MOTIVATION: The availability of very large numbers of protein structures from accurate computational methods poses new challenges in storing, searching and detecting relationships between these structures. In particular, the new-found abundance of multi-domain structures in the AlphaFold structure database introduces challenges for traditional structure comparison methods. RESULTS: We address these challenges using a fast, embedding-based structure comparison method called Foldclass which detects structural similarity between protein domains. We demonstrate the accuracy of Foldclass embeddings for homology detection. In combination with a recently developed deep learning-based automatic domain segmentation tool Merizo, we develop Merizo-search, which first segments multi-domain query structures into domains, and then searches a Foldclass embedding database to determine the top matches for each constituent domain. Combining the ability of Merizo to accurately segment complete chains into domains, and Foldclass to embed and detect similar domains, the Merizo-search tool can be used to rapidly detect per-domain similarities for complete chains, taking as little as 2 min to search all 365 million domains from the Encyclopedia of Domains. We anticipate that these tools will enable many analyses using the wealth of predicted structural data now available. AVAILABILITY AND IMPLEMENTATION: Foldclass and Merizo-search are available at https://github.com/psipred/merizo_search. The version used in this publication is archived at https://doi.org/10.5281/zenodo.15120830. Merizo-search is also available on the PSIPRED web server at http://bioinf.cs.ucl.ac.uk/psipred.

We propose a Perceiver-based sequence classifier to detect abnormalities in speech reflective of several neurological disorders. We combine this classifier with a Universal Speech Model (USM) that is trained on 12 million hours of diverse audio recordings. Our model compresses long sequences into a small set of class-specific latent representations and a factorized projection is used to predict different attributes of the disordered input speech. The benefit of our approach is that it allows us to model different regions of the input for different classes and is at the same time data efficient. We evaluated the proposed model extensively on a curated corpus from the Mayo Clinic. Our model outperforms standard transformer (80.9%) and perceiver (81.8%) models and achieves an average accuracy of 83.1%. With limited task-specific data, we find that pretraining is important and surprisingly pretraining with the un-related automatic speech recognition (ASR) task is also beneficial. Encodings from the middle layers provide a mix of both acoustic and phonetic information and achieve best prediction results compared to just using the final layer encodings (83.1% vs 79.6%). The results are promising and with further refinements may help clinicians detect speech abnormalities without needing access to highly specialized speech-language pathologists.

Pathological speech analysis with Automatic Speech Recognition (ASR) is a long-standing research domain based on the expectation that there is a close relationship between pathological speech abnormalities and reduced ASR performance. This has led to interest in ASR-based clinical analyses such as intelligibility assessments and disfluency detection. However, these analyses are premised on transcriptions that accurately reflect abnormalities. We find that the choices of ASR model and speech task significantly affect transcription errors for both typical and pathological speech, which is reflected in the word error rate (WER). Notably, language restriction (e.g., dictionary of words) decreases a model’s ability to capture abnormalities while more complex tasks increase both model and speaker-driven transcription errors. These findings highlight possible variability in clinical analysis with ASR and suggest the importance of controlling for model type and task complexity.

We model Alzheimer’s disease (AD) progression by combining differential equations (DEs) and reinforcement learning (RL) with domain knowledge. DEs provide relationships between some, but not all, factors relevant to AD. We assume that the missing relationships must satisfy general criteria about the working of the brain, for e.g., maximizing cognition while minimizing the cost of supporting cognition. This allows us to extract the missing relationships by using RL to optimize an objective (reward) function that captures the above criteria. We use our model consisting of DEs (as a simulator) and the trained RL agent to predict individualized 10-year AD progression using baseline (year 0) features on synthetic and real data. The model was comparable or better at predicting 10-year cognition trajectories than state-of-the-art learning-based models. Our interpretable model demonstrated, and provided insights into, "recovery/compensatory" processes that mitigate the effect of AD, even though those processes were not explicitly encoded in the model. Our framework combines DEs with RL for modelling AD progression and has broad applicability for understanding other neurological disorders.

