The first step in understanding the molecular biology of an inherited disease is to identify which gene or genes are carrying variants. This process starts with locating the mutations in a chromosomal band, as narrow as possible, and follows with the manual analysis of all the genes mapping in this region. Usually this is not an easy task, but it can be facilitated by complementary computational approaches that evaluate all genes in a region of interest. We present here a method that combines literature mining, gene annotations, and sequence homology searches to prioritize candidate genes involved in a given genetic disorder. The method progresses in two steps. Firstly, we compute associations of molecular and phenotypic features as taken from MEDLINE. Secondly, for a disease with a given phenotype and linked to a chromosomal region, sequence homology based searches are carried on the chromosomal region to identify potential candidates that are scored using the precomputed associations. The scoring of associations between biological concepts using links across databases can be extended to other databases in Molecular Biology and to nondisease phenotypes.

One of the major challenges of functional genomics is to unravel the connection between genotype and phenotype. So far no global analysis has attempted to explore those connections in the light of the large phenotypic variability seen in nature. Here, we use an unsupervised, systematic approach for associating genes and phenotypic characteristics that combines literature mining with comparative genome analysis. We first mine the MEDLINE literature database for terms that reflect phenotypic similarities of species. Subsequently we predict the likely genomic determinants: genes specifically present in the respective genomes. In a global analysis involving 92 prokaryotic genomes we retrieve 323 clusters containing a total of 2,700 significant gene-phenotype associations. Some clusters contain mostly known relationships, such as genes involved in motility or plant degradation, often with additional hypothetical proteins associated with those phenotypes. Other clusters comprise unexpected associations; for example, a group of terms related to food and spoilage is linked to genes predicted to be involved in bacterial food poisoning. Among the clusters, we observe an enrichment of pathogenicity-related associations, suggesting that the approach reveals many novel genes likely to play a role in infectious diseases.

BACKGROUND: Human inherited diseases can be associated by genetic linkage with one or more genomic regions. The availability of the complete sequence of the human genome allows examining those locations for an associated gene. We previously developed an algorithm to prioritize genes on a chromosomal region according to their possible relation to an inherited disease using a combination of data mining on biomedical databases and gene sequence analysis. RESULTS: We have implemented this method as a web application in our site G2D (Genes to Diseases). It allows users to inspect any region of the human genome to find candidates genes related to a genetic disease of their interest. In addition, the G2D server includes pre-computed analyses of candidate genes for 552 linked monogenic diseases without an associated gene, and the analysis of 18 asthma loci. CONCLUSIONS: G2D can be publicly accessed at http://www.ogic.ca/projects/g2d_2/.

Whole-genome sequences are now available for >100 bacterial species, giving unprecedented power to comparative genomics approaches. We have applied genome-context methods to predict target processes that are regulated by transcription factors (TFs). Of 128 orthologous groups of proteins annotated as TFs, to date, 36 are functionally uncharacterized; in our analysis we predict a probable cellular target process or biochemical pathway for half of these functionally uncharacterized TFs.

MOTIVATION: The description of genes in databases by keywords helps the non-specialist to quickly grasp the properties of a gene and increases the efficiency of computational tools that are applied to gene data (e.g. searching a gene database for sequences related to a particular biological process). However, the association of keywords to genes or protein sequences is a difficult process that ultimately implies examination of the literature related to a gene. RESULTS: To support this task, we present a procedure to derive keywords from the set of scientific abstracts related to a gene. Our system is based on the automated extraction of mappings between related terms from different databases using a model of fuzzy associations that can be applied with all generality to any pair of linked databases. We tested the system by annotating genes of the SWISS-PROT database with keywords derived from the abstracts linked to their entries (stored in the MEDLINE database of scientific references). The performance of the annotation procedure was much better for SWISS-PROT keywords (recall of 47%, precision of 68%) than for Gene Ontology terms (recall of 8%, precision of 67%). AVAILABILITY: The algorithm can be publicly accessed and used for the annotation of sequences through a web server at http://www.bork.embl.de/kat

