Automatic rule generation for protein annotation with the C4.5 data mining algorithm applied on SWISS-PROT.

MOTIVATION: The gap between the amount of newly submitted protein data and reliable functional annotation in public databases is growing. Traditional manual annotation by literature curation and sequence analysis tools without the use of automated annotation systems is not able to keep up with the ever increasing quantity of data that is submitted. Automated supplements to manually curated databases such as TrEMBL or GenPept cover raw data but provide only limited annotation. To improve this situation automatic tools are needed that support manual annotation, automatically increase the amount of reliable information and help to detect inconsistencies in manually generated annotations. RESULTS: A standard data mining algorithm was successfully applied to gain knowledge about the Keyword annotation in SWISS-PROT. 11 306 rules were generated, which are provided in a database and can be applied to yet unannotated protein sequences and viewed using a web browser. They rely on the taxonomy of the organism, in which the protein was found and on signature matches of its sequence. The statistical evaluation of the generated rules by cross-validation suggests that by applying them on arbitrary proteins 33% of their keyword annotation can be generated with an error rate of 1.5%. The coverage rate of the keyword annotation can be increased to 60% by tolerating a higher error rate of 5%. AVAILABILITY: The results of the automatic data mining process can be browsed on http://golgi.ebi.ac.uk:8080/Spearmint/ Source code is available upon request. CONTACT: kretsch@ebi.ac.uk.     

A database of unique protein sequence identifiers for proteome studies

In proteome studies, identification of proteins requires searching protein sequence databases. The public protein sequence databases (e.g., NCBInr, UniProt) each contain millions of entries, and private databases add thousands more. Although much of the sequence information in these databases is redundant, each database uses distinct identifiers for the identical protein sequence and often contains unique annotation information. Users of one database obtain a database-specific sequence identifier that is often difficult to reconcile with the identifiers from a different database. When multiple databases are used for searches or the databases being searched are updated frequently, interpreting the protein identifications and associated annotations can be problematic. We have developed a database of unique protein sequence identifiers called Sequence Globally Unique Identifiers (SEGUID) derived from primary protein sequences. These identifiers serve as a common link between multiple sequence databases and are resilient to annotation changes in either public or private databases throughout the lifetime of a given protein sequence. The SEGUID Database can be downloaded (http://bioinformatics.anl.gov/SEGUID/) or easily generated at any site with access to primary protein sequence databases. Since SEGUIDs are stable, predictions based on the primary sequence information (e.g., pI, Mr) can be calculated just once; we have generated approximately 500 different calculations for more than 2.5 million sequences. SEGUIDs are used to integrate MS and 2-DE data with bioinformatics information and provide the opportunity to search multiple protein sequence databases, thereby providing a higher probability of finding the most valid protein identifications.     

The role SWISS-PROT and TrEMBL play in the genome research environment

SWISS-PROT, a curated protein sequence data bank, contains not only sequence data but also annotation relevant to a particular sequence. The annotation added to each entry is done by a team of biologists and comes, primarily, from articles in journals reporting the actual sequencing and sometimes characterisation. Review articles and collaboration with external experts also play a role along with the use of secondary databases like PROSITE and Pfam in addition to a variety of feature prediction methods. Annotation added by these methods is checked for relevance and likelihood to a particular sequence. The onset of genome sequencing has led to a dramatic increase in sequence data to be included in SWISS-PROT. This has led to the production of TrEMBL (Translation of the EMBL database). TrEMBL consists of entries in a SWISS-PROT format that are derived from the translation of all coding sequences in the EMBL nucleotide sequence database, that are not in SWISS-PROT. Unlike SWISS-PROT entries those in TrEMBL are awaiting manual annotation. However, rather than just representing basic sequence and source information, steps have been taken to add features and annotation automatically. In taking these steps it is hoped that TrEMBL entries are enhanced with some indication as to what a protein is, could or may be.     

The MetaFam Server: a comprehensive protein family resource

MetaFam is a comprehensive relational database of protein family information. This web-accessible resource integrates data from several primary sequence and secondary protein family databases. By pooling together the information from these disparate sources, MetaFam is able to provide the most complete protein family sets available. Users are able to explore the interrelationships among these primary and secondary databases using a powerful graphical visualization tool, MetaFamView. Additionally, users can identify corresponding sequence entries among the sequence databases, obtain a quick summary of corresponding families (and their sequence members) among the family databases, and even attempt to classify their own unassigned sequences. Hypertext links to the appropriate source databases are provided at every level of navigation. Global family database statistics and information are also provided. Public access to the data is available at http://metafam.ahc.umn.edu/.     

A systematic comparison of protein structure classifications: SCOP, CATH and FSSP.

