Identifying diverse metal oxide nanomaterials with lethal effects on embryonic zebrafish using machine learning.

Manufacturers of nanomaterial-enabled products need models of endpoints that are relevant to human safety to support the "safe by design" paradigm and avoid late-stage attrition. Increasingly, embryonic zebrafish (Danio Rerio) are recognised as a key human safety relevant in vivo test system. Hence, machine learning models were developed for identifying metal oxide nanomaterials causing lethality to embryonic zebrafish up to 24 hours post-fertilisation, or excess lethality in the period of 24-120 hours post-fertilisation, at concentrations of 250 ppm or less. Models were developed using data from the Nanomaterial Biological-Interactions Knowledgebase for a dataset of 44 diverse, coated and uncoated metal or, in one case, metalloid oxide nanomaterials. Different modelling approaches were evaluated using nested cross-validation on this dataset. Models were initially developed for both lethality endpoints using multiple descriptors representing the composition of the core, shell and surface functional groups, as well as particle characteristics. However, interestingly, the 24 hours post-fertilisation data were found to be harder to predict, which could reflect different exposure routes. Hence, subsequent analysis focused on the prediction of excess lethality at 120 hours-post fertilisation. The use of two data augmentation approaches, applied for the first time in nano-QSAR research, was explored, yet both failed to boost predictive performance. Interestingly, it was found that comparable results to those originally obtained using multiple descriptors could be obtained using a model based upon a single, simple descriptor: the Pauling electronegativity of the metal atom. Since it is widely recognised that a variety of intrinsic and extrinsic nanomaterial characteristics contribute to their toxicological effects, this is a surprising finding. This may partly reflect the need to investigate more sophisticated descriptors in future studies. Future studies are also required to examine how robust these modelling results are on truly external data, which were not used to select the single descriptor model. This will require further laboratory work to generate comparable data to those studied herein.

Risk Management Framework for Nano-Biomaterials Used in Medical Devices and Advanced Therapy Medicinal Products.

The convergence of nanotechnology and biotechnology has led to substantial advancements in nano-biomaterials (NBMs) used in medical devices (MD) and advanced therapy medicinal products (ATMP). However, there are concerns that applications of NBMs for medical diagnostics, therapeutics and regenerative medicine could also pose health and/or environmental risks since the current understanding of their safety is incomplete. A scientific strategy is therefore needed to assess all risks emerging along the life cycles of these products. To address this need, an overarching risk management framework (RMF) for NBMs used in MD and ATMP is presented in this paper, as a result of a collaborative effort of a team of experts within the EU Project BIORIMA and with relevant inputs from external stakeholders. The framework, in line with current regulatory requirements, is designed according to state-of-the-art approaches to risk assessment and management of both nanomaterials and biomaterials. The collection/generation of data for NBMs safety assessment is based on innovative integrated approaches to testing and assessment (IATA). The framework can support stakeholders (e.g., manufacturers, regulators, consultants) in systematically assessing not only patient safety but also occupational (including healthcare workers) and environmental risks along the life cycle of MD and ATMP. The outputs of the framework enable the user to identify suitable safe(r)-by-design alternatives and/or risk management measures and to compare the risks of NBMs to their (clinical) benefits, based on efficacy, quality and cost criteria, in order to inform robust risk management decision-making.

Characterization of Nanoparticle Batch-To-Batch Variability.

A central challenge for the safe design of nanomaterials (NMs) is the inherent variability of NM properties, both as produced and as they interact with and evolve in, their surroundings. This has led to uncertainty in the literature regarding whether the biological and toxicological effects reported for NMs are related to specific NM properties themselves, or rather to the presence of impurities or physical effects such as agglomeration of particles. Thus, there is a strong need for systematic evaluation of the synthesis and processing parameters that lead to potential variability of different NM batches and the reproducible production of commonly utilized NMs. The work described here represents over three years of effort across 14 European laboratories to assess the reproducibility of nanoparticle properties produced by the same and modified synthesis routes for four of the OECD priority NMs (silica dioxide, zinc oxide, cerium dioxide and titanium dioxide) as well as amine-modified polystyrene NMs, which are frequently employed as positive controls for nanotoxicity studies. For 46 different batches of the selected NMs, all physicochemical descriptors as prioritized by the OECD have been fully characterized. The study represents the most complete assessment of NMs batch-to-batch variability performed to date and provides numerous important insights into the potential sources of variability of NMs and how these might be reduced.

