Facial reconstruction

Search LJMU Research Online

Browse Repository | Browse E-Theses

Investigation of in Silico Modelling to Predict the Human Health Effects of Cosmetics Ingredients

Aljallal, M (2020) Investigation of in Silico Modelling to Predict the Human Health Effects of Cosmetics Ingredients. Doctoral thesis, Liverpool John Moores University.

[img]
Preview
Text
2019 aljallal PhD.pdf - Published Version

Download (3MB) | Preview

Abstract

Animal experiments have been the standard method to assess the safety of chemicals used in cosmetic products for decades. However, public opinion has continued to demand that in vivo hazard identification methods conducted on animals are replaced with alternative methods. Research on alternative methods to replace in vivo toxicity testing continually increased over the past few decades with different alternatives developed, such as in vitro, in chemico and in silico approaches. Although different alternative techniques can be employed, no single technique can solely replace the complexity and an in vivo test, especially for chronic effects. Therefore, integrated testing strategies that can utilise the information from all available alternative testing approaches have been developed. Within the Adverse Outcome Pathway (AOP) paradigm, the molecular initiating event(s) MIE can be induced by several chemical key features which can be captured by structural alerts. When structural alerts for a MIE are compiled and supported by mechanistic and toxicity information confirming the induction of the same MIE, then they can be considered as an in silico profiler. The overall aim of the work presented in this thesis was to assess the current in silico profilers for carcinogenicity (both genotoxic and non-genotoxic), mutagenicity and skin sensitisation through assessment using multiple high-quality experimental databases. The research presented herein demonstrates the ability to assess the positive predictivity of two types of structural alert, mechanism- and chemistry-based that pertain to the endpoints and proposes ways to improve the overall accuracy of these profilers. In this context, this study has given an insight to those alerts that may be found equally in endpoint-positive or negative compounds, and those which may be more effectively utilised to form groups of analogues for read across predictions. A detailed analysis of positive predictivity of the available mutagenicity, carcinogenicity and skin sensitisation structural alerts and profilers Page 3 within the OECD QSAR Toolbox against experimental data is presented. This investigation showed the structural alerts that are accurate as such, and those that may need further refinement, or their use may need to be reconsidered. In addition, the relationship between scaffolds of a range of diverse compounds and carcinogenicity showed that a total of 17 carcinogenicity scaffolds could be identified from the available databases and could be used as a base for an in silico profiler. This work has also determined the need for further in-depth research in this area to study the suitability and merits of each of the alerts within the profilers currently included in the OECD QSAR Toolbox, and other in silico toxicity platforms, to identify the possibilities for improvement in their performance. This will, by implication, also improve the reliability of chemical read-across and grouping/categorisation for classification, labelling and risk assessment for regulatory use of the in silico methods.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: QSAR; Cosmetic; Carcinogenicity; Structure activity relationship; Mutagenicity; Molecular Descriptor; Structural alerts
Subjects: R Medicine > RM Therapeutics. Pharmacology
R Medicine > RS Pharmacy and materia medica
Divisions: Pharmacy & Biomolecular Sciences
Date Deposited: 31 Jan 2020 10:31
Last Modified: 31 Jan 2020 10:32
DOI or Identification number: 10.24377/LJMU.t.00012139
Supervisors: Cronin, M, Madden, J, Enoch, S and Chaudhry, Q
URI: http://researchonline.ljmu.ac.uk/id/eprint/12139

Actions (login required)

View Item View Item