MOTIVATION: Over the past 50 years, our ability to model protein sequences with evolutionary information has progressed in leaps and bounds. However, even with the latest deep learning methods, the modelling of a critically important class of proteins, single orphan sequences, remains unsolved. RESULTS: By taking a bioinformatics approach to semi-supervised machine learning, we develop Profile Augmentation of Single Sequences (PASS), a simple but powerful framework for building accurate single-sequence methods. To demonstrate the effectiveness of PASS we apply it to the mature field of secondary structure prediction. In doing so we develop S4PRED, the successor to the open-source PSIPRED-Single method, which achieves an unprecedented Q3 score of 75.3% on the standard CB513 test. PASS provides a blueprint for the development of a new generation of predictive methods, advancing our ability to model individual protein sequences. AVAILABILITY AND IMPLEMENTATION: The S4PRED model is available as open source software on the PSIPRED GitHub repository (https://github.com/psipred/s4pred), along with documentation. It will also be provided as a part of the PSIPRED web service (http://bioinf.cs.ucl.ac.uk/psipred/). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Motivation: In addition to substitution frequency data from protein sequence alignments, many state-of-the-art methods for contact prediction rely on additional sources of information, or features, of protein sequences in order to predict residue-residue contacts, such as solvent accessibility, predicted secondary structure, and scores from other contact prediction methods. It is unclear how much of this information is needed to achieve state-of-the-art results. Here, we show that using deep neural network models, simple alignment statistics contain sufficient information to achieve state-of-the-art precision. Our prediction method, DeepCov, uses fully convolutional neural networks operating on amino-acid pair frequency or covariance data derived directly from sequence alignments, without using global statistical methods such as sparse inverse covariance or pseudolikelihood estimation. Results: Comparisons against CCMpred and MetaPSICOV2 show that using pairwise covariance data calculated from raw alignments as input allows us to match or exceed the performance of both of these methods. Almost all of the achieved precision is obtained when considering relatively local windows (around 15 residues) around any member of a given residue pairing; larger window sizes have comparable performance. Assessment on a set of shallow sequence alignments (fewer than 160 effective sequences) indicates that the new method is substantially more precise than CCMpred and MetaPSICOV2 in this regime, suggesting that improved precision is attainable on smaller sequence families. Overall, the performance of DeepCov is competitive with the state of the art, and our results demonstrate that global models, which employ features from all parts of the input alignment when predicting individual contacts, are not strictly needed in order to attain precise contact predictions. Availability and implementation: DeepCov is freely available at https://github.com/psipred/DeepCov. Supplementary information: Supplementary data are available at Bioinformatics online.

MOTIVATION: Protein fold recognition when appropriate, evolutionarily-related, structural templates can be identified is often trivial and may even be viewed as a solved problem. However in cases where no homologous structural templates can be detected, fold recognition is a notoriously difficult problem ( Moult et al., 2014 ). Here we present EigenTHREADER, a novel fold recognition method capable of identifying folds where no homologous structures can be identified. EigenTHREADER takes a query amino acid sequence, generates a map of intra-residue contacts, and then searches a library of contact maps of known structures. To allow the contact maps to be compared, we use eigenvector decomposition to resolve the principal eigenvectors these can then be aligned using standard dynamic programming algorithms. The approach is similar to the Al-Eigen approach of Di Lena et al. (2010) , but with improvements made both to speed and accuracy. With this search strategy, EigenTHREADER does not depend directly on sequence homology between the target protein and entries in the fold library to generate models. This in turn enables EigenTHREADER to correctly identify analogous folds where little or no sequence homology information is. RESULTS: EigenTHREADER outperforms well-established fold recognition methods such as pGenTHREADER and HHSearch in terms of True Positive Rate in the difficult task of analogous fold recognition. This should allow template-based modelling to be extended to many new protein families that were previously intractable to homology based fold recognition methods. AVAILABILITY AND IMPLEMENTATION: All code used to generate these results and the computational protocol can be downloaded from https://github.com/DanBuchan/eigen_scripts . EigenTHREADER, the benchmark code and the data this paper is based on can be downloaded from: http://bioinfadmin.cs.ucl.ac.uk/downloads/eigenTHREADER/ . CONTACT: [email protected].

Accurate gene or protein function prediction is a key challenge in the post-genome era. Most current methods perform well on molecular function prediction, but struggle to provide useful annotations relating to biological process functions due to the limited power of sequence-based features in that functional domain. In this work, we systematically evaluate the predictive power of temporal transcription expression profiles for protein function prediction in Drosophila melanogaster. Our results show significantly better performance on predicting protein function when transcription expression profile-based features are integrated with sequence-derived features, compared with the sequence-derived features alone. We also observe that the combination of expression-based and sequence-based features leads to further improvement of accuracy on predicting all three domains of gene function. Based on the optimal feature combinations, we then propose a novel multi-classifier-based function prediction method for Drosophila melanogaster proteins, FFPred-fly+. Interpreting our machine learning models also allows us to identify some of the underlying links between biological processes and developmental stages of Drosophila melanogaster.

MOTIVATION: A sizeable fraction of eukaryotic proteins contain intrinsically disordered regions (IDRs), which act in unfolded states or by undergoing transitions between structured and unstructured conformations. Over time, sequence-based classifiers of IDRs have become fairly accurate and currently a major challenge is linking IDRs to their biological roles from the molecular to the systems level. RESULTS: We describe DISOPRED3, which extends its predecessor with new modules to predict IDRs and protein-binding sites within them. Based on recent CASP evaluation results, DISOPRED3 can be regarded as state of the art in the identification of IDRs, and our self-assessment shows that it significantly improves over DISOPRED2 because its predictions are more specific across the whole board and more sensitive to IDRs longer than 20 amino acids. Predicted IDRs are annotated as protein binding through a novel SVM based classifier, which uses profile data and additional sequence-derived features. Based on benchmarking experiments with full cross-validation, we show that this predictor generates precise assignments of disordered protein binding regions and that it compares well with other publicly available tools.