MOTIVATION: Genome-wide functional annotation either by manual or automatic means has raised considerable concerns regarding the accuracy of assignments and the reproducibility of methodologies. In addition, a performance evaluation of automated systems that attempt to tackle sequence analyses rapidly and reproducibly is generally missing. In order to quantify the accuracy and reproducibility of function assignments on a genome-wide scale, we have re-annotated the entire genome sequence of Chlamydia trachomatis (serovar D), in a collaborative manner. RESULTS: We have encoded all annotations in a structured format to allow further comparison and data exchange and have used a scale that records the different levels of potential annotation errors according to their propensity to propagate in the database due to transitive function assignments. We conclude that genome annotation may entail a considerable amount of errors, ranging from simple typographical errors to complex sequence analysis problems. The most surprising result of this comparative study is that automatic systems might perform as well as the teams of experts annotating genome sequences.

Entries in biological databases are usually linked to scientific references. To generate those links and to keep them up-to-date, database maintainers have to continuously scan the scientific literature to select references that are relevant for each single database entry. The continuous growth of both the corpus of scientific literature and the size of biological databases makes this task very hard. We present a protocol intended to assist the updating of an existing set of literature (abstract) links from a single database entry with new references. It consists of taking the set of MEDLINE neighbour references of the existing linked abstracts and evaluating their relevance according to the existing set of abstracts. To test the applicability of the algorithm, we did a simple benchmark of the system using the references associated with the entries of a protein domain database. Human experts found the references that the algorithm scored highly were more relevant to the database entry than those scored lowly, suggesting that the algorithm was useful.

As scientific literature databases like MEDLINE increase in size, so does the time required to search them. Scientists must frequently inspect long lists of references manually, often just reading the titles. XplorMed is a web tool that aids MEDLINE searching by summarizing the subjects contained in the results, thus allowing users to focus on subjects of interest. Here we describe new features added to XplorMed during the last 2 years (http://www.bork.embl-heidelberg.de/xplormed/).

BACKGROUND: To date, many of the methods for information extraction of biological information from scientific articles are restricted to the abstract of the article. However, full text articles in electronic version, which offer larger sources of data, are currently available. Several questions arise as to whether the effort of scanning full text articles is worthy, or whether the information that can be extracted from the different sections of an article can be relevant. RESULTS: In this work we addressed those questions showing that the keyword content of the different sections of a standard scientific article (abstract, introduction, methods, results, and discussion) is very heterogeneous. CONCLUSIONS: Although the abstract contains the best ratio of keywords per total of words, other sections of the article may be a better source of biologically relevant data.

BACKGROUND: Iron uptake from the host is essential for bacteria that infect animals. To find potential targets for drugs active against pathogenic bacteria, we have searched all completely sequenced genomes of pathogenic bacteria for genes relevant for iron transport. RESULTS: We identified a protein domain that appears in variable copy number in bacterial genes that are usually in the vicinity of a putative Fe3+ siderophore transporter. Accordingly, we have denoted this domain NEAT for 'near transporter'. Most of the bacterial species containing this domain are pathogenic. Sequence features indicate that the domain is anchored to the extracellular side of the membrane. The domain seems to be under high selective pressure for rapid independent duplications that are typical of sequences involved in signaling and binding. CONCLUSIONS: The NEAT domain might be functionally related to iron transport. The taxonomic specificity of this domain and its predicted extracellular position could make it an interesting target for designing new drugs against some highly pathogenic bacteria.

Although approximately one-quarter of the roughly 4,000 genetically inherited diseases currently recorded in respective databases (LocusLink, OMIM) are already linked to a region of the human genome, about 450 have no known associated gene. Finding disease-related genes requires laborious examination of hundreds of possible candidate genes (sometimes, these are not even annotated; see, for example, refs 3,4). The public availability of the human genome draft sequence has fostered new strategies to map molecular functional features of gene products to complex phenotypic descriptions, such as those of genetically inherited diseases. Owing to recent progress in the systematic annotation of genes using controlled vocabularies, we have developed a scoring system for the possible functional relationships of human genes to 455 genetically inherited diseases that have been mapped to chromosomal regions without assignment of a particular gene. In a benchmark of the system with 100 known disease-associated genes, the disease-associated gene was among the 8 best-scoring genes with a 25% chance, and among the best 30 genes with a 50% chance, showing that there is a relationship between the score of a gene and its likelihood of being associated with a particular disease. The scoring also indicates that for some diseases, the chance of identifying the underlying gene is higher.

