Search for relevant subsets of binary predictors in high dimensional regression for discovering the lead molecule

One of the main problems that the drug discovery research field confronts is to identify small molecules, modulators of protein function, which are likely to be therapeutically useful. Common practices rely on the screening of vast libraries of small molecules (often 1–2 million molecules) in order to identify a molecule, known as a lead molecule, which specifically inhibits or activates the protein function. To search for the lead molecule, we investigate the molecular structure, which generally consists of an extremely large number of fragments. Presence or absence of particular fragments, or groups of fragments, can strongly affect molecular properties. We study the relationship between molecular properties and its fragment composition by building a regression model, in which predictors, represented by binary variables indicating the presence or absence of fragments, are grouped in subsets and a bi-level penalization term is introduced for the high dimensionality of the problem. We evaluate the performance of this model in two simulation studies, comparing different penalization terms and different clustering techniques to derive the best predictor subsets structure. Both studies are characterized by small sets of data relative to the number of predictors under consideration. From the results of these simulation studies, we show that our approach can generate models able to identify key features and provide accurate predictions. The good performance of these models is then exhibited with real data about the MMP–12 enzyme.     

Unpacking the Displacement-Migration Nexus: Evidence from Northern Iraq

Recent conflicts have increasingly produced a twofold, forced movement of people: internal displacement and forced migration. While the co-occurrence of these phenomena is an acknowledged fact, there is not yet an understanding of their relation in terms of either the scale or the process contributing to it. This article seeks to fill this gap by questioning the relation between conflict-induced displacement and migration in and from Iraq. The analysis is based on survey data (500 questionnaires) and 29 semi-structured interviews with Iraqi IDPs residing in the Kurdistan Region of Iraq and national and international key informants. Based on the findings, the study concludes that the aspiration to leave Iraq is primarily attributable to the structural enduring condition of Iraq, rather than an individual condition – a finding that uneasily translates into the new framework for intervention in conflict-affected and fragile contexts premised on deterring migration.     

Spectral density-based and measure-preserving ABC for partially observed diffusion processes. An illustration on Hamiltonian SDEs

Approximate Bayesian computation (ABC) has become one of the major tools of likelihood-free statistical inference in complex mathematical models. Simultaneously, stochastic differential equations (SDEs) have developed to an established tool for modelling time-dependent, real-world phenomena with underlying random effects. When applying ABC to stochastic models, two major difficulties arise: First, the derivation of effective summary statistics and proper distances is particularly challenging, since simulations from the stochastic process under the same parameter configuration result in different trajectories. Second, exact simulation schemes to generate trajectories from the stochastic model are rarely available, requiring the derivation of suitable numerical methods for the synthetic data generation. To obtain summaries that are less sensitive to the intrinsic stochasticity of the model, we propose to build up the statistical method (e.g. the choice of the summary statistics) on the underlying structural properties of the model. Here, we focus on the existence of an invariant measure and we map the data to their estimated invariant density and invariant spectral density. Then, to ensure that these model properties are kept in the synthetic data generation, we adopt measure-preserving numerical splitting schemes. The derived property-based and measure-preserving ABC method is illustrated on the broad class of partially observed Hamiltonian type SDEs, both with simulated data and with real electroencephalography data. The derived summaries are particularly robust to the model simulation, and this fact, combined with the proposed reliable numerical scheme, yields accurate ABC inference. In contrast, the inference returned using standard numerical methods (Euler–Maruyama discretisation) fails. The proposed ingredients can be incorporated into any type of ABC algorithm and directly applied to all SDEs that are characterised by an invariant distribution and for which a measure-preserving numerical method can be derived.