MOTIVATION: Recent developments of statistical techniques to infer direct evolutionary couplings between residue pairs have rendered covariation-based contact prediction a viable means for accurate 3D modelling of proteins, with no information other than the sequence required. To extend the usefulness of contact prediction, we have designed a new meta-predictor (MetaPSICOV) which combines three distinct approaches for inferring covariation signals from multiple sequence alignments, considers a broad range of other sequence-derived features and, uniquely, a range of metrics which describe both the local and global quality of the input multiple sequence alignment. Finally, we use a two-stage predictor, where the second stage filters the output of the first stage. This two-stage predictor is additionally evaluated on its ability to accurately predict the long range network of hydrogen bonds, including correctly assigning the donor and acceptor residues. RESULTS: Using the original PSICOV benchmark set of 150 protein families, MetaPSICOV achieves a mean precision of 0.54 for top-L predicted long range contacts-around 60% higher than PSICOV, and around 40% better than CCMpred. In de novo protein structure prediction using FRAGFOLD, MetaPSICOV is able to improve the TM-scores of models by a median of 0.05 compared with PSICOV. Lastly, for predicting long range hydrogen bonding, MetaPSICOV-HB achieves a precision of 0.69 for the top-L/10 hydrogen bonds compared with just 0.26 for the baseline MetaPSICOV. AVAILABILITY AND IMPLEMENTATION: MetaPSICOV is available as a freely available web server at http://bioinf.cs.ucl.ac.uk/MetaPSICOV. Raw data (predicted contact lists and 3D models) and source code can be downloaded from http://bioinf.cs.ucl.ac.uk/downloads/MetaPSICOV. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Accurate protein function annotation is a severe bottleneck when utilizing the deluge of high-throughput, next generation sequencing data. Keeping database annotations up-to-date has become a major scientific challenge that requires the development of reliable automatic predictors of protein function. The CAFA experiment provided a unique opportunity to undertake comprehensive 'blind testing' of many diverse approaches for automated function prediction. We report on the methodology we used for this challenge and on the lessons we learnt. METHODS: Our method integrates into a single framework a wide variety of biological information sources, encompassing sequence, gene expression and protein-protein interaction data, as well as annotations in UniProt entries. The methodology transfers functional categories based on the results from complementary homology-based and feature-based analyses. We generated the final molecular function and biological process assignments by combining the initial predictions in a probabilistic manner, which takes into account the Gene Ontology hierarchical structure. RESULTS: We propose a novel scoring function called COmbined Graph-Information Content similarity (COGIC) score for the comparison of predicted functional categories and benchmark data. We demonstrate that our integrative approach provides increased scope and accuracy over both the component methods and the naïve predictors. In line with previous studies, we find that molecular function predictions are more accurate than biological process assignments. CONCLUSIONS: Overall, the results indicate that there is considerable room for improvement in the field. It still remains for the community to invest a great deal of effort to make automated function prediction a useful and routine component in the toolbox of life scientists. As already witnessed in other areas, community-wide blind testing experiments will be pivotal in establishing standards for the evaluation of prediction accuracy, in fostering advancements and new ideas, and ultimately in recording progress.

BACKGROUND: Recent increases in the number of deposited membrane protein crystal structures necessitate the use of automated computational tools to position them within the lipid bilayer. Identifying the correct orientation allows us to study the complex relationship between sequence, structure and the lipid environment, which is otherwise challenging to investigate using experimental techniques due to the difficulty in crystallising membrane proteins embedded within intact membranes. RESULTS: We have developed a knowledge-based membrane potential, calculated by the statistical analysis of transmembrane protein structures, coupled with a combination of genetic and direct search algorithms, and demonstrate its use in positioning proteins in membranes, refinement of membrane protein models and in decoy discrimination. CONCLUSIONS: Our method is able to quickly and accurately orientate both alpha-helical and beta-barrel membrane proteins within the lipid bilayer, showing closer agreement with experimentally determined values than existing approaches. We also demonstrate both consistent and significant refinement of membrane protein models and the effective discrimination between native and decoy structures. Source code is available under an open source license from http://bioinf.cs.ucl.ac.uk/downloads/memembed/.