BACKGROUND: Several methods of structural classification have been developed to introduce some order to the large amount of data present in the Protein Data Bank. Such methods facilitate structural comparisons and provide a greater understanding of structure and function. The most widely used and comprehensive databases are SCOP, CATH and FSSP, which represent three unique methods of classifying protein structures: purely manual, a combination of manual and automated, and purely automated, respectively. In order to develop reliable template libraries and benchmarks for protein-fold recognition, a systematic comparison of these databases has been carried out to determine their overall agreement in classifying protein structures. RESULTS: Approximately two-thirds of the protein chains in each database are common to all three databases. Despite employing different methods, and basing their systems on different rules of protein structure and taxonomy, SCOP, CATH and FSSP agree on the majority of their classifications. Discrepancies and inconsistencies are accounted for by a small number of explanations. Other interesting features have been identified, and various differences between manual and automatic classification methods are presented. CONCLUSIONS: Using these databases requires an understanding of the rules upon which they are based; each method offers certain advantages depending on the biological requirements and knowledge of the user. The degree of discrepancy between the systems also has an impact on reliability of prediction methods that employ these schemes as benchmarks. To generate accurate fold templates for threading, we extract information from a consensus database, encompassing agreements between SCOP, CATH and FSSP.     

Protein Information and Knowledge Extractor: Discovering biological information from proteomics data

One of the main goals in proteomics is to solve biological and molecular questions regarding a set of identified proteins. In order to achieve this goal, one has to extract and collect the existing biological data from public repositories for every protein and afterward, analyze and organize the collected data. Due to the complexity of this task and the huge amount of data available, it is not possible to gather this information by hand, making it necessary to find automatic methods of data collection. Within a proteomic context, we have developed Protein Information and Knowledge Extractor (PIKE) which solves this problem by automatically accessing several public information systems and databases across the Internet. PIKE bioinformatics tool starts with a set of identified proteins, listed as the most common protein databases accession codes, and retrieves all relevant and updated information from the most relevant databases. Once the search is complete, PIKE summarizes the information for every single protein using several file formats that share and exchange the information with other software tools. It is our opinion that PIKE represents a great step forward for information procurement and drastically reduces manual database validation for large proteomic studies. It is available at http://proteo.cnb.csic.es/pike .     

Construction of validated, non-redundant composite protein sequence databases

A strategy has been developed for the construction of a validated, comprehensive composite protein sequence database. Entries are amalgamated from primary source data bases by a largely automated set of processes in which redundant and trivially different entries are eliminated. A modular approach has been adopted to allow scientific judgement to be used at each stage of database processing and amalgamation. Source databases are assigned a priority depending on the quality of sequence validation and commenting. Rejection of entries from the lower priority database, in each pairwise comparison of databases, is carried out according to optionally defined redundancy criteria based on sequence segment mismatches. Efficient algorithms for this methodology are embodied in the COMPO software system. COMPO has been applied for over 2 years in construction and regular updating of the OWL composite protein sequence database from the source databases NBRF-PIR, SWISS-PROT, a GenBank translation retrieved from the feature tables, NBRF-NEW, NEWAT86, PSD-KYOTO and the sequences contained in the Brookhaven protein structure databank. OWL is part of the ISIS integrated data resource of protein sequence and structure [Akrigg et al. (1988) Nature, 335, 745-746]. The modular nature of the integration process greatly facilitates the frequent updating of OWL following releases of the source databases. The extent of redundancy in these sources is revealed by the comparison process. The advantages of a robust composite database for sequence similarity searching and information retrieval are discussed.     

GenBank

GenBank (R) is a comprehensive sequence database that contains publicly available DNA sequences for more than 119 000 different organisms, obtained primarily through the submission of sequence data from individual laboratories and batch submissions from large-scale sequencing projects. Most submissions are made using the BankIt (web) or Sequin programs and accession numbers are assigned by GenBank staff upon receipt. Daily data exchange with the EMBL Data Library in the UK and the DNA Data Bank of Japan helps ensure worldwide coverage. GenBank is accessible through NCBI's retrieval system, Entrez, which integrates data from the major DNA and protein sequence databases along with taxonomy, genome, mapping, protein structure and domain information, and the biomedical journal literature via PubMed. BLAST provides sequence similarity searches of GenBank and other sequence databases. Complete bimonthly releases and daily updates of the GenBank database are available by FTP. To access GenBank and its related retrieval and analysis services, go to the NCBI home page at: http://www.ncbi.nlm.nih.gov.     

