Stuart Aitken

Real-time imaging of cotranscriptional splicing reveals a kinetic model that reduces noise: implications for alternative splicing regulation.

A combination of several rate-limiting steps allows for efficient control of alternative splicing.

Mining housekeeping genes with a Naive Bayes classifier

BackgroundTraditionally, housekeeping and tissue specific genes have been classified using direct assay of mRNA presence across different tissues, but these experiments are costly and the results not easy to compare and reproduce.ResultsIn this work, a Naive Bayes classifier based only on physical and functional characteristics of genes already available in databases, like exon length and measures of chromatin compactness, has achieved a 97% success rate in classification of human housekeeping genes (93% for mouse and 90% for fruit fly).ConclusionThe newly obtained lists of housekeeping and tissue specific genes adhere to the expected functions and tissue expression patterns for the two classes. Overall, the classifier shows promise, and in the future additional attributes might be included to improve its discriminating power.

Mapping between the OBO and OWL ontology languages

BackgroundOntologies are commonly used in biomedicine to organize concepts to describe domains such as anatomies, environments, experiment, taxonomies etc. NCBO BioPortal currently hosts about 180 different biomedical ontologies. These ontologies have been mainly expressed in either the Open Biomedical Ontology (OBO) format or the Web Ontology Language (OWL). OBO emerged from the Gene Ontology, and supports most of the biomedical ontology content. In comparison, OWL is a Semantic Web language, and is supported by the World Wide Web consortium together with integral query languages, rule languages and distributed infrastructure for information interchange. These features are highly desirable for the OBO content as well. A convenient method for leveraging these features for OBO ontologies is by transforming OBO ontologies to OWL.ResultsWe have developed a methodology for translating OBO ontologies to OWL using the organization of the Semantic Web itself to guide the work. The approach reveals that the constructs of OBO can be grouped together to form a similar layer cake. Thus we were able to decompose the problem into two parts. Most OBO constructs have easy and obvious equivalence to a construct in OWL. A small subset of OBO constructs requires deeper consideration. We have defined transformations for all constructs in an effort to foster a standard common mapping between OBO and OWL. Our mapping produces OWL-DL, a Description Logics based subset of OWL with desirable computational properties for efficiency and correctness. Our Java implementation of the mapping is part of the official Gene Ontology project source.ConclusionsOur transformation system provides a lossless roundtrip mapping for OBO ontologies, i.e. an OBO ontology may be translated to OWL and back without loss of knowledge. In addition, it provides a roadmap for bridging the gap between the two ontology languages in order to enable the use of ontology content in a language independent manner.

Nested sampling for parameter inference in systems biology: application to an exemplar circadian model

BackgroundModel selection and parameter inference are complex problems that have yet to be fully addressed in systems biology. In contrast with parameter optimisation, parameter inference computes both the parameter means and their standard deviations (or full posterior distributions), thus yielding important information on the extent to which the data and the model topology constrain the inferred parameter values.ResultsWe report on the application of nested sampling, a statistical approach to computing the Bayesian evidence Z, to the inference of parameters, and the estimation of log Z in an established model of circadian rhythms. A ten-fold difference in the coefficient of variation between degradation and transcription parameters is demonstrated. We further show that the uncertainty remaining in the parameter values is reduced by the analysis of increasing numbers of circadian cycles of data, up to 4 cycles, but is unaffected by sampling the data more frequently. Novel algorithms for calculating the likelihood of a model, and a characterisation of the performance of the nested sampling algorithm are also reported. The methods we develop considerably improve the computational efficiency of the likelihood calculation, and of the exploratory step within nested sampling.ConclusionsWe have demonstrated in an exemplar circadian model that the estimates of posterior parameter densities (as summarised by parameter means and standard deviations) are influenced predominately by the length of the time series, becoming more narrowly constrained as the number of circadian cycles considered increases. We have also shown the utility of the coefficient of variation for discriminating between highly-constrained and less-well constrained parameters.

EnzML : multi-label prediction of enzyme classes using InterPro signatures

BackgroundManual annotation of enzymatic functions cannot keep up with automatic genome sequencing. In this work we explore the capacity of InterPro sequence signatures to automatically predict enzymatic function.ResultsWe present EnzML, a multi-label classification method that can efficiently account also for proteins with multiple enzymatic functions: 50,000 in UniProt. EnzML was evaluated using a standard set of 300,747 proteins for which the manually curated Swiss-Prot and KEGG databases have agreeing Enzyme Commission (EC) annotations. EnzML achieved more than 98% subset accuracy (exact match of all correct Enzyme Commission classes of a protein) for the entire dataset and between 87 and 97% subset accuracy in reannotating eight entire proteomes: human, mouse, rat, mouse-ear cress, fruit fly, the S. pombe yeast, the E. coli bacterium and the M. jannaschii archaebacterium. To understand the role played by the dataset size, we compared the cross-evaluation results of smaller datasets, either constructed at random or from specific taxonomic domains such as archaea, bacteria, fungi, invertebrates, plants and vertebrates. The results were confirmed even when the redundancy in the dataset was reduced using UniRef100, UniRef90 or UniRef50 clusters.ConclusionsInterPro signatures are a compact and powerful attribute space for the prediction of enzymatic function. This representation makes multi-label machine learning feasible in reasonable time (30 minutes to train on 300,747 instances with 10,852 attributes and 2,201 class values) using the Mulan Binary Relevance Nearest Neighbours algorithm implementation (BR-kNN).

