To investigate the occurrence of glycosylphosphatidylinositol (GPI) lipid anchor modification in various taxonomic ranges, potential substrate proteins have been searched for in completely sequenced genomes. We applied the big-pi predictor for the recognition of propeptide cleavage and anchor attachment sites with a new, generalized analytical form of the extreme-value distribution for evaluating false-positive prediction rates. (i) We find that GPI modification is present among lower and higher Eukaryota (approximately 0.5% of all proteins) but it seems absent in all eubacterial and three archaeobacterial species studied. Four other archaean genomes appear to encode such a fraction of substrate proteins (in the range of eukaryots) that they cannot be explained as false-positive predictions. This result supports the possible existence of GPI anchor modification in an archaean subgroup. (ii) The frequency of GPI-modified proteins on various chromosomes of a given eukaryotic species is different. (iii) Lists of potentially GPI-modified proteins in complete genomes with their predicted cleavage sites are available at http://mendel.imp.univie.ac.at/gpi/gpi_genomes.html. (iv) Orthologues of known transamidase subunits have been found only for EUKARYA: Inconsistencies in domain structure among homologues some of which may indicate sequencing errors are described. We present a refined model of the transamidase complex.

Glycosylphosphatidylinositol (GPI) lipid anchoring is a common posttranslational modification known mainly from extracellular eukaryotic proteins. Attachment of the GPI moiety to the carboxyl terminus (omega-site) of the polypeptide follows after proteolytic cleavage of a C-terminal propeptide. For the first time, a new prediction technique locating potential GPI-modification sites in precursor sequences has been applied for large-scale protein sequence database searches. The composite prediction function (with separate parametrisation for metazoan and protozoan proteins) consists of terms evaluating both amino acid type preferences at sequence positions near a supposed omega-site as well as the concordance with general physical properties encoded in multi-residue correlation within the motif sequence. The latter terms are especially successful in rejecting non-appropriate sequences from consideration. The algorithm has been validated with a self-consistency and two jack-knife tests for the learning set of fully annotated sequences from the SWISS-PROT database as well as with a newly created database "big-Pi" (more than 300 GPI-motif mutations extracted from original literature sources). The accuracy of predicting the effect of mutations in the GPI sequence motif was above 83 %. Lists of potential precursor proteins which are non-annotated in SWISS-PROT and SPTrEMBL are presented on the WWW-page http://www.embl-heidelberg.de/beisenha/gpi/gpi_p rediction. html The algorithm has been implemented in the prototype software "big-Pi predictor" which may find application as a genome annotation and target selection tool.

MOTIVATION: Computer-based selection of entries from sequence databases with respect to a related functional description, e.g. with respect to a common cellular localization or contributing to the same phenotypic function, is a difficult task. Automatic semantic analysis of annotations is not only hampered by incomplete functional assignments. A major problem is that annotations are written in a rich, non-formalized language and are meant for reading by a human expert. This person can extract from the text considerably more information than is immediately apparent due to his extended biological background knowledge and logical reasoning. APPROACH: A technique of automated annotation evaluation based on a combination of lexical analysis and the usage of biological rule libraries has been developed. The proposed algorithm generates new functional descriptors from the annotation of a given entry using the semantic units of the annotation as prepositions for implications executed in accordance with the rule library. RESULTS: The prototype of a software system, the Meta_A(nnotator) program, is described and the results of its application to sequence attribute assignment and sequence selection problems, such as cellular localization and sequence domain annotation of SWISS-PROT entries, are presented. The current software version assigns useful subcellular localization qualifiers to approximately 88% of all SWISS-PROT entries. As shown by demonstrative examples, the combination of sequence and annotation analysis is a powerful approach for the detection of mutual annotation/sequence inconsistencies. AVAILABILITY: Results for the cellular localization assignment can be viewed at the URL http://www.bork. embl-heidelberg.de/CELL_LOC/CELL_LOC.html.

Sequence weighting techniques are aimed at balancing redundant observed information from subsets of similar sequences in multiple alignments. Traditional approaches apply the same weight to all positions of a given sequence, hence equal efficiency of phylogenetic changes is assumed along the whole sequence. This restrictive assumption is not required for the new method PSIC (position-specific independent counts) described in this paper. The number of independent observations (counts) of an amino acid type at a given alignment position is calculated from the overall similarity of the sequences that share the amino acid type at this position with the help of statistical concepts. This approach allows the fast computation of position-specific sequence weights even for alignments containing hundreds of sequences. The PSIC approach has been applied to profile extraction and to the fold family assignment of protein sequences with known structures. Our method was shown to be very productive in finding distantly related sequences and more powerful than Hidden Markov Models or the profile methods in WiseTools and PSI-BLAST in many cases. The profile extraction routine is available on the WWW (http://www.bork.embl-heidelberg. de/PSIC or http://www.imb.ac.ru/PSIC).

