-222), overlapping the preferred CD2BP2-GYF-binding motif (R/K/G)XXPPGX(R/K), characteristic of splicosomal proteins. We have noticed that the same motif (residues 397-403) is mimicked in EBNA-1 and overlaps epitope 398-404, involved in anti-Sm B/B' autoimmunity. The majority of recognized T- and B-cell epitopes in analyzed autoantigens or tumor-associated antigens appertain to the ordered or transient protein structures. The congruence between certain B- and T-cell epitopes and predicted disordered binding sites or protein-binding eukaryotic motifs in the antigens participating in molecular complexes might influence the capture of antigens, their processing and subsequent presentation and immunodominance.

Epitope distribution in ordered and disordered protein regions - part A. T-cell epitope frequency, affinity and hydropathy.

Highly disordered protein regions are prevalently hydrophilic, extremely sensitive to proteolysis in vitro, and are expected to be under-represented as T-cell epitopes. The aim of this research was to find out whether disorder and hydropathy prediction methods could help in understanding epitope processing and presentation. According to the pan-specific T-cell epitope predictors NetMHCpan and NetMHCIIpan and 9 publicly available disorder predictors, frequency of epitopes presented by human leukocyte antigens (HLA) class-I or -II was found to be more than 2.5 times higher in ordered than in disordered protein regions (depending on the disorder predictor). Both HLA class-I and HLA class-II binding epitopes are prevalently hydrophilic in disordered and prevalently hydrophobic in ordered protein regions, whereas epitopes recognized by HLA class-II alleles are more hydrophobic than those recognized by HLA class-I. As regards both classes of HLA molecules, high-affinity binding epitopes display more hydrophobicity than low affinity-binding epitopes (in both ordered and disordered regions). Epitopes belonging to disordered protein regions were not predicted to have poor affinity to HLA class-II molecules, as expected from disorder intrinsic proteolytic instability. The relation of epitope hydrophobicity and order/disorder location was also valid if alleles were grouped according to the HLA class-I and HLA class-II supertypes, except for the class-I supertype A3 in which the main part of recognized epitopes was prevalently hydrophilic. Regarding specific supertypes, the affinity of epitopes belonging to ordered regions varies only slightly (depending on the disorder predictor) compared to the affinity of epitopes in corresponding disordered regions. The distribution of epitopes in ordered and disordered protein regions has revealed that the curves of order-epitope distribution were convex-like while the curves of disorder-epitope distribution were concave-like. The percentage of prevalently hydrophobic epitopes increases with the enhancement of epitope promiscuity level and moving from disordered to ordered regions. These data suggests that reverse vaccinology, oriented towards promiscuous and high-affinity epitopes, is also oriented towards prevalently hydrophobic, ordered regions. The analysis of predicted and experimentally evaluated epitopes of cancer-testis antigen MAGE-A3 has confirmed that the majority of T-cell epitopes, particularly those that are promiscuous or naturally processed, was located in ordered and disorder/order boundary protein regions overlapping hydrophobic regions.

Bioinformatics analysis of disordered proteins in prokaryotes.

BACKGROUND: A significant number of proteins have been shown to be intrinsically disordered, meaning that they lack a fixed 3 D structure or contain regions that do not posses a well defined 3 D structure. It has also been proven that a protein's disorder content is related to its function. We have performed an exhaustive analysis and comparison of the disorder content of proteins from prokaryotic organisms (i.e., superkingdoms Archaea and Bacteria) with respect to functional categories they belong to, i.e., Clusters of Orthologous Groups of proteins (COGs) and groups of COGs-Cellular processes (Cp), Information storage and processing (Isp), Metabolism (Me) and Poorly characterized (Pc). We also analyzed the disorder content of proteins with respect to various genomic, metabolic and ecological characteristics of the organism they belong to. We used correlations and association rule mining in order to identify the most confident associations between specific modalities of the characteristics considered and disorder content. RESULTS: Bacteria are shown to have a somewhat higher level of protein disorder than archaea, except for proteins in the Me functional group. It is demonstrated that the Isp and Cp functional groups in particular (L-repair function and N-cell motility and secretion COGs of proteins in specific) possess the highest disorder content, while Me proteins, in general, posses the lowest. Disorder fractions have been confirmed to have the lowest level for the so-called order-promoting amino acids and the highest level for the so-called disorder promoters. For each pair of organism characteristics, specific modalities are identified with the maximum disorder proteins in the corresponding organisms, e.g., high genome size-high GC content organisms, facultative anaerobic-low GC content organisms, aerobic-high genome size organisms, etc. Maximum disorder in archaea is observed for high GC content-low genome size organisms, high GC content-facultative anaerobic or aquatic or mesophilic organisms, etc. Maximum disorder in bacteria is observed for high GC content-high genome size organisms, high genome size-aerobic organisms, etc. Some of the most reliable association rules mined establish relationships between high GC content and high protein disorder, medium GC content and both medium and low protein disorder, anaerobic organisms and medium protein disorder, Gammaproteobacteria and low protein disorder, etc. A web site Prokaryote Disorder Database has been designed and implemented at the address http://bioinfo.matf.bg.ac.rs/disorder, which contains complete results of the analysis of protein disorder performed for 296 prokaryotic completely sequenced genomes. CONCLUSIONS: Exhaustive disorder analysis has been performed by functional classes of proteins, for a larger dataset of prokaryotic organisms than previously done. Results obtained are well correlated to those previously published, with some extension in the range of disorder level and clear distinction between functional classes of proteins. Wide correlation and association analysis between protein disorder and genomic and ecological characteristics has been performed for the first time. The results obtained give insight into multi-relationships among the characteristics and protein disorder. Such analysis provides for better understanding of the evolutionary process and may be useful for taxon determination. The main drawback of the approach is the fact that the disorder considered has been predicted and not experimentally established.

n-Gram characterization of genomic islands in bacterial genomes.