Literature Review of (Q)SAR Modelling of Nanomaterial Toxicity.

Despite the clear benefits that nanotechnology can bring to various sectors of industry, there are serious concerns about the potential health risks associated with engineered nanomaterials (ENMs), intensified by the limited understanding of what makes ENMs toxic and how to make them safe. As the use of ENMs for commercial purposes and the number of workers/end-users being exposed to these materials on a daily basis increases, the need for assessing the potential adverse effects of multifarious ENMs in a time- and cost-effective manner becomes more apparent. One strategy to alleviate the problem of testing a large number and variety of ENMs in terms of their toxicological properties is through the development of computational models that decode the relationships between the physicochemical features of ENMs and their toxicity. Such data-driven models can be used for hazard screening, early identification of potentially harmful ENMs and the toxicity-governing physicochemical properties, and accelerating the decision-making process by maximising the use of existing data. Moreover, these models can also support industrial, regulatory and public needs for designing inherently safer ENMs. This chapter is mainly concerned with the investigation of the applicability of (quantitative) structure-activity relationship ((Q)SAR) methods to modelling of ENMs' toxicity. It summarizes the key components required for successful application of data-driven toxicity prediction techniques to ENMs, the published studies in this field and the current limitations of this approach.

Accurate and interpretable nanoSAR models from genetic programming-based decision tree construction approaches.

The number of engineered nanomaterials (ENMs) being exploited commercially is growing rapidly, due to the novel properties they exhibit. Clearly, it is important to understand and minimize any risks to health or the environment posed by the presence of ENMs. Data-driven models that decode the relationships between the biological activities of ENMs and their physicochemical characteristics provide an attractive means of maximizing the value of scarce and expensive experimental data. Although such structure-activity relationship (SAR) methods have become very useful tools for modelling nanotoxicity endpoints (nanoSAR), they have limited robustness and predictivity and, most importantly, interpretation of the models they generate is often very difficult. New computational modelling tools or new ways of using existing tools are required to model the relatively sparse and sometimes lower quality data on the biological effects of ENMs. The most commonly used SAR modelling methods work best with large datasets, are not particularly good at feature selection, can be relatively opaque to interpretation, and may not account for nonlinearity in the structure-property relationships. To overcome these limitations, we describe the application of a novel algorithm, a genetic programming-based decision tree construction tool (GPTree) to nanoSAR modelling. We demonstrate the use of GPTree in the construction of accurate and interpretable nanoSAR models by applying it to four diverse literature datasets. We describe the algorithm and compare model results across the four studies. We show that GPTree generates models with accuracies equivalent to or superior to those of prior modelling studies on the same datasets. GPTree is a robust, automatic method for generation of accurate nanoSAR models with important advantages that it works with small datasets, automatically selects descriptors, and provides significantly improved interpretability of models.

Nano(Q)SAR: Challenges, pitfalls and perspectives.

Regulation for nanomaterials is urgently needed, and the drive to adopt an intelligent testing strategy is evident. Such a strategy will not only provide economic benefits but will also reduce moral and ethical concerns arising from animal testing. For regulatory purposes, such an approach is promoted by REACH, particularly the use of quantitative structure-activity relationships [(Q)SAR] as a tool for the categorisation of compounds according to their physicochemical and toxicological properties. In addition to compounds, (Q)SAR has also been applied to nanomaterials in the form of nano(Q)SAR. Although (Q)SAR in chemicals is well established, nano(Q)SAR is still in early stages of development and its successful uptake is far from reality. This article aims to identify some of the pitfalls and challenges associated with nano-(Q)SARs in relation to the categorisation of nanomaterials. Our findings show clear gaps in the research framework that must be addressed if we are to have reliable predictions from such models. Three major barriers were identified: the need to improve quality of experimental data in which the models are developed from, the need to have practical guidelines for the development of the nano(Q)SAR models and the need to standardise and harmonise activities for the purpose of regulation. Of these three, the first, i.e. the need to improve data quality requires immediate attention, as it underpins activities associated with the latter two. It should be noted that the usefulness of data in the context of nano-(Q)SAR modelling is not only about the quantity of data but also about the quality, consistency and accessibility of those data.