Predicting function from sequence using computational tools is a highly complicated procedure that is generally done for each gene individually. This review focuses on the added value that is provided by completely sequenced genomes in function prediction. Various levels of sequence annotation and function prediction are discussed, ranging from genomic sequence to that of complex cellular processes. Protein function is currently best described in the context of molecular interactions. In the near future it will be possible to predict protein function in the context of higher order processes such as the regulation of gene expression, metabolic pathways and signalling cascades. The analysis of such higher levels of function description uses, besides the information from completely sequenced genomes, also the additional information from proteomics and expression data. The final goal will be to elucidate the mapping between genotype and phenotype.

Glycosylphosphatidylinositol (GPI) anchoring is a common post-translational modification of extracellular eukaryotic proteins. Attachment of the GPI moiety to the carboxyl terminus (omega-site) of the polypeptide occurs after proteolytic cleavage of a C-terminal propeptide. In this work, the sequence pattern for GPI-modification was analyzed in terms of physical amino acid properties based on a database analysis of annotated proprotein sequences. In addition to a refinement of previously described sequence signals, we report conserved sequence properties in the regions omega - 11...omega - 1 and omega + 4...omega + 5. We present statistical evidence for volume-compensating residue exchanges with respect to the positions omega - 1...omega + 2. Differences between protozoan and metazoan GPI-modification motifs consist mainly in variations of preferences to amino acid types at the positions near the omega-site and in the overall motif length. The variations of polypeptide substrates are exploited to suggest a model of the polypeptide binding site of the putative transamidase, the enzyme catalyzing the GPI-modification. The volume of the active site cleft accommodating the four residues omega - 1...omega + 2 appears to be approximately 540 A3.

Homology search techniques based on the iterative PSI-BLAST method in combination with various filters for low sequence complexity are applied to assign folds to all Mycoplasma genitalium proteins. The resulting procedure (implemented as a web server) is able to predict at least one domain in 37% of these proteins automatically, with an estimated accuracy higher than 98%. Taking structural features such as coiled coil or transmembrane regions aside, folds can be assigned to more than half of the globular proteins in a bacterium just by iterative sequence comparison.

The parametric description of residue environments through solvent accessibility, backbone conformation, or pairwise residue-residue distances is the key to the comparison between amino acid types at protein sequence positions and residue locations in structural templates (condition of protein sequence-structure match). For the first time, the research results presented in this study clarify and allow to quantify, on a rigorous statistical basis, to what extent the amino acid type-specific distributions of commonly used environment parameters are discriminative with respect to the 20 amino acid types. Relying on the Bahadur theory, we estimate the probability of error in a single-sequence-structure alignment based on weak or absent discriminative power in a learning database of protein structure. We present the results for many residue environment variables and demonstrate that each fold description parameter is sensitive with respect to only a few amino acid types while indifferent to most of the other amino acid types. Even complex structural characteristics combining solvent-accessible surface area, backbone conformation, and pairwise distances distinguish only some amino acid types, whereas the others remain nondiscriminated. We find that the knowledge-based potentials currently in use treat especially Ala, Asp, Gln, His, Ser, Thr, and Tyr as essentially "average" amino acids. Thus, highly discriminative amino acid types define the alignment register in gapless sequence-structure alignments. The introduction of gaps leads to alignment ambiguities at sequence positions occupied by nondiscriminated amino acid types. Therefore, local sequence-structure alignments produced by techniques with gaps cannot be reliable. Conceptionally new and more sensitive environment parameters must be invented.

For the first time, a direct approach for the derivation of an atomic solvation parameter from macromolecular structural data alone is presented. The specific free energy of solvation for hydrophobic surface regions of proteins is delineated from the area distribution of hydrophobic surface patches. The resulting value is 18 cal/(mol.A2), with a statistical uncertainty of +/-2 cal/mol.A2) at the 5% significance level. It compares favorably with the parameters for carbon obtained by other authors who use the the crystal geometry of succinic acid or energies of transfer from hydrophobic solvent to water for small organic compounds. Thus, the transferability of atomic solvation parameters for hydrophobic atoms to macromolecules has been directly demonstrated. A careful statistical analysis demonstrates that surface energy parameters derived from thermodynamic data of protein mutation experiments are clearly less confident.