Modelling reveals kinetic advantages of co-transcriptional splicing.

Messenger RNA splicing is an essential and complex process for the removal of intron sequences. Whereas the composition of the splicing machinery is mostly known, the kinetics of splicing, the catalytic activity of splicing factors and the interdependency of transcription, splicing and mRNA 3′ end formation are less well understood. We propose a stochastic model of splicing kinetics that explains data obtained from high-resolution kinetic analyses of transcription, splicing and 3′ end formation during induction of an intron-containing reporter gene in budding yeast. Modelling reveals co-transcriptional splicing to be the most probable and most efficient splicing pathway for the reporter transcripts, due in part to a positive feedback mechanism for co-transcriptional second step splicing. Model comparison is used to assess the alternative representations of reactions. Modelling also indicates the functional coupling of transcription and splicing, because both the rate of initiation of transcription and the probability that step one of splicing occurs co-transcriptionally are reduced, when the second step of splicing is abolished in a mutant reporter.

Intersensory substitution in the blind

Abstract This study looks at the implications for theories of development of an intersensory surrogate for the blind—the Sonicguide. Results of both short- and longterm studies suggest that there is a critical age, before which use of the Sonicguide is qualitatively more effective than attempts at intervention begun after this age. Similarly the young infants in the sample show a much more rapid acquisition of ability to use the guide effectively. The importance of such results for theories of perceptual development is discussed.

A bioinformatics approach for identifying candidate transcriptional regulators of mesenchyme-to-epithelium transitions in mouse embryos†‡

This article reports a method for identifying groups of genes associated with tissues undergoing a particular process during mouse development. Given the Theiler stage at which each tissue starts the process, Boolean intersection analysis identifies genes expressed in some or all of these tissues both before the process starts and once it has started. This analysis is implemented in GXD‐search; this tool downloads appropriate gene sets from GXD, the mouse gene expression database, and performs the calculations. Applied to mesenchyme‐to‐epithelium transitions (MET), GXD‐search has identified Crabp1 and six transcriptional regulators (Cited1, Cited2, Meox1, Lhx1, Foxc1, and Foxc2) that are usually expressed in tissues about to undergo this process. Expression pattern analysis of these transcriptional regulators, mutations in each of which affect epithelial development, shows that this gene set is expressed in no other tissues and they are, thus, candidates for regulating MET. GXD‐search is downloadable from http://www.aiai.ed.ac.uk/project/biosphere/GXD‐search.html. Developmental Dynamics 237:2748–2754, 2008.

The SOFG Anatomy Entry List (SAEL): An Annotation Tool for Functional Genomics Data

A great deal of data in functional genomics studies needs to be annotated with low-resolution anatomical terms. For example, gene expression assays based on manually dissected samples (microarray, SAGE, etc.) need high-level anatomical terms to describe sample origin. First-pass annotation in high-throughput assays (e.g. large-scale in situ gene expression screens or phenotype screens) and bibliographic applications, such as selection of keywords, would also benefit from a minimum set of standard anatomical terms. Although only simple terms are required, the researcher faces serious practical problems of inconsistency and confusion, given the different aims and the range of complexity of existing anatomy ontologies. A Standards and Ontologies for Functional Genomics (SOFG) group therefore initiated discussions between several of the major anatomical ontologies for higher vertebrates. As we report here, one result of these discussions is a simple, accessible, controlled vocabulary of gross anatomical terms, the SOFG Anatomy Entry List (SAEL). The SAEL is available from http://www.sofg.org and is intended as a resource for biologists, curators, bioinformaticians and developers of software supporting functional genomics. It can be used directly for annotation in the contexts described above. Importantly, each term is linked to the corresponding term in each of the major anatomy ontologies. Where the simple list does not provide enough detail or sophistication, therefore, the researcher can use the SAEL to choose the appropriate ontology and move directly to the relevant term as an entry point. The SAEL links will also be used to support computational access to the respective ontologies.

Processivity and Coupling in Messenger RNA Transcription

Background The complexity of messenger RNA processing is now being uncovered by experimental techniques that are capable of detecting individual copies of mRNA in cells, and by quantitative real-time observations that reveal the kinetics. This processing is commonly modelled by permitting mRNA to be transcribed only when the promoter is in the on state. In this simple on/off model, the many processes involved in active transcription are represented by a single reaction. These processes include elongation, which has a minimum time for completion and processing that is not captured in the model. Methodology In this paper, we explore the impact on the mRNA distribution of representing the elongation process in more detail. Consideration of the mechanisms of elongation leads to two alternative models of the coupling between the elongating polymerase and the state of the promoter: Processivity allows polymerases to complete elongation irrespective of the promoter state, whereas coupling requires the promoter to be active to produce a full-length transcript. We demonstrate that these alternatives have a significant impact on the predicted distributions. Models are simulated by the Gillespie algorithm, and the third and fourth moments of the resulting distribution are computed in order to characterise the length of the tail, and sharpness of the peak. By this methodology, we show that the moments provide a concise summary of the distribution, showing statistically-significant differences across much of the feasible parameter range. Conclusions We conclude that processivity is not fully consistent with the on/off model unless the probability of successfully completing elongation is low—as has been observed. The results also suggest that some form of coupling between the promoter and a rate-limiting step in transcription may explain the cells inability to maintain high mRNA levels at low noise—a prediction of the on/off model that has no supporting evidence.