Celebrating 40 Years of Career

Abstract

Nowadays in silico approaches based on QSAR (Quantitative Structure Activity Relationships) are core elements of Intelligent Testing Strategies (ITS). These methods are based on the combination of multiple data streams, which include data generated through the application of alternatives to animal testing such as QSAR models and related predictions. In Europe their application represents a main innovation introduced by REACH when compared to former regulations. The approval and publication of the OECD Principles for the regulatory use of QSAR, represented a milestone in the worldwide application of in silico strategies for the modelling and the prediction of properties and activities for traditional and industrial chemicals. The current use of computational chemistry and QSARs is already providing key elements in the determination of missing data, which can serve to support and integrate ITSbased frameworks. Furthermore it allows to focus experiments thereby limiting costs in terms of time, money and animal lives, to plan the synthesis of new chemicals, avoiding the creation, marketing and use of substances with effects adverse to human health and the environment (safe by design approach), or to find possible alternatives to known or potentially hazardous compounds. Regulatory and scientific communities are making a big effort to maximize the use of in silico techniques calling for fast, efficient, and transparent modelling tools, which are expected to support the exponential growth of information made available through public databases as well as highthroughput and -omics techniques. Machine learning methods based on linear and nonlinear approaches as well as data mining techniques have been widely applied to this end. Nowadays thousands of QSARs are available to predict different endpoints of (eco)toxicological and regulatory interest. The increased availability of complex data poses new challenges, which involve big data handling and modelling, and call for an increase in the expertise of developers and users, including regulators, to cope with highly complex information and computational approaches. On the other hand these new challenges push computational scientists to deliver better results for instance by addressing data uncertainty through data curation and multi modelling approaches. This special issue celebrates the career and contribution to science and to the field of QSAR of Prof. Paola Gramatica who retired in march 2018, after more than 40 years of dedicated hard work in organic and computational chemistry. From her early career at the University of Milan (Italy) until 1994, Prof. Gramatica carried out research on the synthesis, purification, structural elucidation, and biotransformation of organic natural products in the Department of Organic and Industrial Chemistry. In the late ‘90s she