The hydrophobic part of the solvent-accessible surface of a typical monomeric globular protein consists of a single, large interconnected region formed from faces of apolar atoms and constituting approximately 60% of the solvent-accessible surface area. Therefore, the direct delineation of the hydrophobic surface patches on an atom-wise basis is impossible. Experimental data indicate that, in a two-state hydration model, a protein can be considered to be unified with its first hydration shell in its interaction with bulk water. We show that, if the surface area occupied by water molecules bound at polar protein atoms as generated by AUTOSOL is removed, only about two-thirds of the hydrophobic part of the protein surface remains accessible to bulk solvent. Moreover, the organization of the hydrophobic part of the solvent-accessible surface experiences a drastic change, such that the single interconnected hydrophobic region disintegrates into many smaller patches, i.e. the physical definition of a hydrophobic surface region as unoccupied by first hydration shell water molecules can distinguish between hydrophobic surface clusters and small interconnecting channels. It is these remaining hydrophobic surface pieces that probably play an important role in intra- and intermolecular recognition processes such as ligand binding, protein folding and protein-protein association in solution conditions. These observations have led to the development of an accurate and quick analytical technique for the automatic determination of hydrophobic surface patches of proteins. This technique is not aggravated by the limiting assumptions of the methods for generating explicit water hydration positions. Formation of the hydrophobic surface regions owing to the structure of the first hydration shell can be computationally simulated by a small radial increment in solvent-accessible polar atoms, followed by calculation of the remaining exposed hydrophobic patches. We demonstrate that a radial increase of 0.35-0.50 A resembles the effect of tightly bound water on the organization of the hydrophobic part of the solvent-accessible surface.

The success rates reported for secondary structural class prediction with different methods are contradictory. On one side, the problem of recognizing the secondary structural class of a protein knowing only its amino acid composition appears completely solved by simply applying jury decision with an elliptically scaled distance function. Chou and coworkers repeatedly (see Crit. Rev. Biochem. Mol. Biol. 30:275-349, 1995) published prediction accuracies near 100%. On the other hand, traditional secondary structure prediction techniques achieve success rates of about 70% for the secondary structural state per residue and about 75% for structural class only with extensive input information (full sequence of the query protein, its amino acid composition and length, multiple alignments with homologous sequences). In this article, we resolve the paradox and consider (1) the question of the secondary structural class definition, (2) the role of the representativity of the test set of protein tertiary structure for the current state of the Protein Data Bank (PDB); and (3) we estimate the real impact of amino acid composition on secondary structural class. We formulate three objective criteria for a reasonable definition of secondary structural classes and show that only the criterion of Nakashima et al. (J. Biochem. 99:153-162, 1986) complies with all of them. Only this definition matches the distribution of secondary structural content in representative PDB subsets, whereas other criteria leave many proteins (up to 65% of all PDB entries) simply unassigned. We review critically specialized secondary-structural class prediction methods, especially those of Chou and coworkers, which claim almost 100% accuracy using only amino acid composition, and resolve the paradox that these prediction accuracies are better than those from secondary structure predictions from multiple alignments. We show (i) that these techniques rely on a preselection of test sets which removes irregular proteins and other proteins without any class assignment (about 35% of all PDB entries); and (ii) that even for preselected representative test sets, the success rate drops to 60% and lower for a 4-type classification (alpha, beta, alpha + beta, alpha/beta). The prediction accuracies fall to about 50% if the secondary structural class definition of Nakashima et al. is applied and only few irregular proteins are preselected and removed from automatically generated, representative subsets of the PDB. We have applied two new vector decomposition methods for secondary structural content prediction from amino acid composition alone, with and without consideration of amino acid compositional coupling in the learning set of tertiary structures respectively, to the problem of class prediction and achieve about 60% correct assignment among four classes (alpha, beta, mixed, irregular) as well as single sequence-based secondary structure prediction methods like GORIII and COMBI. Our results demonstrate that 60% correctness is the upper limit for a 4-type class prediction from amino acid composition alone for an unknown query protein and that consideration of compositional coupling does not improve the prediction success. The prediction program SSCP offering secondary structural class assignment for query compositions and sequences has been made available as a World Wide Web and E-mail service.

