A proof-of-concept study applying machine learning methods to putative risk factors for eating disorders: results from the multi-centre European project on healthy eating

BACKGROUND

Despite a wide range of proposed risk factors and theoretical models, prediction of eating disorder (ED) onset remains poor. This study undertook the first comparison of two machine learning (ML) approaches [penalised logistic regression (LASSO), and prediction rule ensembles (PREs)] to conventional logistic regression (LR) models to enhance prediction of ED onset and differential ED diagnoses from a range of putative risk factors.

METHOD

Data were part of a European Project and comprised 1402 participants, 642 ED patients [52% with anorexia nervosa (AN) and 40% with bulimia nervosa (BN)] and 760 controls. The Cross-Cultural Risk Factor Questionnaire, which assesses retrospectively a range of sociocultural and psychological ED risk factors occurring before the age of 12 years (46 predictors in total), was used.

RESULTS

All three statistical approaches had satisfactory model accuracy, with an average area under the curve (AUC) of 86% for predicting ED onset and 70% for predicting AN v. BN. Predictive performance was greatest for the two regression methods (LR and LASSO), although the PRE technique relied on fewer predictors with comparable accuracy. The individual risk factors differed depending on the outcome classification (EDs v. non-EDs and AN v. BN).

CONCLUSIONS

Even though the conventional LR performed comparably to the ML approaches in terms of predictive accuracy, the ML methods produced more parsimonious predictive models. ML approaches offer a viable way to modify screening practices for ED risk that balance accuracy against participant burden.

Application of fragment based virtual screening towards inhibition of bacterial N-acetyglucosaminidase$

A structure-based approach is applied for the development of inhibitors of bacterial N-acetyglucosaminidase (autolysin). Autolysins are enzymes involved in the degradation of peptidoglycan and therefore participate in bacterial cell growth and different lysis phenomena. Several studies indicate that by the inhibition of autolysins, and consequently of bacterial cell division, antibacterial activity can be obtained, thus paving the road to a novel group of therapeutics against human pathogens. As crystal structures of the autolysin E (AtlE)-ligand complexes were obtained in our laboratories, fragment-based virtual screening was the method of choice for the initial studies. Fragment libraries from various databases were merged to increase the number of compounds for the virtual screening. Twenty-four commercially available virtual hits were selected and subjected to quantitative analysis of binding interactions using the surface plasmon resonance technique. Twelve fragments showed fragment-AtlE interactions. For F1, the top hit of the virtual screening, a K_D of 228 µM was determined, while other fragments displayed non-stoichiometric binding. Blind docking of potential binders uncovers three possible allosteric sites. Ligands of N-acetyglucosaminidase identified in our study represent valuable information for the further development of AtlE inhibitors, which could in future represent antibacterial agents acting by a novel mode of action.

Lifetime course of eating disorders: design and validity testing of a new strategy to define the eating disorders phenotype

BACKGROUND

Aetiological studies of eating disorders would benefit from a solution to the problem of instability of eating disorder symptoms. We present an approach to defining an eating disorders phenotype based on the retrospective assessment of lifetime eating disorders symptoms to define a lifetime pattern of illness. We further validate this approach by testing the most common lifetime categories for differences in the prevalence of specific childhood personality traits.

METHOD

Ninety-seven females participated in this study, 35 with a current diagnosis of restricting anorexia nervosa, 32 with binge/purging subtype of anorexia nervosa and 30 with bulimia nervosa. Subjects were interviewed by a newly developed EATATE Lifetime Diagnostic Interview for a retrospective assessment of the lifetime course of eating disorders symptoms and childhood traits reflecting obsessive-compulsive personality.

RESULTS

The data illustrate the extensive instability of the eating disorders diagnosis. Four most common lifetime diagnostic categories were identified that significantly differ in the prevalence of childhood traits. Perfectionism and rigidity were more common in groups with a longer duration of underweight status, longer episodes of severe food restriction, excessive exercising, and shorter duration of binge eating.

CONCLUSIONS

The assessment of lifetime symptoms may produce a more accurate definition of the eating disorders phenotype. Obsessive-compulsive traits in childhood may moderate the course producing longer periods of underweight status. These findings may have important implications for nosology, treatment and future aetiological studies of eating disorders.

Thrombin receptor antagonists; recent advances in PAR-1 antagonist development

The receptor for the serine protease thrombin, the protease-activated receptor-1 (PAR-1), has been recently characterized. Its key roles in thrombin-stimulated human platelet activation, vascular endothelial and smooth muscle proliferation, inflammatory responses and neurodegeneration suggest receptor involvement in various disorders such as arterial thrombosis, atherosclerosis, restenosis, inflammation and myocardial infarction. It has been established that thrombin elicits the majority of its effects via PAR-1. PAR-1 has a novel mechanism of activation. The receptor, a member of the seven-transmembrane domain receptor family, is cleaved by thrombin at a specific site on the N-terminal extension, and a newly exposed N-terminus acts as a tethered ligand to activate the receptor itself. The need for development of a PAR-1 antagonist that may be valuable as a therapeutic agent has been recognized. An intriguing challenge is the necessity of the antagonist to compete with an intramolecular ligand while showing no intrinsic activity. The lead compounds were found to be synthetic peptides containing N-terminal hexapeptide or pentapeptide (Ser-Phe-Leu-Leu-Arg-Asn, Ser-Phe-Leu-Leu-Arg) or modified sequences (TRAPs; thrombin receptor-activating peptides), which exhibit full PAR-1 agonist activity. Selective PAR-1 antagonists have already been synthesized. Though their potency is still not enough to justify therapeutic use, it is clear that future progress will bring a novel class of drugs-thrombin receptor antagonists. The emphasis of this review, therefore, will be placed on advances in the discovery of potent and selective PAR-1 antagonists.