Pesticide Indirect Photodegradation Database: A Data-Sharing Platform for Screening Existing and Discovering New Agrochemicals

Abstract

The ability to control and tune physicochemical properties that underscore chemical behavior in living systems and the environment is at the "heart" of green chemistry. This is especially true for chemical classes designed a priori to be biologically active, such as pesticides, where the chance of unintended adverse outcomes is high. We recently proposed a design-vectoring framework, leveraging validated computational models of ecotoxicity and indirect photodegradation as a useful, quasisystems-based tool for screening existing and designing new agrochemicals. Here, we describe the development of a database that integrates our models, which link structural and substructural features to process metrics, and corresponding predicated data for all agrochemicals with photodegradable cores on the U.S. Environmental Protection Agency's registry (785 compounds and over 18,000 pairwise interactions with chromophoric dissolved organic matter, CDOM). The database is searchable by structural and nonstructural identifiers (e.g., chemical class, oxidizable core, physicochemical and electronic properties, etc.) to aid in chemical selection, hazard, and alternative assessment. Crucially, it can be easily updated and augmented to aid in interactive datasharing across industry, government, and academia. The overarching goal of this project is to spur grander efforts in systems-based design of pesticides that would see this platform paired with target-based computational methods and incorporated into the discovery phase of new product development across industry sectors.