Sonia Migliorati

Toxicity on the luminescent bacterium Vibrio Fischeri (Beijerinck). I: QSAR equation for narcotics and polar narcotics

Toxicity data on chemicals, supposed to have a narcotic or polar narcotic toxicological mode of action, have been produced on the luminescent bacterium Vibrio fischeri using the Microtox test procedure. Advanced statistical methods have been used to calculate statistically sound values for ecotoxicological endpoints. Simple quantitative structure activity relationship (QSAR) equations were developed for narcotics and polar narcotics. These equations were compared with those proposed by the European Technical Guidance Document on Risk Assessment for other aquatic organisms (algae, Daphnia, and fish). Similarities and differences are discussed. The need for including the bacterial component in the ecotoxicological risk assessment for aquatic ecosystems is highlighted.

Toxicity on the luminescent bacterium Vibrio fischeri (Beijerinck). II: Response to complex mixtures of heterogeneous chemicals at low levels of individual components

The toxicity of eight complex mixtures of chemicals with different chemical structures and toxicological modes of action (narcotics, polar narcotics, herbicides, insecticides, fungicides) was tested on the luminescent bacterium Vibrio fischeri. There were maximum 84 individual chemicals in the mixtures. Suitable statistical approaches were applied for the comparison between experimental results and theoretical predictions. The results demonstrated that the two models of Concentration Addition (CA) and Independent Action (IA) are suitable to explain the effect of the mixtures.Even extremely lower concentrations of individual chemicals contributed to the effect of the mixtures. Synergistic effects were not observed in any of the tested mixtures. In particular, the CA approach well predicted the effects of six out of eight mixtures and slightly overestimated the effects of the remaining two mixtures. Therefore, the CA model can be proposed as a pragmatic and adequately protective approach for regulatory purposes.

SSD-based rating system for the classification of pesticide risk on biodiversity

A novel approach, based on Species sensitivity distribution (SSD), is proposed for the development of an index for classifying ecotoxicological pesticide risk in surface waters. In this approach, the concept of TER (Toxicity Exposure Ratio), commonly used in traditional risk indices, is substituted by the concept of PAF (Potentially Affected Fraction), which takes into account several species within the biological community of interest, rather than just a small number of indicator species assumed as being representative of the ecosystem. The procedure represents a probabilistic tool to quantitatively assess the ecotoxicological risk on biodiversity considering the distribution of toxicological sensitivity. It can be applied to assess chemical risk on generic aquatic and terrestrial communities as well as on site-specific natural communities. Examples of its application are shown for some pesticides in freshwater ecosystems. In order to overcome the problem of insufficient reliable ecotoxicological data, a methodology and related algorithms are proposed for predicting SSD curves for chemicals that do not have sufficient available data. The methodology is applicable within congeneric classes of chemicals and has been tested and statistically validated on a group of organophosphorus insecticides. Values and limitations of the approach are discussed.

Risk of POP mixtures on the Arctic food chain

The exposure of the Arctic ecosystem to persistent organic pollutants (POPs) was assessed through a review of literature data. Concentrations of 19 chemicals or congeneric groups were estimated for the highest levels of the Arctic food chain (Arctic cod, ringed seals, and polar bears). The ecotoxicological risk for seals, bears, and bear cubs was estimated by applying the concentration addition (CA) concept. The risk of POP mixtures was very low in seals. By contrast, the risk was 2 orders of magnitude higher than the risk threshold for adult polar bears and even more (3 orders of magnitude above the threshold) for bear cubs fed with contaminated milk. Based on the temporal trends available for many of the chemicals, the temporal trend of the mixture risk for bear cubs was calculated. Relative to the 1980s, a decrease in risk from the POP mixture is evident, mainly because of international control measures. However, the composition of the mixture substantially changes, and the contribution of new POPs (particularly perfluorooctane sulfonate) increases. These results support the effectiveness of control measures, such as those promulgated in the Stockholm Convention, as well as the urgent need for their implementation for new and emerging POPs. Environ Toxicol Chem 2017;36:1181-1192.

Adjusting p-Values when n Is Large in the Presence of Nuisance Parameters

A precise null hypothesis formulation (instead of the more realistic interval one) is usually adopted by statistical packages although it generally leads to excessive (and often misleading) rates of rejection whenever the sample size is large. In a previous paper (Migliorati and Ongaro, 2007) we proposed a calibration procedure aimed at adjusting test levels and p-values when testing the mean of a Normal model with known variance. We now address the more complicated calibration issues arising when a nuisance parameter (e.g., the variance) is present. As procedures for testing the interval null hypothesis available in the literature are shown to be unsatisfactory for calibration purposes, this entails, in particular, the construction of suitable new tests.