MOTIVATION: The accurate prediction of residue-residue contacts, critical for maintaining the native fold of a protein, remains an open problem in the field of structural bioinformatics. Interest in this long-standing problem has increased recently with algorithmic improvements and the rapid growth in the sizes of sequence families. Progress could have major impacts in both structure and function prediction to name but two benefits. Sequence-based contact predictions are usually made by identifying correlated mutations within multiple sequence alignments (MSAs), most commonly through the information-theoretic approach of calculating mutual information between pairs of sites in proteins. These predictions are often inaccurate because the true covariation signal in the MSA is often masked by biases from many ancillary indirect-coupling or phylogenetic effects. Here we present a novel method, PSICOV, which introduces the use of sparse inverse covariance estimation to the problem of protein contact prediction. Our method builds on work which had previously demonstrated corrections for phylogenetic and entropic correlation noise and allows accurate discrimination of direct from indirectly coupled mutation correlations in the MSA. RESULTS: PSICOV displays a mean precision substantially better than the best performing normalized mutual information approach and Bayesian networks. For 118 out of 150 targets, the L/5 (i.e. top-L/5 predictions for a protein of length L) precision for long-range contacts (sequence separation >23) was ≥ 0.5, which represents an improvement sufficient to be of significant benefit in protein structure prediction or model quality assessment. AVAILABILITY: The PSICOV source code can be downloaded from http://bioinf.cs.ucl.ac.uk/downloads/PSICOV.

BACKGROUND: Alpha-helical transmembrane channel and transporter proteins play vital roles in a diverse range of essential biological processes and are crucial in facilitating the passage of ions and molecules across the lipid bilayer. However, the experimental difficulties associated with obtaining high quality crystals has led to their significant under-representation in structural databases. Computational methods that can identify structural features from sequence alone are therefore of high importance. RESULTS: We present a method capable of automatically identifying pore-lining regions in transmembrane proteins from sequence information alone, which can then be used to determine the pore stoichiometry. By labelling pore-lining residues in crystal structures using geometric criteria, we have trained a support vector machine classifier to predict the likelihood of a transmembrane helix being involved in pore formation. Results from testing this approach under stringent cross-validation indicate that prediction accuracy of 72% is possible, while a support vector regression model is able to predict the number of subunits participating in the pore with 62% accuracy. CONCLUSION: To our knowledge, this is the first tool capable of identifying pore-lining regions in proteins and we present the results of applying it to a data set of sequences with available crystal structures. Our method provides a way to characterise pores in transmembrane proteins and may even provide a starting point for discovering novel routes of therapeutic intervention in a number of important diseases. This software is freely available as source code from: http://bioinf.cs.ucl.ac.uk/downloads/memsat-svm/.

MOTIVATION: The experimental difficulties of alpha-helical transmembrane protein structure determination make this class of protein an important target for sequence-based structure prediction tools. The MEMPACK prediction server allows users to submit a transmembrane protein sequence and returns transmembrane topology, lipid exposure, residue contacts, helix-helix interactions and helical packing arrangement predictions in both plain text and graphical formats using a number of novel machine learning-based algorithms. AVAILABILITY: The server can be accessed as a new component of the PSIPRED portal by at http://bioinf.cs.ucl.ac.uk/psipred/.

Alpha-helical transmembrane proteins constitute roughly 30% of a typical genome and are involved in a wide variety of important biological processes including cell signalling, transport of membrane-impermeable molecules and cell recognition. Despite significant efforts to predict transmembrane protein topology, comparatively little attention has been directed toward developing a method to pack the helices together. Here, we present a novel approach to predict lipid exposure, residue contacts, helix-helix interactions and finally the optimal helical packing arrangement of transmembrane proteins. Using molecular dynamics data, we have trained and cross-validated a support vector machine (SVM) classifier to predict per residue lipid exposure with 69% accuracy. This information is combined with additional features to train a second SVM to predict residue contacts which are then used to determine helix-helix interaction with up to 65% accuracy under stringent cross-validation on a non-redundant test set. Our method is also able to discriminate native from decoy helical packing arrangements with up to 70% accuracy. Finally, we employ a force-directed algorithm to construct the optimal helical packing arrangement which demonstrates success for proteins containing up to 13 transmembrane helices. This software is freely available as source code from http://bioinf.cs.ucl.ac.uk/memsat/mempack/.

MOTIVATION: Generation of structural models and recognition of homologous relationships for unannotated protein sequences are fundamental problems in bioinformatics. Improving the sensitivity and selectivity of methods designed for these two tasks therefore has downstream benefits for many other bioinformatics applications. RESULTS: We describe the latest implementation of the GenTHREADER method for structure prediction on a genomic scale. The method combines profile-profile alignments with secondary-structure specific gap-penalties, classic pair- and solvation potentials using a linear combination optimized with a regression SVM model. We find this combination significantly improves both detection of useful templates and accuracy of sequence-structure alignments relative to other competitive approaches. We further present a second implementation of the protocol designed for the task of discriminating superfamilies from one another. This method, pDomTHREADER, is the first to incorporate both sequence and structural data directly in this task and improves sensitivity and selectivity over the standard version of pGenTHREADER and three other standard methods for remote homology detection.