The predictive limits of the amino acid composition for the secondary structural content (percentage of residues in the secondary structural states helix, sheet, and coil) in proteins are assessed quantitatively. For the first time, techniques for prediction of secondary structural content are presented which rely on the amino acid composition as the only information on the query protein. In our first method, the amino acid composition of an unknown protein is represented by the best (in a least square sense) linear combination of the characteristic amino acid compositions of the three secondary structural types computed from a learning set of tertiary structures. The second technique is a generalization of the first one and takes into account also possible compositional couplings between any two sorts of amino acids. Its mathematical formulation results in an eigenvalue/eigenvector problem of the second moment matrix describing the amino acid compositional fluctuations of secondary structural types in various proteins of a learning set. Possible correlations of the principal directions of the eigenspaces with physical properties of the amino acids were also checked. For example, the first two eigenvectors of the helical eigenspace correlate with the size and hydrophobicity of the residue types respectively. As learning and test sets of tertiary structures, we utilized representative, automatically generated subsets of Protein Data Bank (PDB) consisting of non-homologous protein structures at the resolution thresholds < or = 1.8A, < or = 2.0A, < or = 2.5A, and < or = 3.0 A. We show that the consideration of compositional couplings improves prediction accuracy, albeit not dramatically. Whereas in the self-consistency test (learning with the protein to be predicted), a clear decrease of prediction accuracy with worsening resolution is observed, the jackknife test (leave the predicted protein out) yielded best results for the largest dataset (< or = 3.0A, almost no difference to the self-consistency test!), i.e., only this set, with more than 400 proteins, is sufficient for stable computation of the parameters in the prediction function of the second method. The average absolute error in predicting the fraction of helix, sheet, and coil from amino acid composition of the query protein are 13.7, 12.6, and 11.4%, respectively with r.m.s. deviations in the range of 8.6 divided by 11.8% for the 3.0 A dataset in a jackknife test. The absolute precision of the average absolute errors is in the range of 1 divided by 3% as measured for other representative subsets of the PDB. Secondary structural content prediction methods found in the literature have been clustered in accordance with their prediction accuracies. To our surprise, much more complex secondary structure prediction methods utilized for the same purpose of secondary structural content prediction achieve prediction accuracies very similar to those of the present analytic techniques, implying that all the information beyond the amino acid composition is, in fact, mainly utilized for positioning the secondary structural state in the sequence but not for determination of the overall number of residues in a secondary structural type. This result implies that higher prediction accuracies cannot be achieved relying solely on the amino acid composition of an unknown query protein as prediction input. Our prediction program SSCP has been made available as a World Wide Web and E-mail service.

The geometry of helix-helix packing in globular proteins is comprehensively analysed within the model of the superposition of two helix lattices which result from unrolling the helix cylinders onto a plane containing points representing each residue. The requirements for the helix geometry (the radius R, the twist angle omega and the rise per residue delta) under perfect match of the lattices are studied through a consistent mathematical model that allows consideration of all possible associations of all helix types (alpha-, pi- and 3(10)). The corresponding equations have three well-separated solutions for the interhelical packing angle, omega, as a function of the helix geometric parameters allowing optimal packing. The resulting functional relations also show unexpected behaviour. For a typically observed alpha-helix (omega = 99.1 degrees, delta = 1.45 A), the three optimal packing angles are omega a,b,c = -37.1 degrees, -97.4 degrees and +22.0 degrees with a periodicity of 180 degrees and respective helix radii Ra,b,c = 3.0 A, 3.5 A and 4.3 A. However, the resulting radii are very sensitive to variations in the twist angle omega. At omega triple = 96.9 degrees, all three solutions yield identical radii at delta = 1.45 A where Rtriple = 3.46 A. This radius is close to that of a poly(Ala) helix, indicating a great packing flexibility when alanine is involved in the packing core, and omega triple is close to the mean observed twist angle. In contrast, the variety of possible theoretical solutions is limited for the other two helix types. Besides the perfect matches, novel suboptimal "knobs into holes" hydrophobic packing patterns as a function of the helix radius are described. Alternative "knobs onto knobs" and mixed models can be applied in cases where salt bridges, hydrogen bonds, disulphide bonds and tight hydrophobic head-to-head contacts are involved in helix-helix associations. An analysis of the experimentally observed packings in proteins confirmed the conclusions of the theoretical model. Nonetheless, the observed alpha-helix packings showed deviations from the 180 degrees periodicity expected from the model. An investigation of the actual three-dimensional geometry of helix-helix packing revealed an explanation for the observed discrepancies where a decisive role was assigned to the defined orientation of the C alpha-C beta vectors of the side-chains. As predicted form the model, helices with different radii (differently sized side-chains in the packing core) were observed to utilize different packing cells (packing patterns). In agreement with the coincidence between Rtriple and the radius of a poly(Ala) helix, Ala was observed to show greatest propensity to build the packing core. The application of the helix lattice superposition model suggests that the packing of amino acid residues is best described by a "knobs into holes" scheme rather than "ridges into grooves". The various specific packing modes made salient by the model should be useful in protein engineering and design.