A generalization of the Dirichlet distribution

A new parametric family of distributions on the unit simplex is proposed and investigated. Such family, called flexible Dirichlet, is obtained by normalizing a correlated basis formed by a mixture of independent gamma random variables. The Dirichlet distribution is included as an inner point. The flexible Dirichlet is shown to exhibit a rich dependence pattern, capable of discriminating among many of the independence concepts relevant for compositional data. At the same time it can model multi-modality. A number of stochastic representations are given, disclosing its remarkable tractability. In particular, it is closed under marginalization, conditioning, subcomposition, amalgamation and permutation.

A structured Dirichlet mixture model for compositional data: inferential and applicative issues

The flexible Dirichlet (FD) distribution (Ongaro and Migliorati in J. Multvar. Anal. 114: 412–426, 2013) makes it possible to preserve many theoretical properties of the Dirichlet one, without inheriting its lack of flexibility in modeling the various independence concepts appropriate for compositional data, i.e. data representing vectors of proportions. In this paper we tackle the potential of the FD from an inferential and applicative viewpoint. In this regard, the key feature appears to be the special structure defining its Dirichlet mixture representation. This structure determines a simple and clearly interpretable differentiation among mixture components which can capture the main features of a large variety of data sets. Furthermore, it allows a substantially greater flexibility than the Dirichlet, including both unimodality and a varying number of modes. Very importantly, this increased flexibility is obtained without sharing many of the inferential difficulties typical of general mixtures. Indeed, the FD displays the identifiability and likelihood behavior proper to common (non-mixture) models. Moreover, thanks to a novel non random initialization based on the special FD mixture structure, an efficient and sound estimation procedure can be devised which suitably combines EM-types algorithms. Reliable complete-data likelihood-based estimators for standard errors can be provided as well.

NATURAL VARIABILITY OF BIOCHEMICAL BIOMARKERS IN THE MACRO-ZOOBENTHOS: DEPENDENCE ON LIFE STAGE AND ENVIRONMENTAL FACTORS

Abstract Biomarkers are widely used in ecotoxicology as indicators of exposure to toxicants. However, their ability to provide ecologically relevant information remains controversial. One of the major problems is understanding whether the measured responses are determined by stress factors or lie within the natural variability range. In a previous work, the natural variability of enzymatic levels in invertebrates sampled in pristine rivers was proven to be relevant across both space and time. In the present study, the experimental design was improved by considering different life stages of the selected taxa and by measuring more environmental parameters. The experimental design considered sampling sites in 2 different rivers, 8 sampling dates covering the whole seasonal cycle, 4 species from 3 different taxonomic groups (Plecoptera, Perla grandis; Ephemeroptera, Baetis alpinus and Epeorus alpicula; Tricoptera, Hydropsyche pellucidula), different life stages for each species, and 4 enzymes (acetylcholinesterase, glutathione S‐transferase, alkaline phosphatase, and catalase). Biomarker levels were related to environmental (physicochemical) parameters to verify any kind of dependence. Data were statistically elaborated using hierarchical multilevel Bayesian models. Natural variability was found to be relevant across both space and time. The results of the present study proved that care should be paid when interpreting biomarker results. Further research is needed to better understand the dependence of the natural variability on environmental parameters. Environ Toxicol Chem 2017;36:3158–3167.

Log-ratio approach in curve fitting for concentration-response experiments

The assessment of the ecological risk of chemical contamination by pollutants, pesticides or toxicants is of primary interest in environmental statistics. Concentration-response models play a fundamental role in computing the risk values connected with some exposure levels of a particular contaminant in living organisms. The present paper proposes a regression model called simplicial regression. This model is able to cope with the relative character of the explanatory and response parts via the logratio methodology of compositional data. Consequently, it allows performance of the corresponding statistical inference under the assumption of normality. Some real-world examples show that simplicial regression even outperforms the existing well-established methodologies on standard accuracy and quality-of-fit criteria. The better fit is due to the change of scale entailed by the new model.

A Dirichlet Mixture Model for Compositions Allowing for Dependence on the Size

The Dirichlet is the most well known distribution for compositional data, i.e. data representing vectors of proportions. The flexible Dirichlet distribution (FD) generalizes the Dirichlet one allowing to preserve its main mathematical and compositional properties. At the same time, it does not inherit its lack of flexibility in modeling the dependence concepts appropriate for compositional data. The present paper introduces a new model obtained by extending the basis of positive random variables generating the FD by normalization. Specifically, the new basis exhibits a more sophisticated mixture (latent) representation, which leads to a twofold result. On the one side, a more general distribution for compositional data, called EFD, is obtained by normalization. In particular, the EFD allows for a significantly wider differentiation among the clusters defining its mixture representation. On the other side, the generalized basis induces a tractable model for the dependence between composition and size: the conditional distribution of the composition given the size is still an EFD, the size affecting it in a simple fashion through the cluster weights.