BACKGROUND: The classification of protein domains in the CATH resource is primarily based on structural comparisons, sequence similarity and manual analysis. One of the main bottlenecks in the processing of new entries is the evaluation of 'borderline' cases by human curators with reference to the literature, and better tools for helping both expert and non-expert users quickly identify relevant functional information from text are urgently needed. A text based method for protein classification is presented, which complements the existing sequence and structure-based approaches, especially in cases exhibiting low similarity to existing members and requiring manual intervention. The method is based on the assumption that textual similarity between sets of documents relating to proteins reflects biological function similarities and can be exploited to make classification decisions. RESULTS: An optimal strategy for the text comparisons was identified by using an established gold standard enzyme dataset. Filtering of the abstracts using a machine learning approach to discriminate sentences containing functional, structural and classification information that are relevant to the protein classification task improved performance. Testing this classification scheme on a dataset of 'borderline' protein domains that lack significant sequence or structure similarity to classified proteins showed that although, as expected, the structural similarity classifiers perform better on average, there is a significant benefit in incorporating text similarity in logistic regression models, indicating significant orthogonality in this additional information. Coverage was significantly increased especially at low error rates, which is important for routine classification tasks: 15.3% for the combined structure and text classifier compared to 10% for the structural classifier alone, at 10-3 error rate. Finally when only the highest scoring predictions were used to infer classification, an extra 4.2% of correct decisions were made by the combined classifier. CONCLUSION: We have described a simple text based method to classify protein domains that demonstrates an improvement over existing methods. The method is unique in incorporating structural and text based classifiers directly and is particularly useful in cases where inconclusive evidence from sequence or structure similarity requires laborious manual classification.

BACKGROUND: Alanine scanning mutagenesis is a powerful experimental methodology for investigating the structural and energetic characteristics of protein complexes. Individual amino-acids are systematically mutated to alanine and changes in free energy of binding (DeltaDeltaG) measured. Several experiments have shown that protein-protein interactions are critically dependent on just a few residues ("hot spots") at the interface. Hot spots make a dominant contribution to the free energy of binding and if mutated they can disrupt the interaction. As mutagenesis studies require significant experimental efforts, there is a need for accurate and reliable computational methods. Such methods would also add to our understanding of the determinants of affinity and specificity in protein-protein recognition. RESULTS: We present a novel computational strategy to identify hot spot residues, given the structure of a complex. We consider the basic energetic terms that contribute to hot spot interactions, i.e. van der Waals potentials, solvation energy, hydrogen bonds and Coulomb electrostatics. We treat them as input features and use machine learning algorithms such as Support Vector Machines and Gaussian Processes to optimally combine and integrate them, based on a set of training examples of alanine mutations. We show that our approach is effective in predicting hot spots and it compares favourably to other available methods. In particular we find the best performances using Transductive Support Vector Machines, a semi-supervised learning scheme. When hot spots are defined as those residues for which DeltaDeltaG >or= 2 kcal/mol, our method achieves a precision and a recall respectively of 56% and 65%. CONCLUSION: We have developed an hybrid scheme in which energy terms are used as input features of machine learning models. This strategy combines the strengths of machine learning and energy-based methods. Although so far these two types of approaches have mainly been applied separately to biomolecular problems, the results of our investigation indicate that there are substantial benefits to be gained by their integration.

BACKGROUND: Alpha-helical transmembrane (TM) proteins are involved in a wide range of important biological processes such as cell signaling, transport of membrane-impermeable molecules, cell-cell communication, cell recognition and cell adhesion. Many are also prime drug targets, and it has been estimated that more than half of all drugs currently on the market target membrane proteins. However, due to the experimental difficulties involved in obtaining high quality crystals, this class of protein is severely under-represented in structural databases. In the absence of structural data, sequence-based prediction methods allow TM protein topology to be investigated. RESULTS: We present a support vector machine-based (SVM) TM protein topology predictor that integrates both signal peptide and re-entrant helix prediction, benchmarked with full cross-validation on a novel data set of 131 sequences with known crystal structures. The method achieves topology prediction accuracy of 89%, while signal peptides and re-entrant helices are predicted with 93% and 44% accuracy respectively. An additional SVM trained to discriminate between globular and TM proteins detected zero false positives, with a low false negative rate of 0.4%. We present the results of applying these tools to a number of complete genomes. Source code, data sets and a web server are freely available from http://bioinf.cs.ucl.ac.uk/psipred/. CONCLUSION: The high accuracy of TM topology prediction which includes detection of both signal peptides and re-entrant helices, combined with the ability to effectively discriminate between TM and globular proteins, make this method ideally suited to whole genome annotation of alpha-helical transmembrane proteins.

