MHC Class II Binding Prediction by Molecular Docking.

Proteins of the Major Histocompatibility Complex (MHC) bind self and nonself peptide antigens or epitopes within the cell and present them at the cell surface for recognition by T cells. All T-cell epitopes are MHC binders but not all MCH binders are T-cell epitopes. The MHC class II proteins are extremely polymorphic. Polymorphic residues cluster in the peptide-binding region and largely determine the MHC's peptide selectivity. The peptide binding site on MHC class II proteins consist of five binding pockets. Using molecular docking, we have modelled the interactions between peptide and MHC class II proteins from locus DRB1. A combinatorial peptide library was generated by mutation of residues at peptide positions which correspond to binding pockets (so called anchor positions). The binding affinities were assessed using different scoring functions. The normalized scoring functions for each amino acid at each anchor position were used to construct quantitative matrices (QM) for MHC class II binding prediction. Models were validated by external test sets comprising 4540 known binders. Eighty percent of the known binders are identified in the best predicted 15 % of all overlapping peptides, originating from one protein.

Hierarchical classification of G-protein-coupled receptors with data-driven selection of attributes and classifiers.

We address the important bioinformatics problem of predicting protein function from a protein's primary sequence. We consider the functional classification of G-Protein-Coupled Receptors (GPCRs), whose functions are specified in a class hierarchy. We tackle this task using a novel top-down hierarchical classification system where, for each node in the class hierarchy, the predictor attributes to be used in that node and the classifier to be applied to the selected attributes are chosen in a data-driven manner. Compared with a previous hierarchical classification system selecting classifiers only, our new system significantly reduced processing time without significantly sacrificing predictive accuracy.

Towards the in silico identification of class II restricted T-cell epitopes: a partial least squares iterative self-consistent algorithm for affinity prediction.

MOTIVATION: The immunogenicity of peptides depends on their ability to bind to MHC molecules. MHC binding affinity prediction methods can save significant amounts of experimental work. The class II MHC binding site is open at both ends, making epitope prediction difficult because of the multiple binding ability of long peptides. RESULTS: An iterative self-consistent partial least squares (PLS)-based additive method was applied to a set of 66 peptides no longer than 16 amino acids, binding to DRB1*0401. A regression equation containing the quantitative contributions of the amino acids at each of the nine positions was generated. Its predictability was tested using two external test sets which gave r(pred) = 0.593 and r(pred) = 0.655, respectively. Furthermore, it was benchmarked using 25 known T-cell epitopes restricted by DRB1*0401 and we compared our results with four other online predictive methods. The additive method showed the best result finding 24 of the 25 T-cell epitopes. AVAILABILITY: Peptides used in the study are available from http://www.jenner.ac.uk/JenPep. The PLS method is available commercially in the SYBYL molecular modelling software package. The final model for affinity prediction of peptides binding to DRB1*0401 molecule is available at http://www.jenner.ac.uk/MHCPred. Models developed for DRB1*0101 and DRB1*0701 also are available in MHCPred.

PRINTS and its automatic supplement, prePRINTS.

The PRINTS database houses a collection of protein fingerprints. These may be used to assign uncharacterised sequences to known families and hence to infer tentative functions. The September 2002 release (version 36.0) includes 1800 fingerprints, encoding approximately 11 000 motifs, covering a range of globular and membrane proteins, modular polypeptides and so on. In addition to its continued steady growth, we report here the development of an automatic supplement, prePRINTS, designed to increase the coverage of the resource and reduce some of the manual burdens inherent in its maintenance. The databases are accessible for interrogation and searching at http://www.bioinf.man.ac.uk/dbbrowser/PRINTS/.

PRINTS and PRINTS-S shed light on protein ancestry.

The PRINTS database houses a collection of protein fingerprints. These may be used to make family and tentative functional assignments for uncharacterised sequences. The September 2001 release (version 32.0) includes 1600 fingerprints, encoding approximately 10 000 motifs, covering a range of globular and membrane proteins, modular polypeptides and so on. In addition to its continued steady growth, we report here its use as a source of annotation in the InterPro resource, and the use of its relational cousin, PRINTS-S, to model relationships between families, including those beyond the reach of conventional sequence analysis approaches. The database is accessible for BLAST, fingerprint and text searches at http://www.bioinf.man.ac.uk/dbbrowser/PRINTS/.

Toward the quantitative prediction of T-cell epitopes: coMFA and coMSIA studies of peptides with affinity for the class I MHC molecule HLA-A*0201.