XplorMed is a publicly available web tool conceived to make life easier for MEDLINE(c) users looking for scientific information. Searching scientific literature is an information retrieval problem. Abstracts that are of possible interest to the user are usually selected by a keyword search followed by manual screening, which often results in the retrieval of a large number of abstracts. Interesting references can be buried among irrelevant ones because of nonspecific queries. XplorMed is intended to extract dependency relations between the words of the abstracts. These relations can be filtered and arranged to deduce different subjects in the query and offer a condensed view of the abstract, allowing users to select texts of interest without having to read them all. XplorMed is available http://www.bork. embl-heidelberg.de/xplormed.

The increase of information in biology makes it difficult for researchers in any field to keep current with the literature. The MEDLINE database of scientific abstracts can be quickly scanned using electronic mechanisms. Potentially interesting abstracts can be selected by matching words joined by Boolean operators. However this means of selecting documents is not optimal. Nonspecific queries have to be effected, resulting in large numbers of irrelevant abstracts that have to be manually scanned To facilitate this analysis, we have developed a system that compiles a summary of subjects and related documents on the results of a MEDLINE query. For this, we have applied a fuzzy binary relation formalism that deduces relations between words present in a set of abstracts preprocessed with a standard grammatical tagger. Those relations are used to derive ensembles of related words and their associated subsets of abstracts. The algorithm can be used publicly at http:// www.bork.embl-heidelberg.de/xplormed/.

In the sequences released by the Arabidopsis Genome Initiative (AGI), we have discovered a new large gene family (48 genes as of July 2000). A detailed computational and biochemical analysis of the predicted gene products reveals a novel family of plant F-box proteins, where the amino (N)-terminal F-box motif is followed by four kelch repeats and a characteristic carboxy-terminal domain. F-box proteins are an expanding family of eukaryotic proteins, which have been shown in some cases to be critical for the controlled degradation of cellular regulatory proteins via the ubiquitin pathway. The F-box motif of the At5g48990 gene product, a member of the family, was shown to be functionally active by its ability to mediate the in vitro interaction between At5g48990 and ASK1 proteins. F-box proteins specifically recruit the targets to be ubiquitinated, mainly through protein-protein interaction modules such as WD-40 domains or leucine-rich repeats (LRRs). The kelch repeats of the family described here form a potential protein-protein interaction domain, as molecular modelling of the kelch repeats according to the galactose oxidase crystal structure (the only solved structure containing kelch repeats) predicts a beta-propeller. The identification of this family of F-box proteins greatly expands the field of plant F-box proteins and suggests that controlled degradation of cellular proteins via the ubiquitin pathway could play a critical role in multiple plant cellular processes.

Retinal plasticity has been shown in the adult visual nervous system in mammals. Following a retinal lesion (scotoma) there is a reorganization of the cortical receptive field distribution: cortical neurons selective to visual stimuli in the area of the visual field corresponding to the retinal lesion, become selective to other parts of the visual field. In this work, we study this effect with a self-organizing neural network. In a first stage, the network reaches a pattern of connectivity that represents normal development of neuronal selectivity. The scotoma is simulated by perturbing accordingly the properties of a region of the input layer representing the retina. The system evolves to a new receptive field distribution mainly by means of the reorganization of the intra cortical connectivity. No major change of the geniculo cortical connectivity is detected. This may explain the surprisingly short time scale of the event.