UniProt archive

UniProt Archive (UniParc) is the most comprehensive, non-redundant protein sequence database available. Its protein sequences are retrieved from predominant, publicly accessible resources. All new and updated protein sequences are collected and loaded daily into UniParc for full coverage. To avoid redundancy, each unique sequence is stored only once with a stable protein identifier, which can be used later in UniParc to identify the same protein in all source databases. When proteins are loaded into the database, database cross-references are created to link them to the origins of the sequences. As a result, performing a sequence search against UniParc is equivalent to performing the same search against all databases cross-referenced by UniParc. UniParc contains only protein sequences and database cross-references; all other information must be retrieved from the source databases.     

Functional and structural genomics using PEDANT

MOTIVATION: Enormous demand for fast and accurate analysis of biological sequences is fuelled by the pace of genome analysis efforts. There is also an acute need in reliable up-to-date genomic databases integrating both functional and structural information. Here we describe the current status of the PEDANT software system for high-throughput analysis of large biological sequence sets and the genome analysis server associated with it. RESULTS: The principal features of PEDANT are: (i) completely automatic processing of data using a wide range of bioinformatics methods, (ii) manual refinement of annotation, (iii) automatic and manual assignment of gene products to a number of functional and structural categories, (iv) extensive hyperlinked protein reports, and (v) advanced DNA and protein viewers. The system is easily extensible and allows to include custom methods, databases, and categories with minimal or no programming effort. PEDANT is actively used as a collaborative environment to support several on-going genome sequencing projects. The main purpose of the PEDANT genome database is to quickly disseminate well-organized information on completely sequenced and unfinished genomes. It currently includes 80 genomic sequences and in many cases serves as the only source of exhaustive information on a given genome. The database also acts as a vehicle for a number of research projects in bioinformatics. Using SQL queries, it is possible to correlate a large variety of pre-computed properties of gene products encoded in complete genomes with each other and compare them with data sets of special scientific interest. In particular, the availability of structural predictions for over 300 000 genomic proteins makes PEDANT the most extensive structural genomics resource available on the web.     

GeneReporter--sequence-based document retrieval and annotation

GeneReporter is a web tool that reports functional information and relevant literature on a protein-coding sequence of interest. Its purpose is to support both manual genome annotation and document retrieval. PubMed references corresponding to a sequence are detected by the extraction of query words from UniProt entries of homologous sequences. Data on protein families, domains, potential cofactors, structure, function, cellular localization, metabolic contribution and corresponding DNA binding sites complement the information on a given gene product of interest. Availability and implementation: GeneReporter is available at http://www.genereporter.tu-bs.de. The web site integrates databases and analysis tools as SOAP-based web services from the EBI (European Bioinformatics Institute) and NCBI (National Center for Biotechnology Information).     

In Silico Characterization of Proteins: UniProt,InterPro and Integr8

Nucleic acid sequences from genome sequencing projects are submitted as raw data, from which biologists attempt to elucidate the function of the predicted gene products. The protein sequences are stored in public databases, such as the UniProt Knowledgebase (UniProtKB), where curators try to add predicted and experimental functional information. Protein function prediction can be done using sequence similarity searches, but an alternative approach is to use protein signatures, which classify proteins into families and domains. The major protein signature databases are available through the integrated InterPro database, which provides a classification of UniProtKB sequences. As well as characterization of proteins through protein families, many researchers are interested in analyzing the complete set of proteins from a genome (i.e. the proteome), and there are databases and resources that provide non-redundant proteome sets and analyses of proteins from organisms with completely sequenced genomes. This article reviews the tools and resources available on the web for single and large-scale protein characterization and whole proteome analysis.     

CHPVDB ‐ a sequence annotation database for Chandipura Virus

Databases containing proteomic information have become indispensable for virology studies. As the gap between the amount of sequence information and functional characterization widens, increasing efforts are being directed to the development of databases. For virologist, it is therefore desirable to have a single data collection point which integrates research related data from different domains. CHPVDB is our effort to provide virologist such a one‐step information center. We describe herein the creation of CHPVDB, a new database that integrates information of different proteins in to a single resource. For basic curation of protein information, the database relies on features from other selected databases, servers and published reports. This database facilitates significant relationship between molecular analysis, cleavage sites, possible protein functional families assigned to different proteins of Chandipura virus (CHPV) by SVMProt and related tools.     

Subdivision of the MDR superfamily of medium-chain dehydrogenases/reductases through iterative hidden Markov model refinement

Background  

The Medium-chain Dehydrogenases/Reductases (MDR) form a protein superfamily whose size and complexity defeats traditional means of subclassification; it currently has over 15000 members in the databases, the pairwise sequence identity is typically around 25%, there are members from all kingdoms of life, the chain-lengths vary as does the oligomericity, and the members are partaking in a multitude of biological processes. There are profile hidden Markov models (HMMs) available for detecting MDR superfamily members, but none for determining which MDR family each protein belongs to. The current torrential influx of new sequence data enables elucidation of more and more protein families, and at an increasingly fine granularity. However, gathering good quality training data usually requires manual attention by experts and has therefore been the rate limiting step for expanding the number of available models.     