A set of 102 peptides with affinity for the class I MHC HLA-A0201 molecule was subjected to three-dimensional quantitative structure-affinity relationship (3D QSAR) studies using comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA). A test set of 50 peptides was used to determine the predictive value of the models. The CoMFA models gave q(2) and r(2)pred below 0.5. The best CoMSIA model has q(2) = 0.542 and r(2)pred = 0.679, and includes hydrophobic, steric, and H-bond donor fields. The hydrophobic interactions play a dominant role in peptide-MHC molecule binding. CoMSIA coefficient contour maps were used to analyze the structural features of the peptides accounting for the affinity in terms of the three positively contributing physicochemical properties: local hydrophobicity, steric bulk and hydrogen-bond-donor ability.

A deletion in the gene encoding the CD45 antigen in a patient with SCID.

SCID is a heterogeneous group of hereditary diseases. Mutations in the common gamma-chain (gamma(c)) of cytokine receptors, including those for IL-2, IL-4, IL-7, IL-9, and IL-15, are responsible for an X-linked form of the disease, while mutations of several other genes, including Janus-associated kinase-3, may cause autosomal recessive forms of SCID. We investigated the first SCID patient to be described with minimal cell surface expression of the leukocyte common (CD45) Ag. CD45 is an abundant transmembrane tyrosine phosphatase, expressed on all leukocytes, and is required for efficient lymphocyte signaling. CD45-deficient mice are severely immunodeficient and have very few peripheral T lymphocytes. We report here that a homozygous 6-bp deletion in the gene encoding CD45 (PTPRC, gene map locus 1q31-32), which results in a loss of glutamic acid 339 and tyrosine 340 in the first fibronectin type III module of the extracellular domain of CD45, is associated with failure of surface expression of CD45 and SCID. Molecular modeling suggests that tyrosine 340 is crucial for the structural integrity of CD45 protein. This is the second description of a clinically relevant CD45 mutation, provides direct evidence for the importance of CD45 in immune function in humans, and suggests that abnormalities in CD45 expression are a possible cause of SCID in humans.

The lipocalin protein family: structural and sequence overview.

Lipocalins are remarkably diverse at the sequence level yet have highly conserved structures. Most lipocalins share three characteristic conserved sequence motifs - the kernel lipocalins - while others are more divergent family members - the outlier lipocalins - typically sharing only one or two. This classification is a useful tool for analysing the family, and within these large sets are smaller groups sharing much higher levels of sequence similarity. The lipocalins are also part of a larger protein superfamily: the calycins, which includes the fatty acid binding proteins, avidins, a group of metalloproteinase inhibitors, and triabin. The superfamily is characterised by a similar structure (a repeated +1 topology beta-barrel) and by the conservation of a remarkable structural signature.

Experimentally determined lipocalin structures.

Twelve structures of distinct members of the lipocalin protein family have been solved experimentally. These structures have revolutionised our understanding of the properties of the lipocalins. Many more members of the family have been crystallised and now await structure solution. The number of solved lipocalin structures is steadily increasing, and with it increases our knowledge of this enigmatic and challenging protein family.

Beyond the superfamily: the lipocalin receptors.

Lipocalins are characterized by multiple molecular recognition properties including the ability to bind to cell surface receptors. Receptors for a number of lipocalins have been identified. These include receptors for alpha-1-microglobulin, insecticyanin, glycodelin, retinol-binding protein, alpha-1-acid glycoprotein, beta-lactoglobulin and odorant-binding protein. The properties of these receptors are summarized and discussed.

The lipocalin website.

We introduce a website devoted to the lipocalins. The website contains data on lipocalin structures and sequences, as well as reviewing lipocalin biology and biochemistry. Our hope is that it can act as a focus for future research into the lipocalin protein family. The website can be accessed at the following URL: http://www. jenner.ac.uk/lipocalin.htm.

PRINTS-S: the database formerly known as PRINTS.

The PRINTS database houses a collection of protein family fingerprints. These are groups of motifs that together are diagnostically more potent than single motifs by virtue of the biological context afforded by matching motif neighbours. Around 1200 fingerprints have now been created and stored in the database. The September 1999 release (version 24.0) encodes approximately 7200 motifs, covering a range of globular and membrane proteins, modular polypeptides and so on. In addition to its continued steady growth, we report here several major changes to the resource, including the design of an automated strategy for database maintenance, and implementation of an object-relational schema for more efficient data management. The database is accessible for BLAST, fingerprint and text searches at http://www.bioinf.man.ac. uk/dbbrowser/PRINTS/

Modelling G-protein-coupled receptors for drug design.

The G-protein coupled receptors form a large and diverse multi-gene superfamily with many important physiological functions. As such, they have become important targets in pharmaceutical research. Molecular modelling and site-directed mutagenesis have played an important role in our increasing understanding of the structural basis of drug action at these receptors. Aspects of this understanding, how these techniques can be used within a drug-design programme, and remaining challenges for the future are reviewed.

Lead generation using pharmacophore mapping and three-dimensional database searching: application to muscarinic M(3) receptor antagonists.

