To study basic principles of transcriptome organization in bacteria, we analyzed one of the smallest self-replicating organisms, Mycoplasma pneumoniae. We combined strand-specific tiling arrays, complemented by transcriptome sequencing, with more than 252 spotted arrays. We detected 117 previously undescribed, mostly noncoding transcripts, 89 of them in antisense configuration to known genes. We identified 341 operons, of which 139 are polycistronic; almost half of the latter show decaying expression in a staircase-like manner. Under various conditions, operons could be divided into 447 smaller transcriptional units, resulting in many alternative transcripts. Frequent antisense transcripts, alternative transcripts, and multiple regulators per gene imply a highly dynamic transcriptome, more similar to that of eukaryotes than previously thought.

SMART (Simple Modular Architecture Research Tool) is a web tool (http://smart.embl.de/) for the identification and annotation of protein domains, and provides a platform for the comparative study of complex domain architectures in genes and proteins. The January 2004 release of SMART contains 685 protein domains. New developments in SMART are centred on the integration of data from completed metazoan genomes. SMART now uses predicted proteins from complete genomes in its source sequence databases, and integrates these with predictions of orthology. New visualization tools have been developed to allow analysis of gene intron-exon structure within the context of protein domain structure, and to align these displays to provide schematic comparisons of orthologous genes, or multiple transcripts from the same gene. Other improvements include the ability to query SMART by Gene Ontology terms, improved structure database searching and batch retrieval of multiple entries.

In all genome-sequencing projects completed to date, a considerable number of 'gaps' have been found in the biochemical pathways of the respective species. In many instances, missing enzymes are displaced by analogs, functionally equivalent proteins that have evolved independently and lack sequence and structural similarity. Here we fill such gaps by analyzing anticorrelating occurrences of genes across species. Our approach, applied to the thiamin biosynthesis pathway comprising approximately 15 catalytic steps, predicts seven instances in which known enzymes have been displaced by analogous proteins. So far we have verified four predictions by genetic complementation, including three proteins for which there was no previous experimental evidence of a role in the thiamin biosynthesis pathway. For one hypothetical protein, biochemical characterization confirmed the predicted thiamin phosphate synthase (ThiE) activity. The results demonstrate the ability of our computational approach to predict specific functions without taking into account sequence similarity.

The evolution of enzymes and pathways is under debate. Recent studies show that recruitment of single enzymes from different pathways could be the driving force for pathway evolution. Other mechanisms of evolution, such as pathway duplication, enzyme specialization, de novo invention of pathways or retro-evolution of pathways, appear to be less abundant. Twenty percent of enzyme superfamilies are quite variable, not only in changing reaction chemistry or metabolite type but in changing both at the same time. These variable superfamilies account for nearly half of all known reactions. The most frequently occurring metabolites provide a helping hand for such changes because they can be accommodated by many enzyme superfamilies. Thus, a picture is emerging in which new pathways are evolving from central metabolites by preference, thereby keeping the overall topology of the metabolic network.

Functional links between proteins can often be inferred from genomic associations between the genes that encode them: groups of genes that are required for the same function tend to show similar species coverage, are often located in close proximity on the genome (in prokaryotes), and tend to be involved in gene-fusion events. The database STRING is a precomputed global resource for the exploration and analysis of these associations. Since the three types of evidence differ conceptually, and the number of predicted interactions is very large, it is essential to be able to assess and compare the significance of individual predictions. Thus, STRING contains a unique scoring-framework based on benchmarks of the different types of associations against a common reference set, integrated in a single confidence score per prediction. The graphical representation of the network of inferred, weighted protein interactions provides a high-level view of functional linkage, facilitating the analysis of modularity in biological processes. STRING is updated continuously, and currently contains 261 033 orthologs in 89 fully sequenced genomes. The database predicts functional interactions at an expected level of accuracy of at least 80% for more than half of the genes; it is online at http://www.bork.embl-heidelberg.de/STRING/.

The WEB tool "AnDom" assigns to a given protein sequence all experimentally determined structural domains contained within it, including multidomain and large proteins. The server uses profile specific matrices from custom generated multiple sequence alignments of all known SCOP domains (SCOP version 1.50). Prediction time is short allowing numerous applications for structural genomics including investigation of complex eucaryotic protein families. The WWW server is at http://www.bork.embl-heidelberg.de/AnDom, and profiles can be downloaded at ftp.bork.embl-heidelberg.de/pub/users/schmidt/AnDom.

By exploiting the rapid increase in available sequence data, the definition of medically relevant protein targets has been improved by a combination of: (i) differential genome analysis (target list): and (ii) analysis of individual proteins (target analysis). Fast sequence comparisons, data mining, and genetic algorithms further promote these procedures. Mycobacterium tuberculosis proteins were chosen as applied examples.