Ecological niche models are widely used in ecology and biogeography. Maxent is one of the most frequently used niche modeling tools, and many studies have aimed to optimize its performance. However, scholars have conflicting views on the treatment of predictor collinearity in Maxent modeling. Despite this lack of consensus, quantitative examinations of the effects of collinearity on Maxent modeling, especially in model transfer scenarios, are lacking. To address this knowledge gap, here we quantify the effects of collinearity under different scenarios of Maxent model training and projection. We separately examine the effects of predictor collinearity, collinearity shifts between training and testing data, and environmental novelty on model performance. We demonstrate that excluding highly correlated predictor variables does not significantly influence model performance. However, we find that collinearity shift and environmental novelty have significant negative effects on the performance of model transfer. We thus conclude that (a) Maxent is robust to predictor collinearity in model training; (b) the strategy of excluding highly correlated variables has little impact because Maxent accounts for redundant variables; and (c) collinearity shift and environmental novelty can negatively affect Maxent model transferability. We therefore recommend to quantify and report collinearity shift and environmental novelty to better infer model accuracy when models are spatially and/or temporally transferred.

lye@genome.wustl.edu

Climate change and human-mediated dispersal are increasingly influencing species' geographic distributions. Ecological niche models (ENMs) are widely used in forecasting species' distributions, but are weak in extrapolation to novel environments because they rely on available distributional data and do not incorporate mechanistic information, such as species' physiological response to abiotic conditions. To improve accuracy of ENMs, we incorporated physiological knowledge through Bayesian analysis. In a case study of the zebra mussel Dreissena polymorpha, we used native and global occurrences to obtain native and global models representing narrower and broader understanding of zebra mussel' response to temperature. We also obtained thermal limit and survival information for zebra mussel from peer-reviewed literature and used the two types of information separately and jointly to calibrate native models. We showed that, compared to global models, native models predicted lower relative probability of presence along zebra mussel's upper thermal limit, suggesting the shortcoming of native models in predicting zebra mussel's response to warm temperature. We also found that native models showed improved prediction of relative probability of presence when thermal limit was used alone, and best approximated global models when both thermal limit and survival data were used. Our result suggests that integration of physiological knowledge enhances extrapolation of ENM in novel environments. Our modeling framework can be generalized for other species or other physiological limits and may incorporate evolutionary information (e.g. evolved thermal tolerance), thus has the potential to improve predictions of species' invasive potential and distributional response to climate change.

No abstract

Diabetic retinopathy (DR) is a leading cause for blindness among working-aged adults. The growing prevalence of diabetes urges for cost-effective tools to improve the compliance of eye examinations for early detection of DR. The objective of this research is to identify essential predictors and develop predictive technologies for DR using electronic health records. We conducted a retrospective analysis on a derivation cohort with 3749 DR and 94,127 non-DR diabetic patients. In the analysis, an ensemble predictor selection method was employed to find essential predictors among 26 variables in demographics, duration of diabetes, complications and laboratory results. A predictive model and a risk index were built based on the selected, essential predictors, and then validated using another independent validation cohort with 869 DR and 6448 non-DR diabetic patients. Out of the 26 variables, 10 were identified to be essential for predicting DR. The predictive model achieved a 0.85 AUC on the derivation cohort and a 0.77 AUC on the validation cohort. For the risk index, the AUCs were 0.81 and 0.73 on the derivation and validation cohorts, respectively. The predictive technologies can provide an early warning sign that motivates patients to comply with eye examinations for early screening and potential treatments.