Internal repetition within proteins has been a successful strategem on multiple separate occasions throughout evolution. Such protein repeats possess regular secondary structures and form multirepeat assemblies in three dimensions of diverse sizes and functions. In general, however, internal repetition affords a protein enhanced evolutionary prospects due to an enlargement of its available binding surface area. Constraints on sequence conservation appear to be relatively lax, due to binding functions ensuing from multiple, rather than, single repeats. Considerable sequence divergence as well as the short lengths of sequence repeats mean that repeat detection can be a particularly arduous task. We also consider the conundrum of how multiple repeats, which show strong structural and functional interdependencies, ever evolved from a single repeat ancestor. In this review, we illustrate each of these points by referring to six prolific repeat types (repeats in beta-propellers and beta-trefoils and tetratricopeptide, ankyrin, armadillo/HEAT, and leucine-rich repeats) and in other less-prolific but nonetheless interesting repeats.

ARM and HEAT motifs are tandemly repeated sequences of approximately 50 amino acid residues that occur in a wide variety of eukaryotic proteins. An exhaustive search of sequence databases detected new family members and revealed that at least 1 in 500 eukaryotic protein sequences contain such repeats. It also rendered the similarity between ARM and HEAT repeats, believed to be evolutionarily related, readily apparent. All the proteins identified in the database searches could be clustered by sequence similarity into four groups: canonical ARM-repeat proteins and three groups of the more divergent HEAT-repeat proteins. This allowed us to build improved sequence profiles for the automatic detection of repeat motifs. Inspection of these profiles indicated that the individual repeat motifs of all four classes share a common set of seven highly conserved hydrophobic residues, which in proteins of known three-dimensional structure are buried within or between repeats. However, the motifs differ at several specific residue positions, suggesting important structural or functional differences among the classes. Our results illustrate that ARM and HEAT-repeat proteins, while having a common phylogenetic origin, have since diverged significantly. We discuss evolutionary scenarios that could account for the great diversity of repeats observed.

To assess how automatic function assignment will contribute to genome annotation in the next five years, we have performed an analysis of 31 available genome sequences. An emerging pattern is that function can be predicted for almost two-thirds of the 73,500 genes that were analyzed. Despite progress in computational biology, there will always be a great need for large-scale experimental determination of protein function.

The most frequent access to the MEDLINE database of scientific abstracts is by keyword search. However, this is often not sufficient because although the user might find all the useful abstracts, these are buried in hundreds that are irrelevant. The exploratory tool XplorMed has been developed to analyse the result of any MEDLINE query. It suggests main groups of related topics and documents, sparing the user the need of reading all abstracts.

We review data mining techniques in molecular biology, specifically those that extract information from the scientific literature itself. As more of the biological literature is published electronically, there is an opportunity, and even a need, to automatically summarize the literature in a customized way, for example by associating keywords to a topic. These keywords can be extracted from relevant publications. The process of keyword extraction can be automated and optimized to keep literature pointers automatically up-to-date or to filter relevant information from the literature. To illustrate these points, OMIM (Online Mendelian Inheritance in Man), a database of human inherited diseases, was linked to the literature and keywords were derived that covered distinct aspects such as genetic information on the one hand and disease-specific protein and phenotypic information on the other. They were used to extract information that is helpful for keeping entries about disease up-to-date.

Short protein repeats, frequently with a length between 20 and 40 residues, represent a significant fraction of known proteins. Many repeats appear to possess high amino acid substitution rates and thus recognition of repeat homologues is highly problematic. Even if the presence of a certain repeat family is known, the exact locations and the number of repetitive units often cannot be determined using current methods. We have devised an iterative algorithm based on optimal and sub-optimal score distributions from profile analysis that estimates the significance of all repeats that are detected in a single sequence. This procedure allows the identification of homologues at alignment scores lower than the highest optimal alignment score for non-homologous sequences. The method has been used to investigate the occurrence of eleven families of repeats in Saccharomyces cerevisiae, Caenorhabditis elegans and Homo sapiens accounting for 1055, 2205 and 2320 repeats, respectively. For these examples, the method is both more sensitive and more selective than conventional homology search procedures. The method allowed the detection in the SwissProt database of more than 2000 previously unrecognised repeats belonging to the 11 families. In addition, the method was used to merge several repeat families that previously were supposed to be distinct, indicating common phylogenetic origins for these families.