BACKGROUND: We present a novel method of protein fold decoy discrimination using machine learning, more specifically using neural networks. Here, decoy discrimination is represented as a machine learning problem, where neural networks are used to learn the native-like features of protein structures using a set of positive and negative training examples. A set of native protein structures provides the positive training examples, while negative training examples are simulated decoy structures obtained by reversing the sequences of native structures. Various features are extracted from the training dataset of positive and negative examples and used as inputs to the neural networks. RESULTS: Results have shown that the best performing neural network is the one that uses input information comprising of PSI-BLAST 1 profiles of residue pairs, pairwise distance and the relative solvent accessibilities of the residues. This neural network is the best among all methods tested in discriminating the native structure from a set of decoys for all decoy datasets tested. CONCLUSION: This method is demonstrated to be viable, and furthermore evolutionary information is successfully used in the neural networks to improve decoy discrimination.

MOTIVATION: Many important biological processes such as cell signaling, transport of membrane-impermeable molecules, cell-cell communication, cell recognition and cell adhesion are mediated by membrane proteins. Unfortunately, as these proteins are not water soluble, it is extremely hard to experimentally determine their structure. Therefore, improved methods for predicting the structure of these proteins are vital in biological research. In order to improve transmembrane topology prediction, we evaluate the combined use of both integrated signal peptide prediction and evolutionary information in a single algorithm. RESULTS: A new method (MEMSAT3) for predicting transmembrane protein topology from sequence profiles is described and benchmarked with full cross-validation on a standard data set of 184 transmembrane proteins. The method is found to predict both the correct topology and the locations of transmembrane segments for 80% of the test set. This compares with accuracies of 62-72% for other popular methods on the same benchmark. By using a second neural network specifically to discriminate transmembrane from globular proteins, a very low overall false positive rate (0.5%) can also be achieved in detecting transmembrane proteins. AVAILABILITY: An implementation of the described method is available both as a web server (http://www.psipred.net) and as downloadable source code from http://bioinf.cs.ucl.ac.uk/memsat. Both the server and source code files are free to non-commercial users. Benchmark and training data are also available from http://bioinf.cs.ucl.ac.uk/memsat.

Natively unstructured regions are a common feature of eukaryotic proteomes. Between 30% and 60% of proteins are predicted to contain long stretches of disordered residues, and not only have many of these regions been confirmed experimentally, but they have also been found to be essential for protein function. In this study, we directly address the potential contribution of protein disorder in predicting protein function using standard Gene Ontology (GO) categories. Initially we analyse the occurrence of protein disorder in the human proteome and report ontology categories that are enriched in disordered proteins. Pattern analysis of the distributions of disordered regions in human sequences demonstrated that the functions of intrinsically disordered proteins are both length- and position-dependent. These dependencies were then encoded in feature vectors to quantify the contribution of disorder in human protein function prediction using Support Vector Machine classifiers. The prediction accuracies of 26 GO categories relating to signalling and molecular recognition are improved using the disorder features. The most significant improvements were observed for kinase, phosphorylation, growth factor, and helicase categories. Furthermore, we provide predicted GO term assignments using these classifiers for a set of unannotated and orphan human proteins. In this study, the importance of capturing protein disorder information and its value in function prediction is demonstrated. The GO category classifiers generated can be used to provide more reliable predictions and further insights into the behaviour of orphan and unannotated proteins.

BACKGROUND: Many biological processes involve the physical interaction between protein domains. Understanding these functional associations requires knowledge of the molecular structure. Experimental investigations though present considerable difficulties and there is therefore a need for accurate and reliable computational methods. In this paper we present a novel method that seeks to dock protein domains using a contact map representation. Rather than providing a full three dimensional model of the complex, the method predicts contacting residues across the interface. We use a scoring function that combines structural, physicochemical and evolutionary information, where each potential residue contact is assigned a value according to the scoring function and the hypothesis is that the real configuration of contacts is the one that maximizes the score. The search is performed with a simulated annealing algorithm directly in contact space. RESULTS: We have tested the method on interacting domain pairs that are part of the same protein (intra-molecular domains). We show that it correctly predicts some contacts and that predicted residues tend to be significantly closer to each other than other pairs of residues in the same domains. Moreover we find that predicted contacts can often discriminate the best model (or the native structure, if present) among a set of optimal solutions generated by a standard docking procedure. CONCLUSION: Contact docking appears feasible and able to complement other computational methods for the prediction of protein-protein interactions. With respect to more standard docking algorithms it might be more suitable to handle protein conformational changes and to predict complexes starting from protein models.