By using a pharmacophore model, a geometrical representation of the features necessary for molecules to show a particular biological activity, it is possible to search databases containing the 3D structures of molecules and identify novel compounds which may possess this activity. We describe our experiences of establishing a working 3D database system and its use in rational drug design. By using muscarinic M(3) receptor antagonists as an example, we show that it is possible to identify potent novel lead compounds using this approach. Pharmacophore generation based on the structures of known M(3) receptor antagonists, 3D database searching, and medium-throughput screening were used to identify candidate compounds. Three compounds were chosen to define the pharmacophore: a lung-selective M(3) antagonist patented by Pfizer and two Astra compounds which show affinity at the M(3) receptor. From these, a pharmacophore model was generated, using the program DISCO, and this was used subsequently to search a UNITY 3D database of proprietary compounds; 172 compounds were found to fit the pharmacophore. These compounds were then screened, and 1-[2-(2-(diethylamino)ethoxy)phenyl]-2-phenylethanone (pA(2) 6.67) was identified as the best hit, with N-[2-(piperidin-1-ylmethyl)cycohexyl]-2-propoxybenz amide (pA(2) 4. 83) and phenylcarbamic acid 2-(morpholin-4-ylmethyl)cyclohexyl ester (pA(2) 5.54) demonstrating lower activity. As well as its potency, 1-[2-(2-(diethylamino)ethoxy)phenyl]-2-phenylethanone is a simple structure with limited similarity to existing M(3) receptor antagonists.

PRINTS prepares for the new millennium.

PRINTS is a diagnostic collection of protein fingerprints. Fingerprints exploit groups of motifs to build characteristic family signatures, offering improved diagnostic reliability over single-motif approaches by virtue of the mutual context provided by motif neighbours. Around 1000 fingerprints have now been created and stored in PRINTS. The September 1998 release (version 20.0), encodes approximately 5700 motifs, covering a range of globular and membrane proteins, modular polypeptides and so on. The database is accessible via the DbBrowser Web Server at http://www.biochem.ucl.ac.uk/bsm/dbbrowser /. In addition to supporting its continued growth, recent enhancements to the resource include a BLAST server, and more efficient fingerprint search software, with improved statistics for estimating the reliability of retrieved matches. Current efforts are focused on the design of more automated methods for database maintenance; implementation of an object-relational schema for efficient data management; and integration with PROSITE, profiles, Pfam and ProDom, as part of the international InterPro project, which aims to unify protein pattern databases and offer improved tools for genome analysis.

Rotational superposition: a review of methods.

Rotational superposition is one of the most commonly used algorithms in molecular modelling. Many different methods of solving superposition have been suggested. Of these, methods based on the quaternion parameterization of rotation are fast, accurate, and robust. Quaternion parameterization-based methods cannot result in rotation inversion and do not have special cases such as co-linearity or co-planarity of points. Thus, quaternion parameterization-based methods are the best choice for rotational superposition applications.

FingerPRINTScan: intelligent searching of the PRINTS motif database.

MOTIVATION: By identifying an unknown gene or protein as a member of a known family, we can infer a wealth of previously compiled information pertinent to that family and its members. RESULTS: This paper introduces a method that classifies sequences using familial definitions from the PRINTS database, allowing progress to be made with the identification of distant evolutionary relationships. The approach makes use of the contextual information inherent in a multiple-motif method, and has the power to identify hitherto unidentified relationships in mass genome data. We exemplify our method by a comparison of database searches with uncharacterized sequences from the Caenorhabditis elegans and Saccharomyces cerevisiae genome projects. This analysis tool combines a simple, user-friendly interface with the capacity to provide an 'intelligent', biologically relevant result.

The PRINTS protein fingerprint database in its fifth year.

PRINTS is a database of protein family 'fingerprints' offering a diagnostic resource for newly-determined sequences. By contrast with PROSITE, which uses single consensus expressions to characterise particular families, PRINTS exploits groups of motifs to build characteristic signatures. These signatures offer improved diagnostic reliability by virtue of the mutual context provided by motif neighbours. To date, 800 fingerprints have been constructed and stored in PRINTS. The current version, 17.0, encodes approximately 4500 motifs, covering a range of globular and membrane proteins, modular polypeptides, and so on. The database is accessible via the UCL Bioinformatics World Wide Web (WWW) Server at http://www. biochem.ucl.ac.uk/bsm/dbbrowser/ . We have recently enhanced the usefulness of PRINTS by making available new, intuitive search software. This allows both individual query sequence and bulk data submission, permitting easy analysis of single sequences or complete genomes. Preliminary results indicate that use of the PRINTS system is able to assign additional functions not found by other methods, and hence offers a useful adjunct to current genome analysis protocols.