Four years after the original sequence submission, we have re-annotated the genome of Mycoplasma pneumoniae to incorporate novel data. The total number of ORFss has been increased from 677 to 688 (10 new proteins were predicted in intergenic regions, two further were newly identified by mass spectrometry and one protein ORF was dismissed) and the number of RNAs from 39 to 42 genes. For 19 of the now 35 tRNAs and for six other functional RNAs the exact genome positions were re-annotated and two new tRNA(Leu) and a small 200 nt RNA were identified. Sixteen protein reading frames were extended and eight shortened. For each ORF a consistent annotation vocabulary has been introduced. Annotation reasoning, annotation categories and comparisons to other published data on M.pneumoniae functional assignments are given. Experimental evidence includes 2-dimensional gel electrophoresis in combination with mass spectrometry as well as gene expression data from this study. Compared to the original annotation, we increased the number of proteins with predicted functional features from 349 to 458. The increase includes 36 new predictions and 73 protein assignments confirmed by the published literature. Furthermore, there are 23 reductions and 30 additions with respect to the previous annotation. mRNA expression data support transcription of 184 of the functionally unassigned reading frames.

SUMMARY: Network Analysis Interface for Linking HMMER results (NAIL) is a web-based tool for the analysis of results from a HMMER protein database-search. NAIL facilitates the selection of protein hits and the creation of an alignment, which can be used for a new sequence similarity search.

I present in this work an algorithm for deriving protein functional annotations which are position-specific. The input is based on the results of a sequence similarity search of the query sequence against a sequence database. Strings of words are extracted from the descriptions of the proteins, and the correlation between proteins having the same descriptors and the amino acid conservation is used to compute a score that indicates which descriptor is likely to describe better the function of each particular residue. Analysis of the score curves and comparison of different functions allows an easy detection of parts of the sequence associated to different function. Different levels of functional specificity can be compared, allowing to choose the one that suits better the function of the protein. Immediate applications of this algorithm are, support for (automated) methods of protein functional annotation, and database coherence check.

MOTIVATION: Large-scale genome projects generate a rapidly increasing number of sequences, most of them biochemically uncharacterized. Research in bioinformatics contributes to the development of methods for the computational characterization of these sequences. However, the installation and application of these methods require experience and are time consuming. RESULTS: We present here an automatic system for preliminary functional annotation of protein sequences that has been applied to the analysis of sets of sequences from complete genomes, both to refine overall performance and to make new discoveries comparable to those made by human experts. The GeneQuiz system includes a Web-based browser that allows examination of the evidence leading to an automatic annotation and offers additional information, views of the results, and links to biological databases that complement the automatic analysis. System structure and operating principles concerning the use of multiple sequence databases, underlying sequence analysis tools, lexical analyses of database annotations and decision criteria for functional assignments are detailed. The system makes automatic quality assessments of results based on prior experience with the underlying sequence analysis tools; overall error rates in functional assignment are estimated at 2.5-5% for cases annotated with highest reliability ('clear' cases). Sources of over-interpretation of results are discussed with proposals for improvement. A conservative definition for reporting 'new findings' that takes account of database maturity is presented along with examples of possible kinds of discoveries (new function, family and superfamily) made by the system. System performance in relation to sequence database coverage, database dynamics and database search methods is analysed, demonstrating the inherent advantages of an integrated automatic approach using multiple databases and search methods applied in an objective and repeatable manner. AVAILABILITY: The GeneQuiz system is publicly available for analysis of protein sequences through a Web server at http://www.sander.ebi.ac. uk/gqsrv/submit