BACKGROUND: In order to maintain the most comprehensive structural annotation databases we must carry out regular updates for each proteome using the latest profile-profile fold recognition methods. The ability to carry out these updates on demand is necessary to keep pace with the regular updates of sequence and structure databases. Providing the highest quality structural models requires the most intensive profile-profile fold recognition methods running with the very latest available sequence databases and fold libraries. However, running these methods on such a regular basis for every sequenced proteome requires large amounts of processing power. In this paper we describe and benchmark the JYDE (Job Yield Distribution Environment) system, which is a meta-scheduler designed to work above cluster schedulers, such as Sun Grid Engine (SGE) or Condor. We demonstrate the ability of JYDE to distribute the load of genomic-scale fold recognition across multiple independent Grid domains. We use the most recent profile-profile version of our mGenTHREADER software in order to annotate the latest version of the Human proteome against the latest sequence and structure databases in as short a time as possible. RESULTS: We show that our JYDE system is able to scale to large numbers of intensive fold recognition jobs running across several independent computer clusters. Using our JYDE system we have been able to annotate 99.9% of the protein sequences within the Human proteome in less than 24 hours, by harnessing over 500 CPUs from 3 independent Grid domains. CONCLUSION: This study clearly demonstrates the feasibility of carrying out on demand high quality structural annotations for the proteomes of major eukaryotic organisms. Specifically, we have shown that it is now possible to provide complete regular updates of profile-profile based fold recognition models for entire eukaryotic proteomes, through the use of Grid middleware such as JYDE.

MOTIVATION: The ability of a simple method (MODCHECK) to determine the sequence-structure compatibility of a set of structural models generated by fold recognition is tested in a thorough benchmark analysis. Four Model Quality Assessment Programs (MQAPs) were tested on 188 targets from the latest LiveBench-9 automated structure evaluation experiment. We systematically test and evaluate whether the MQAP methods can successfully detect native-like models. RESULTS: We show that compared with the other three methods tested MODCHECK is the most reliable method for consistently performing the best top model selection and for ranking the models. In addition, we show that the choice of model similarity score used to assess a model's similarity to the experimental structure can influence the overall performance of these tools. Although these MQAP methods fail to improve the model selection performance for methods that already incorporate protein three dimension (3D) structural information, an improvement is observed for methods that are purely sequence-based, including the best profile-profile methods. This suggests that even the best sequence-based fold recognition methods can still be improved by taking into account the 3D structural information. CONTACT: [email protected]

MOTIVATION: Converting the vast quantity of free-format text found in journals into a concise, structured format makes the researcher's quest for information easier. Recently, several information extraction systems have been developed that attempt to simplify the retrieval and analysis of biological and medical data. Most of this work has used the abstract alone, owing to the convenience of access and the quality of data. Abstracts are generally available through central collections with easy direct access (e.g. PubMed). The full-text papers contain more information, but are distributed across many locations (e.g. publishers' web sites, journal web sites and local repositories), making access more difficult. In this paper, we present BioRAT, a new information extraction (IE) tool, specifically designed to perform biomedical IE, and which is able to locate and analyse both abstracts and full-length papers. BioRAT is a Biological Research Assistant for Text mining, and incorporates a document search ability with domain-specific IE. RESULTS: We show first, that BioRAT performs as well as existing systems, when applied to abstracts; and second, that significantly more information is available to BioRAT through the full-length papers than via the abstracts alone. Typically, less than half of the available information is extracted from the abstract, with the majority coming from the body of each paper. Overall, BioRAT recalled 20.31% of the target facts from the abstracts with 55.07% precision, and achieved 43.6% recall with 51.25% precision on full-length papers.

Abstract Summary: The Genomic Threading Database currently contains structural annotations for the genomes of over 100 recently sequenced organisms. Annotations are carried out by using our modified GenTHREADER software and through implementing grid technology. Availability: http://bioinf.cs.ucl.ac.uk/GTD Supplementary information: http://bioinf.cs.ucl.ac.uk/GTD/demo.avi, http://bioinf.cs.ucl.ac.uk/GTD/summaryView.html, http://bioinf.cs.ucl.ac.uk/GTD/figure1.pdf, http://www.e-protein.org

UNLABELLED: Dynamically disordered regions appear to be relatively abundant in eukaryotic proteomes. The DISOPRED server allows users to submit a protein sequence, and returns a probability estimate of each residue in the sequence being disordered. The results are sent in both plain text and graphical formats, and the server can also supply predictions of secondary structure to provide further structural information. AVAILABILITY: The server can be accessed by non-commercial users at http://bioinf.cs.ucl.ac.uk/disopred/

MOTIVATION: In order to enhance genome annotation, the fully automatic fold recognition method GenTHREADER has been improved and benchmarked. The previous version of GenTHREADER consisted of a simple neural network which was trained to combine sequence alignment score, length information and energy potentials derived from threading into a single score representing the relationship between two proteins, as designated by CATH. The improved version incorporates PSI-BLAST searches, which have been jumpstarted with structural alignment profiles from FSSP, and now also makes use of PSIPRED predicted secondary structure and bi-directional scoring in order to calculate the final alignment score. Pairwise potentials and solvation potentials are calculated from the given sequence alignment which are then used as inputs to a multi-layer, feed-forward neural network, along with the alignment score, alignment length and sequence length. The neural network has also been expanded to accommodate the secondary structure element alignment (SSEA) score as an extra input and it is now trained to learn the FSSP Z-score as a measurement of similarity between two proteins. RESULTS: The improvements made to GenTHREADER increase the number of remote homologues that can be detected with a low error rate, implying higher reliability of score, whilst also increasing the quality of the models produced. We find that up to five times as many true positives can be detected with low error rate per query. Total MaxSub score is doubled at low false positive rates using the improved method. AVAILABILITY: http://www.psipred.net.

