A current mineral exploration focus is the development of tools to identify magmatic districts predisposed to host porphyry copper deposits. In this paper, we train and test four, common, supervised machine learning algorithms: logistic regression, support vector machines, artificial neural networks (ANN) and Random Forest to classify metallogenic ‘fertility’ in arc magmas based on whole-rock geochemistry. We outline pre-processing steps that can be used to mitigate against the undesirable characteristics of geochemical data (high multicollinearity, sparsity, missing values, class imbalance and compositional data effects) and therefore produce more meaningful results. We evaluate the classification accuracy of each supervised machine learning technique using a tenfold cross-validation technique and by testing the models on deposits unseen during the training process. This yields 81–83% accuracy for all classifiers, and receiver operating characteristic (ROC) curves have mean area under curve (AUC) scores of 87–89% indicating the probability of ranking a ‘fertile’ rock higher than an ‘unfertile’ rock. By contrast, bivariate classification schemes show much lower performance, demonstrating the value of classifying geochemical data in high dimension space. Principal component analysis suggests that porphyry-fertile magmas fractionate deep in the arc crust, and that calc-alkaline magmas associated with Cu-rich porphyries evolve deeper in the crust than more alkaline magmas linked with Au-rich porphyries. Feature analysis of the machine learning classifiers suggests that the most important parameters associated with fertile magmas are low Mn, high Al, high Sr, high K and listric REE patterns. These signatures further highlight the association of porphyry Cu deposits with hydrous arc magmas that undergo amphibole fractionation in the deep arc crust.

No abstract

No abstract

No abstract

IntroductionHypoglycemia elicits coordinated counter-regulatory neuroendocrine responses. The extent to which this process involves an increased drive to eat, together with greater preference for foods high in carbohydrate content, is unclear. Our objective was to examine this effect in children and adolescents (age 5–19 years) without diabetes and no prior known experience of hypoglycemic episodes.Research design and methodsWe administered a computerised task designed to examine preference for high-carbohydrate foods (sweet and savory) to pediatric patients (n=26) undergoing an insulin tolerance test as part of the routine clinical assessment of pituitary hormone secretory capacity. The task was completed at baseline and three time points after intravenous infusion of insulin (approximately 7, 20 and 90 min).ResultsAlthough all patients reached insulin-induced hypoglycemia (mean venous glucose at nadir=1.9 mmol/L), there was moderate evidence of no effect on preference for high-carbohydrate foods (moderate evidence for the null hypothesis) compared with euglycemia. Patients also did not display an increase in selection of foods of high compared with low energy density. Sensitivity of the task was demonstrated by decreased preference for sweet, high-carbohydrate foods after consumption of sweet food and drink.ConclusionsResults support the view that acute hypoglycemia does not automatically prompt the choice of high-carbohydrate foods for rapid glucose restoration, and further stresses the importance that people and families with children vulnerable to hypoglycemic episodes ensure that ‘rapidly absorbed glucose rescue therapy’ is always available.