The evolutionary divergence among the three major domains of life can now be addressed through the first set of complete genomes from representative species. These model species from the three domains of life, Haemophilus influenzae for Bacteria, Saccharomyces cerevisiae for Eukarya, and Methanococcus jannaschii for Archaea, provide the basis for a universal functional classification and analysis. We have chosen 13 functional classes and three superclasses (ENERGY, COMMUNICATION and INFORMATION) as global descriptors of protein function. Compositional comparison of the three complete genomes reveals that functional classes are ubiquitous yet diverse in the three domains of life. Proteins related with ENERGY processes are generally represented in all three domains, while those related with COMMUNICATION represent the most distinctive functional feature of each single domain. Finally, functions related with INFORMATION processing (translation, transcription, and replication) show a complex behaviour. In Archaea, proteins in this superclass are related with proteins in either Eukarya or Bacteria, as recognized previously. The distribution of functional classes in the three domains accurately reflects the principal characteristics of cellular life forms.

We describe the basic design of a system for automatic detection of protein-protein interactions extracted from scientific abstracts. By restricting the problem domain and imposing a number of strong assumptions which include pre-specified protein names and a limited set of verbs that represent actions, we show that it is possible to perform accurate information extraction. The performance of the system is evaluated with different cases of real-world interaction networks, including the Drosophila cell cycle control. The results obtained computationally are in good agreement with current biological knowledge and demonstrate the feasibility of developing a fully automated system able to describe networks of protein interactions with sufficient accuracy.

MOTIVATION: Annotation of the biological function of different protein sequences is a time-consuming process currently performed by human experts. Genome analysis tools encounter great difficulty in performing this task. Database curators, developers of genome analysis tools and biologists in general could benefit from access to tools able to suggest functional annotations and facilitate access to functional information. APPROACH: We present here the first prototype of a system for the automatic annotation of protein function. The system is triggered by collections of s related to a given protein, and it is able to extract biological information directly from scientific literature, i.e. MEDLINE abstracts. Relevant keywords are selected by their relative accumulation in comparison with a domain-specific background distribution. Simultaneously, the most representative sentences and MEDLINE abstracts are selected and presented to the end-user. Evolutionary information is considered as a predominant characteristic in the domain of protein function. Our system consequently extracts domain-specific information from the analysis of a set of protein families. RESULTS: The system has been tested with different protein families, of which three examples are discussed in detail here: 'ataxia-telangiectasia associated protein', 'ran GTPase' and 'carbonic anhydrase'. We found generally good correlation between the amount of information provided to the system and the quality of the annotations. Finally, the current limitations and future developments of the system are discussed. AVAILABILITY: The current system can be considered as a prototype system. As such, it can be accessed as a server at http://columba.ebi.ac. uk:8765/andrade/abx. The system accepts text related to the protein or proteins to be evaluated (optimally, the result of a MEDLINE search by keyword) and the results are returned in the form of Web pages for keywords, sentences and s. SUPPLEMENTARY INFORMATION: Web pages containing full information on the examples mentioned in the text are available at: http://www.cnb.uam.es/ approximately cnbprot/keywords/ CONTACT: valencia@cnb.uam.es

MOTIVATION: The explosive growth of the biological sequences databases stimulated by genome projects has modified the framework of several applications in the biological sequence analysis area. In most cases, this new scenario is characterized by studies on large sets of sequences, suggesting the need for effective and automatic methods for their clustering. A more effective clustering of the database could be followed by the application of common family analysis schemes to the groups so formed. RESULTS: In this work, we present a new strategy to reduce the computational cost associated with the clustering of large sets of sequences which are expected to contain several families. The strategy is based on the grouping of the sequences into families by using a dynamic threshold on a pairwise sequence similarity criterion. Routine clustering of large data sets can now be done very efficiently. The method developed here achieves a computational space reduction of about an order of magnitude over more traditional ones of all-versus-all comparisons. The outcome of this approach produces family groupings that reproduce closely already accepted biological results. Our work includes a parallel implementation for distributed memory multiprocessors with a dynamic scheduling strategy for performance optimization. AVAILABILITY: By anonymous ftp at ftp.ac.uma.es (/pub/ots/pCluster directory), or from our Web site http://www.cnb. uam.es/www/software/software_index.html CONTACT: ots@ac.uma.es