     

Dangerous Climate Change: The Role for Risk Research

The notion of "dangerous climate change" constitutes an important development of the 1992 United Nations Framework Convention on Climate Change. It persists, however, as an ambiguous expression, sustained by multiple definitions of danger. It also implicitly contains the question of how to respond to the complex and multi-disciplinary risk issues that climate change poses. The invaluable role of the climate science community, which relies on risk assessments to characterize system uncertainties and to identify limits beyond which changes may become dangerous, is acknowledged. But this alone will not suffice to develop long-term policy. Decisions need to include other considerations, such as value judgments about potential risks, and societal and individual perceptions of "danger," which are often contested. This article explores links and cross-overs between the climate science and risk communication and perception approaches to defining danger. Drawing upon nine articles in this Special Issue of Risk Analysis, we examine a set of themes: limits of current scientific understanding; differentiated public perceptions of danger from climate change; social and cultural processes amplifying and attenuating perceptions of, and responses to, climate change; risk communication design; and new approaches to climate change decision making. The article reflects upon some of the difficulties inherent in responding to the issue in a coherent, interdisciplinary fashion, concluding nevertheless that action should be taken, while acknowledging the context-specificity of "danger." The need for new policy tools is emphasised, while research on nested solutions should be aimed at overcoming the disjunctures apparent in interpretations of climate change risks.     

The relationships between work characteristics and mental health: examining normal,reversed and reciprocal relationships in a 4-wave study

This longitudinal study examined the causal relationships between job demands, job control and supervisor support on the one hand and mental health on the other. Whereas we assumed that work characteristics affect mental health, we also examined reversed causal relationships (mental health influences work characteristics). Further, the topic of the appropriate time lag for testing causal relationships was addressed. Our hypotheses were tested in a 4-wave study among a heterogeneous sample of 668 Dutch employees using structural equation modelling. The results provide evidence for reciprocal causal relationships between the work characteristics and mental health, although the effects of work characteristics on well-being were causally predominant. The best model fit was found for a 1-year time lag. Compared to earlier—predominantly cross-sectional—results, the present study presents a stronger case for the effects of work characteristics on the development of strain. The results also emphasize the need for a dynamic view of the relationship between work and health; the one-directional viewpoint in many work stress models does not seem to fully capture the relations between work characteristics and well-being.     

Return to work: A comparison of two cognitive behavioural interventions in cases of work-related psychological complaints among the self-employed

Abstract

We investigated the effectiveness of cognitive behavioural therapy (CBT) and a combined intervention of workplace- and individual-focused techniques among self-employed people on sick leave owing to work-related psychological complaints (such as anxiety, depression, and burnout). Both interventions were based on CBT; however, one was conducted by psychotherapists and involved extensive CBT, while the other was delivered by “labour experts” and consisted of a brief CBT-derived intervention combined with both individual-focused and workplace interventions. One hundred and twenty-two self-employed people who had applied for sickness benefit from an insurance company enrolled in a randomized controlled design. These individuals were assessed before the intervention and then at 4 months and 10 months after the onset of the intervention. The outcome was assessed based on duration of sick leave until partial and full return to work and on psychological complaints. Significant effects on partial and full return were found in favour of the combined intervention: partial return occurred 17 and 30 days earlier in this group than in the CBT group and the control group, respectively. For full return to work, the difference was approximately 200 days. A decrease in psychological complaints was present in each condition but we found no significant interaction effects. The results suggest that work resumption should be addressed earlier in individuals receiving CBT. This insight is of value for the (scarce) literature concerning interventions for individuals who are on sick leave owing to work-related psychological complaints.     

Conditional failure occurrence rates for semi-Markov chains*

ABSTRACT

In the present paper, we aim at providing plug-in-type empirical estimators that enable us to quantify the contribution of each operational or/and non-functioning state to the failures of a system described by a semi-Markov model. In the discrete-time and finite state space semi-Markov framework, we study different conditional versions of an important reliability measure for random repairable systems, the failure occurrence rate, which is based on counting processes. The identification of potential failure contributors through the conditional counterparts of the failure occurrence rate is of paramount importance since it could lead to corrective actions that minimize the occurrence of the more important failure modes and therefore improve the reliability of the system. The aforementioned estimators are characterized by appealing asymptotic properties such as strong consistency and asymptotic normality. We further obtain detailed analytical expressions for the covariance matrices of the random vectors describing the conditional failure occurrence rates. As particular cases we present the failure occurrence rates for hidden (semi-) Markov models. We illustrate our results by means of a simulated study. Different applications are presented based on wind, earthquake and vibration data.