Predictive model for aerobic biodegradability developed from a file of evaluated biodegradation data

Abstract

A file of evaluated biodegradation data was used to develop a model for predicting aerobic biodegradability from chemical substructures. Chemicals initially were divided into three groups: (a) chemicals that degrade rapidly under most environmental conditions without requiring acclimation; (b) chemicals that are biodegradable, but only after an acclimation period; and (c) chemicals that degrade slowly or not at all. Chemicals in the first and last groups were then used to develop a model for classifying chemicals as rapidly biodegradable or not rapidly biodegradable. Two models, one based on linear regression and the other based on nonlinear regression, were developed from 35 preselected substructures. Both models predict the biodegradation categories correctly 90% of the time for the training set (264 chemicals) and an independent validation set (27 chemicals), and are 95% correct for chemicals that biodegrade rapidly.

A new biodegradation prediction model specific to petroleum hydrocarbons

“…Table 3 lists the fragments that ultimately were chosen as well as their coefficient value compared with applicable fragments from the BIOWIN linear [16] and MITI linear [15] models. In addition, 95% confidence intervals are provided for each BIOHCWIN fragment.…”

“…These models provide an estimate for the probability of biodegradation based on fragment methodology (BIOWIN linear and nonlinear models, primary and ultimate survey models, and MTTI linear and nonlinear models). The Ministry of International Trade and Industry (MITI) models were developed using Organisation for Economic Co‐operation and Development ready screening test data [15], whereas the BIOWIN linear and nonlinear models [16] are based on experimental biodegradation data collected from BIODEG, a file within Syracuse Research Corporation's Environmental Fate DataBase (EFDB) [17]. The number of fragments from the BIOWIN linear and nonlinear models applicable to the modeling of petroleum hydrocarbons is limited to five, whereas the MITI model contains only 12 fragments that would be useful in the modeling of petroleum hydrocarbons.…”

Predicting ready biodegradability in the Japanese ministry of international trade and industry test

“…We responded to this need in the early 1990s by first developing a weight‐of‐evidence procedure [1] for collecting and evaluating available data and then using these summary evaluations to develop two models for predicting aerobic biode‐gradability from chemical substructures [2,3]. The data and evaluations were made available in BIODEG, a component of the Environmental Fate Data Base [4,5] (http://esc-plaza.syrres.com/efdb.htm, Syracuse Research Corporation, North Syracuse, NY, USA) that now contains information on more than 1,000 discrete organics.…”

From laboratory to environmental conditions: a new approach for chemical’s biodegradability assessment

“…The stakes are high; in addition to the substance-specific properties, the fate of each chemical in the environment differs according to its terms of use and mode of discharge (Bartholomew and Pfaender 1983;Federle et al 1997;Vink and Van Der Zee 1997). In silico models have been developed (Raymond et al 2001) and are available under REACH regulations (Pizzo et al 2013), but the prediction of overall persistence with multimedia fate models remains limited to existing data and may be not adapted to assess environmental persistence as a function of both chemical intrinsic properties and environmental conditions (Howard et al 1992;Fenner et al 2004;Aronson et al 2006). Indeed, the various environmental microbial communities, which are potentially involved in the biodegradation of chemicals, are different from each other (Forney et al 2001;Martiny et al 2006) and are also subject to the variation of environmental parameters associated with the conditions of in situ life (Zogg et al 1997;Ranjard et al 2013).…”