MOTIVATION: A new method that uses support vector machines (SVMs) to predict protein secondary structure is described and evaluated. The study is designed to develop a reliable prediction method using an alternative technique and to investigate the applicability of SVMs to this type of bioinformatics problem. METHODS: Binary SVMs are trained to discriminate between two structural classes. The binary classifiers are combined in several ways to predict multi-class secondary structure. RESULTS: The average three-state prediction accuracy per protein (Q(3)) is estimated by cross-validation to be 77.07 +/- 0.26% with a segment overlap (Sov) score of 73.32 +/- 0.39%. The SVM performs similarly to the 'state-of-the-art' PSIPRED prediction method on a non-homologous test set of 121 proteins despite being trained on substantially fewer examples. A simple consensus of the SVM, PSIPRED and PROFsec achieves significantly higher prediction accuracy than the individual methods.

As the number of completely sequenced genomes rapidly increases, including now the complete Human Genome sequence, the post-genomic problems of genome-scale protein structure determination and the issue of gene function identification become ever more pressing. In fact, these problems can be seen as interrelated in that experimentally determining or predicting or the structure of proteins encoded by genes of interest is one possible means to glean subtle hints as to the functions of these genes. The applicability of this approach to gene characterisation is reviewed, along with a brief survey of the reliability of large-scale protein structure prediction methods and the prospects for the development of new prediction methods.

MOTIVATION: What constitutes a baseline level of success for protein fold recognition methods? As fold recognition benchmarks are often presented without any thought to the results that might be expected from a purely random set of predictions, an analysis of fold recognition baselines is long overdue. Given varying amounts of basic information about a protein-ranging from the length of the sequence to a knowledge of its secondary structure-to what extent can the fold be determined by intelligent guesswork? Can simple methods that make use of secondary structure information assign folds more accurately than purely random methods and could these methods be used to construct viable hierarchical classifications? EXPERIMENTS PERFORMED: A number of rapid automatic methods which score similarities between protein domains were devised and tested. These methods ranged from those that incorporated no secondary structure information, such as measuring absolute differences in sequence lengths, to more complex alignments of secondary structure elements. Each method was assessed for accuracy by comparison with the Class Architecture Topology Homology (CATH) classification. Methods were rated against both a random baseline fold assignment method as a lower control and FSSP as an upper control. Similarity trees were constructed in order to evaluate the accuracy of optimum methods at producing a classification of structure. RESULTS: Using a rigorous comparison of methods with CATH, the random fold assignment method set a lower baseline of 11% true positives allowing for 3% false positives and FSSP set an upper benchmark of 47% true positives at 3% false positives. The optimum secondary structure alignment method used here achieved 27% true positives at 3% false positives. Using a less rigorous Critical Assessment of Structure Prediction (CASP)-like sensitivity measurement the random assignment achieved 6%, FSSP-59% and the optimum secondary structure alignment method-32%. Similarity trees produced by the optimum method illustrate that these methods cannot be used alone to produce a viable protein structural classification system. CONCLUSIONS: Simple methods that use perfect secondary structure information to assign folds cannot produce an accurate protein taxonomy, however they do provide useful baselines for fold recognition. In terms of a typical CASP assessment our results suggest that approximately 6% of targets with folds in the databases could be assigned correctly by randomly guessing, and as many as 32% could be recognised by trivial secondary structure comparison methods, given knowledge of their correct secondary structures.

SUMMARY: The PSIPRED protein structure prediction server allows users to submit a protein sequence, perform a prediction of their choice and receive the results of the prediction both textually via e-mail and graphically via the web. The user may select one of three prediction methods to apply to their sequence: PSIPRED, a highly accurate secondary structure prediction method; MEMSAT 2, a new version of a widely used transmembrane topology prediction method; or GenTHREADER, a sequence profile based fold recognition method. AVAILABILITY: Freely available to non-commercial users at http://globin.bio.warwick.ac.uk/psipred/

An efficient means for generating mutation data matrices from large numbers of protein sequences is presented here. By means of an approximate peptide-based sequence comparison algorithm, the set sequences are clustered at the 85% identity level. The closest relating pairs of sequences are aligned, and observed amino acid exchanges tallied in a matrix. The raw mutation frequency matrix is processed in a similar way to that described by Dayhoff et al. (1978), and so the resulting matrices may be easily used in current sequence analysis applications, in place of the standard mutation data matrices, which have not been updated for 13 years. The method is fast enough to process the entire SWISS-PROT databank in 20 h on a Sun SPARCstation 1, and is fast enough to generate a matrix from a specific family or class of proteins in minutes. Differences observed between our 250 PAM mutation data matrix and the matrix calculated by Dayhoff et al. are briefly discussed.