QSAR modeling of nanomaterials

Abstract

A thorough understanding of the relationships between the physicochemical properties and the behavior of nanomaterials in biological systems is mandatory for designing safe and efficacious nanomedicines. Quantitative structure-activity relationship (QSAR) methods help to establish such relationships, although their application to model the behavior of nanomaterials requires new ideas and applications to account for the novel properties of this class of compounds. This review presents and discusses a number of recent inspiring applications of QSAR modeling and descriptors for nanomaterials with a focus on approaches that attempt to describe the interactions that take place at the nano/bio-interface. The paradigm shift from classic to nano-QSAR currently relies on both theoretically and experimentally derived descriptors, and the solutions adopted for modeling are diverse, mirroring the structural and behavioral heterogeneity of nanomaterials. Research should focus on both aspects of a QSAR study: the generation of nanospecific theoretical descriptors and experimental test data.

(Q)SAR modelling of nanomaterial toxicity: A critical review

“…A set of 30 molecular descriptors were calculated for each NPs and only two of them, the conduction band energy and ionic index, were identified as the key molecular descriptors on which the best performing nano-(Q)SAR model was built. Their conclusion was in a good agreement with the results of previous researchers (Burello & Worth, 2011) who stated that the conduction band energy of oxide NPs is related to their toxicity. Similar findings have also been reported by (Zhang, et al, 2012) who indicated that the oxidative stress induced by MO-NPs could be linked to their conduction and valance band energies.…”

“…As the properties of nanoscale materials are remarkably different from conventional ones, it is very likely that the toxicity of ENMs is also associated with different features (Burello & Worth, 2011). Therefore, the development of nano-specific descriptors with the capability to describe the distinctive properties of NPs is one of the major research needs in the area of computational nanotoxicology.…”

“…The idea suggested by (Glotzer & Solomon, 2007) has inspired other researchers such as (T. ( Burello & Worth, 2011) proposed that different types of spectra (e.g. NMR, IR, Raman, UVVis) can be used as nano-descriptors since they contain fingerprint-like information (Fig.…”

Frameworks and tools for risk assessment of manufactured nanomaterials

“…Quantitative Nano-structure Activity Relationships (QNAR) (Burello and Worth, 2011a, Burello and Worth, 2011b, Puzyn et al, 2011, Puzyn et al, 2009a, Puzyn et al, 2009b, Toropov et al, 2007, Toropov and Leszczynski, 2006, Liu et al, 2013b, Liu et al, 2014, Liu et al, 2013a, Gómez et al, 2013 In silico tools for hazard assessment (Liu et al, 2014, Liu et al, 2013a, Liu et al, 2013b In silico tools for hazard assessment Some of these tools are capable of assessing uncertainties. The Precautionary Matrix for Synthetic Nanomaterials uses a "specific framework conditions" criterion that represents uncertainties resulting from knowledge gaps with respect to the origin of the MNs, their characteristics and uses.…”

“…Such ITS would combine testing and non-testing methods to generate data for RA. In this context the development and application of in silico tools has become prominent and a number of such tools have been proposed (Burello and Worth, 2011a, Burello and Worth, 2011b, Puzyn et al, 2011, Puzyn et al, 2009a, KAR et al, 2014, Richarz et al, 2015, Toropova and Toropov, 2015. Statistical analysis and machine learning methods (e.g.…”