The Jalview Java alignment editor

Multiple sequence alignment remains a crucial method for understanding the function of groups of related nucleic acid and protein sequences. However, it is known that automatic multiple sequence alignments can often be improved by manual editing. Therefore, tools are needed to view and edit multiple sequence alignments. Due to growth in the sequence databases, multiple sequence alignments can often be large and difficult to view efficiently. The Jalview Java alignment editor is presented here, which enables fast viewing and editing of large multiple sequence alignments.     

The quest to deduce protein function from sequence: the role of pattern databases

In the wake of the numerous now-fruitful genome projects, we have witnessed a 'tsunami' of sequence data and with it the birth of the field of bioinformatics. Bioinformatics involves the application of information technology to the management and analysis of biological data. For many of us, this means that databases and their search tools have become an essential part of the research environment. However, the rate of sequence generation and the haphazard proliferation of databases have made it difficult to keep pace with developments, even for the cognoscenti. Moreover, increasing amounts of sequence information do not necessarily equate with an increase in knowledge, and in the panic to automate the route from raw data to biological insight, we may be generating and propagating innumerable errors in our precious databases. In the genome era upon us, researchers want rapid, easy-to-use, reliable tools for functional characterisation of newly determined sequences. For the pharmaceutical industry in particular, the Pandora's box of bioinformatics harbours an information-rich nugget, ripe with potential drug targets and possible new avenues for the development of therapeutic agents. This review outlines the current status of the major pattern databases now used routinely in the analysis of protein sequences. The review is divided into three main sections. In the first, commonly used terms are defined and the methods behind the databases are briefly described; in the second, the structure and content of the principal pattern databases are discussed; and in the final part, several alignment databases, which are frequently confused with pattern databases, are mentioned. For the new-comer, the array of resources, the range of methods behind them and the different tools required to search them can be confusing. The review therefore also briefly mentions a current international endeavour to integrate the diverse databases, which effort should facilitate sequence analysis in the future. This is particularly important for target-discovery programmes, where the challenge is to rationalise the enormous numbers of potential targets generated by sequence database searches. This problem may be addressed, at least in part, by reducing search outputs to the more focused and manageable subsets suggested by searches of integrated groups of family-specific pattern databases.     

Protein Information Crawler (PIC): extensive spidering of multiple protein information resources for large protein sets

Proteomic studies often produce sets of hundreds of proteins. Bioinformatic information for these large protein sets must be collected from multiple online resources. Protein Information Crawler (PIC) automatically bulk-collects such data from multiple databases and prediction servers, based on National Center for Biotechnology Information (NCBI) gi numbers or accession numbers, and summarizes them in a Microsoft Excel spreadsheet and/or HTML table. PIC greatly accelerates information procurement, helps to build customized protein information databases and drastically reduces manual database investigation in extensive proteomic studies. Availability: http://www.zoo.uni-heidelberg.de/mfa/PIC.     

PhosphoBase: a database of phosphorylation sites.

PhosphoBase is a database of experimentally verified phosphorylation sites. Version 1.0 contains 156 entries and 398 experimentally determined phosphorylation sites. Entries are compiled and revised from the literature and from major protein sequence databases such as SwissProt and PIR. The entries provide information about the phosphoprotein and the exact position of its phosphorylation sites. Furthermore, part of the entries contain information about kinetic data obtained from enzyme assays on specific peptides. To illustrate the use of data extracted from PhosphoBase we present a sequence logo displaying the overall conservation of positions around serines phosphorylated by protein kinase A (PKA). PhosphoBase is available on the WWW at http://www.cbs.dtu.dk/databases/PhosphoBase/     

Human cellular protein patterns and their link to genome DNA sequence data: usefulness of two-dimensional gel electrophoresis and microsequencing.

Analysis of cellular protein patterns by computer-aided 2-dimensional gel electrophoresis together with recent advances in protein sequence analysis have made possible the establishment of comprehensive 2-dimensional gel protein databases that may link protein and DNA information and that offer a global approach to the study of the cell. Using the integrated approach offered by 2-dimensional gel protein databases it is now possible to reveal phenotype specific protein (or proteins), to microsequence them, to search for homology with previously identified proteins, to clone the cDNAs, to assign partial protein sequence to genes for which the full DNA sequence and the chromosome location is known, and to study the regulatory properties and function of groups of proteins that are coordinately expressed in a given biological process. Human 2-dimensional gel protein databases are becoming increasingly important in view of the concerted effort to map and sequence the entire genome.