BACKGROUND: Recent advances in data analysis methods based on principles of Mendelian Randomisation, such as Egger regression and the weighted median estimator, add to the researcher's ability to infer cause-effect links from observational data. Now is the time to gauge the potential of these methods within specific areas of biomedical research. In this paper, we choose a study in metabolomics as an illustrative testbed. We apply Mendelian Randomisation methods in the analysis of data from the DILGOM (Dietary, Lifestyle and Genetic determinants of Obesity and Metabolic syndrome) study, in the context of an effort to identify molecular pathways of cardiovascular disease. In particular, our illustrative analysis addresses the question whether body mass, as measured by body mass index (BMI), exerts a causal effect on the concentrations of a collection of 137 cardiometabolic markers with different degrees of atherogenic power, such as the (highly atherogenic) lipoprotein metabolites with very low density (VLDLs) and the (protective) high density lipoprotein metabolites. RESULTS: We found strongest evidence of a positive BMI effect (that is, evidence that an increase in BMI causes an increase in the metabolite concentration) on those metabolites known to represent strong risk factors for coronary artery disease, such as the VLDLs, and evidence of a negative effect on protective biomarkers. CONCLUSIONS: The methods discussed represent a useful scientific tool, although they assume the validity of conditions that are (at best) only partially verifiable. This paper provides a rigorous account of such conditions. The results of our analysis provide a proof-of-concept illustration of the potential usefulness of Mendelian Randomisation in genomic biobank studies aiming to dissect the molecular causes of disease, and to identify candidate pharmacological targets.

MOTIVATION: Isobaric tag for relative and absolute quantitation (iTRAQ) is a widely used method in quantitative proteomics. A robust data analysis strategy is required to determine protein quantification reliability, i.e. changes due to biological regulation rather than technical variation, so that proteins that are differentially expressed can be identified. METHODS: Samples were created by mixing 5, 10, 15 and 20 μg Escherichia coli cell lysate with 100 μg of cell lysate from mouse, corresponding to expected relative fold changes of one for mouse proteins and from 0.25 to 4 for E.coli proteins. Relative quantification was carried out using eight channel isobaric tagging with iTRAQ reagent, and proteins were identified using a TripleTOF 5600 mass spectrometer. Technical variation inherent in this iTRAQ dataset was systematically investigated. RESULTS: A hierarchical statistical model was developed to use quantitative information at peptide level and protein level simultaneously to estimate variation present in each individual peptide and protein. A novel data analysis strategy for iTRAQ, denoted in short as WHATraq, was subsequently proposed with its performance evaluated by the proportion of E.coli proteins that are successfully identified as differentially expressed. Compared with two benchmark data analysis strategies WHATraq was able to identify at least 62.8% more true positive proteins that are differentially expressed. Further validated using a biological iTRAQ dataset including multiple biological replicates from varied murine cell lines, WHATraq performed consistently and identified 375% more proteins as being differentially expressed among different cell lines than the other data analysis strategies.

MOTIVATION: Several methods have recently been developed to analyse large-scale sets of physical interactions between proteins in terms of physical contacts between the constituent domains, often with a view to predicting new pairwise interactions. Our aim is to combine genomic interaction data, in which domain-domain contacts are not explicitly reported, with the domain-level structure of individual proteins, in order to learn about the structure of interacting protein pairs. Our approach is driven by the need to assess the evidence for physical contacts between domains in a statistically rigorous way. RESULTS: We develop a statistical approach that assigns p-values to pairs of domain superfamilies, measuring the strength of evidence within a set of protein interactions that domains from these superfamilies form contacts. A set of p-values is calculated for SCOP superfamily pairs, based on a pooled data set of interactions from yeast. These p-values can be used to predict which domains come into contact in an interacting protein pair. This predictive scheme is tested against protein complexes in the Protein Quaternary Structure (PQS) database, and is used to predict domain-domain contacts within 705 interacting protein pairs taken from our pooled data set.

This paper considers biomedical problems in which a sample of subjects, for example clinical patients, is monitored through time for purposes of individual prediction. Emphasis is on situations in which the monitoring generates data both in the form of a time series and in the form of events (development of a disease, death, etc.) observed on each subject over specified intervals of time. A Bayesian approach to the combined modeling of both types of data for purposes of prediction is presented. The proposed method merges ideas of Bayesian hierarchical modeling, nonparametric smoothing of time series data, survival analysis, and forecasting into a unified framework. Emphasis is on flexible modeling of the time series data based on stochastic process theory. The use of Markov chain Monte Carlo simulation to calculate the predictions of interest is discussed. Conditional independence graphs are used throughout for a clear presentation of the models. An application in the monitoring of transplant patients is presented.

The authors describe GAMEES II (Graphical Modeling Environment for Expert Systems II), a computer system built to manage arrays of Bayesian belief networks (BBNs). With regard to biomedical applications, the system makes it possible to represent probabilistic knowledge both for the individual patient and for population, enlarging the class of problems actually represented through BBNs. In addition, mathematical models can be represented through the BBN formalism. These characteristics were exploited to perform model-based patient monitoring and population pharmacokinetic/pharmacodynamic studies. BBNs are built within GAMEES II through a graphical interface which is provided with some statistical knowledge, in order to facilitate the construction of sound statistical models. Probabilistic inference is performed by stochastic simulation algorithms that do not give rise to computational problems when dealing with complex networks.>