Finding Potential Replacements for TRI Solvents Using the Environmental Impact of the Average Solvent.

A solvent has many different types of impact on the environment. This article describes a method that combines several different types of impacts together into one environmental index so that similar solvents may be compared by their cumulative impact to the environment. The software tool PARIS III (Program for Assisting the Replacement of Industrial Solvents III) initially finds thousands of solvents mixtures with behaviors as close as possible to those of the original solvent entered. The overall environmental impacts of these solvent mixtures are estimated and assigned to environmental indexes. Users of the software tool can then choose replacements for the original solvent with similar activities but with significantly smaller environmental indexes. These solvent mixtures may act as practical substitutes for the industrial solvents but substantially reduce the overall environmental impact of the original harmful solvents. Potential replacements like this are found for three of the U.S. Environmental Protection Agency's Toxic Release Inventory solvents, carbon tetrachloride, toluene, and N-methylpyrrolidone.

An EPA database on the effects of engineered nanomaterials-NaKnowBase.

The US EPA Office of Research and Development (ORD) has conducted a research program assessing potential risks of emerging materials and technologies, including engineered nanomaterials (ENM). As a component of that program, a nanomaterial knowledge base, termed "NaKnowBase", was developed containing the results of published ORD research relevant to the potential environmental and biological actions of ENM. The experimental data address issues such as ENM release into the environment; fate, transport and transformations in environmental media; exposure to ecological species or humans; and the potential for effects on those species. The database captures information on the physicochemical properties of ENM tested, assays performed and their parameters, and the results obtained. NaKnowBase (NKB) is a relational SQL database, and may be queried either with SQL code or through a user-friendly web interface. Filtered results may be output in spreadsheet format for subsequent user-defined analyses. Potential uses of the data might include input to quantitative structure-activity relationships (QSAR), meta-analyses, or other investigative approaches.

The software tool to find greener solvent replacements, PARIS III.

PARIS III (Program for Assisting the Replacement of Industrial Solvents III, Version 1.4.0) is a pollution prevention solvent substitution software tool used to find mixtures of solvents that are less harmful to the environment than the industrial solvents to be replaced. By searching extensively though hundreds of millions of possible solvent combinations, mixtures that perform the same as the original solvents may be found. Greener solvent substitutes may then be chosen from those mixtures that behave similarly but have less environmental impact. These extensive searches may be enhanced by fine-tuning impact weighting factors to better reflect regional environmental concerns; and by adjusting how close the properties of the replacement must be to those of the original solvent. Optimal replacements can then be compared again and selected for better performance, but less environmental impact. This method can be a very effective way of finding greener replacements for harmful solvents used by industry.

Does rational selection of training and test sets improve the outcome of QSAR modeling?

Prior to using a quantitative structure activity relationship (QSAR) model for external predictions, its predictive power should be established and validated. In the absence of a true external data set, the best way to validate the predictive ability of a model is to perform its statistical external validation. In statistical external validation, the overall data set is divided into training and test sets. Commonly, this splitting is performed using random division. Rational splitting methods can divide data sets into training and test sets in an intelligent fashion. The purpose of this study was to determine whether rational division methods lead to more predictive models compared to random division. A special data splitting procedure was used to facilitate the comparison between random and rational division methods. For each toxicity end point, the overall data set was divided into a modeling set (80% of the overall set) and an external evaluation set (20% of the overall set) using random division. The modeling set was then subdivided into a training set (80% of the modeling set) and a test set (20% of the modeling set) using rational division methods and by using random division. The Kennard-Stone, minimal test set dissimilarity, and sphere exclusion algorithms were used as the rational division methods. The hierarchical clustering, random forest, and k-nearest neighbor (kNN) methods were used to develop QSAR models based on the training sets. For kNN QSAR, multiple training and test sets were generated, and multiple QSAR models were built. The results of this study indicate that models based on rational division methods generate better statistical results for the test sets than models based on random division, but the predictive power of both types of models are comparable.

A hierarchical clustering methodology for the estimation of toxicity.

ABSTRACT A quantitative structure-activity relationship (QSAR) methodology based on hierarchical clustering was developed to predict toxicological endpoints. This methodology utilizes Ward's method to divide a training set into a series of structurally similar clusters. The structural similarity is defined in terms of 2-D physicochemical descriptors (such as connectivity and E-state indices). A genetic algorithm-based technique is used to generate statistically valid QSAR models for each cluster (using the pool of descriptors described above). The toxicity for a given query compound is estimated using the weighted average of the predictions from the closest cluster from each step in the hierarchical clustering assuming that the compound is within the domain of applicability of the cluster. The hierarchical clustering methodology was tested using a Tetrahymena pyriformis acute toxicity data set containing 644 chemicals in the training set and with two prediction sets containing 339 and 110 chemicals. The results from the hierarchical clustering methodology were compared to the results from several different QSAR methodologies.