The paper presents a novel, n-gram-based method for analysis of bacterial genome segments known as genomic islands (GIs). Identification of GIs in bacterial genomes is an important task since many of them represent inserts that may contribute to bacterial evolution and pathogenesis. In order to characterize and distinguish GIs from rest of the genome, binary classification of islands based on n-gram frequency distribution have been performed. It consists of testing the agreement of islands n-gram frequency distributions with the complete genome and backbone sequence. In addition, a statistic based on the maximal order Markov model is used to identify significantly overrepresented and underrepresented n-grams in islands. The results may be used as a basis for Zipf-like analysis suggesting that some of the n-grams are overrepresented in a subset of islands and underrepresented in the backbone, or vice versa, thus complementing the binary classification. The method is applied to strain-specific regions in the Escherichia coli O157:H7 EDL933 genome (O-islands), resulting in two groups of O-islands with different n-gram characteristics. It refines a characterization based on other compositional features such as G+C content and codon usage, and may help in identification of GIs, and also in research and development of adequate drugs targeting virulence genes in them.

Could n-gram analysis contribute to genomic island determination?

There are two approaches to identifying genomic and pathogenesis islands (GI/PAIs) in bacterial genomes: the compositional and the functional, based on DNA or protein level composition and gene function, respectively. We applied n-gram analysis in addition to other compositional features, combined them by union and intersection and defined two measures for evaluating the results-recall and precision. Using the best criteria (by training on the Escherichia coli O157:H7 EDL933 genome), we predicted GIs for 14 Enterobacteriaceae family members and for 21 randomly selected bacterial genomes. These predictions were compared with results obtained from HGT DB (based on the compositional approach) and PAI DB (based on the combined approach). The results obtained show that intersecting n-grams with other compositional features improves relative precision by up to 10% in case of HGT DB and up to 60% in case of PAI DB. In addition, it was demonstrated that the union of all compositional features results in maximum recall (up to 37%). Thus, the application of n-gram analysis alongside existing or newly developed methods may improve the prediction of GI/PAIs.

SARS-CoV genome polymorphism: a bioinformatics study.

A dataset of 103 SARS-CoV isolates (101 human patients and 2 palm civets) was investigated on different aspects of genome polymorphism and isolate classification. The number and the distribution of single nucleotide variations (SNVs) and insertions and deletions, with respect to a "profile", were determined and discussed ("profile" being a sequence containing the most represented letter per position). Distribution of substitution categories per codon positions, as well as synonymous and non-synonymous substitutions in coding regions of annotated isolates, was determined, along with amino acid (a.a.) property changes. Similar analysis was performed for the spike (S) protein in all the isolates (55 of them being predicted for the first time). The ratio Ka/Ks confirmed that the S gene was subjected to the Darwinian selection during virus transmission from animals to humans. Isolates from the dataset were classified according to genome polymorphism and genotypes. Genome polymorphism yields to two groups, one with a small number of SNVs and another with a large number of SNVs, with up to four subgroups with respect to insertions and deletions. We identified three basic nine-locus genotypes: TTTT/TTCGG, CGCC/TTCAT, and TGCC/TTCGT, with four subgenotypes. Both classifications proposed are in accordance with the new insights into possible epidemiological spread, both in space and time.

Bioinformatics analysis of SARS coronavirus genome polymorphism.

BACKGROUND: We have compared 38 isolates of the SARS-CoV complete genome. The main goal was twofold: first, to analyze and compare nucleotide sequences and to identify positions of single nucleotide polymorphism (SNP), insertions and deletions, and second, to group them according to sequence similarity, eventually pointing to phylogeny of SARS-CoV isolates. The comparison is based on genome polymorphism such as insertions or deletions and the number and positions of SNPs. RESULTS: The nucleotide structure of all 38 isolates is presented. Based on insertions and deletions and dissimilarity due to SNPs, the dataset of all the isolates has been qualitatively classified into three groups each having their own subgroups. These are the A-group with "regular" isolates (no insertions / deletions except for 5' and 3' ends), the B-group of isolates with "long insertions", and the C-group of isolates with "many individual" insertions and deletions. The isolate with the smallest average number of SNPs, compared to other isolates, has been identified (TWH). The density distribution of SNPs, insertions and deletions for each group or subgroup, as well as cumulatively for all the isolates is also presented, along with the gene map for TWH. Since individual SNPs may have occurred at random, positions corresponding to multiple SNPs (occurring in two or more isolates) are identified and presented. This result revises some previous results of a similar type. Amino acid changes caused by multiple SNPs are also identified (for the annotated sequences, as well as presupposed amino acid changes for non-annotated ones). Exact SNP positions for the isolates in each group or subgroup are presented. Finally, a phylogenetic tree for the SARS-CoV isolates has been produced using the CLUSTALW program, showing high compatibility with former qualitative classification. CONCLUSIONS: The comparative study of SARS-CoV isolates provides essential information for genome polymorphism, indication of strain differences and variants evolution. It may help with the development of effective treatment.