Making reliable negative predictions of human skin sensitisation using an in silico fragmentation approach

ABSTRACT A previously published fragmentation method for making reliable negative in silico predictions has been applied to the problem of predicting skin sensitisation in humans, making use of a dataset of over 2750 chemicals with publicly available skin sensitisation data from 18 in vivo assays. An assay hierarchy was designed to enable the classification of chemicals within this dataset as either sensitisers or non‐sensitisers where data from more than one in vivo test was available. The negative prediction approach was validated internally, using a 5‐fold cross‐validation, and externally, against a proprietary dataset of approximately 1000 chemicals with in vivo reference data shared by members of the pharmaceutical, nutritional, and personal care industries. The negative predictivity for this proprietary dataset was high in all cases (>75%), and the model was also able to identify structural features that resulted in a lower accuracy or a higher uncertainty in the negative prediction, termed misclassified and unclassified features respectively. These features could serve as an aid for further expert assessment of the negative in silico prediction. Graphical abstract Figure. No Caption available. HighlightsReliable negative in silico predictions of human skin sensitisation can be made.A fragmentation approach is used to assess similarity to known false negatives.The negative predictivity was >75% when tested against a large proprietary dataset.Structural features of concern are highlighted for further expert assessment.

[1]  Amparo Nácher,et al.  Alternative Methods to Animal Testing in Safety Evaluation of Cosmetic Products , 2018 .

[2]  Thorsten Meinl,et al.  KNIME - the Konstanz information miner: version 2.0 and beyond , 2009, SKDD.

[3]  I Gerner,et al.  Chemical substances and contact allergy--244 substances ranked according to allergenic potency. , 2003, Toxicology.

[4]  LaquiezeLeslie,et al.  Alternative Methods to Animal Testing and Cosmetic Safety: An Update on Regulations and Ethical Considerations in Brazil , 2015 .

[5]  Timothy F. Malloy,et al.  Policy reforms to update chemical safety testing , 2017, Science.

[6]  Steven J Canipa,et al.  Predicting skin sensitisation using a decision tree integrated testing strategy with an in silico model and in chemico/in vitro assays. , 2016, Regulatory toxicology and pharmacology : RTP.

[7]  João Barroso,et al.  Categorization of Chemicals According to Their Relative Human Skin Sensitizing Potency , 2014, Dermatitis : contact, atopic, occupational, drug.

[8]  Carol A Marchant,et al.  Structure–activity relationships for skin sensitization: recent improvements to Derek for Windows , 2006, Contact dermatitis.

[9]  Nigel Greene,et al.  It's difficult, but important, to make negative predictions. , 2016, Regulatory toxicology and pharmacology : RTP.

[10]  B F Goodwin,et al.  A comparison of three guinea‐pig sensitization procedures for the detection of 19 reported human contact sensitizers , 1981, Contact dermatitis.

[11]  M. Barratt,et al.  Development of an expert system rulebase for identifying contact allergens. , 1994, Toxicology in vitro : an international journal published in association with BIBRA.

[12]  Thomas Steger-Hartmann,et al.  A quantitative in silico model for predicting skin sensitization using a nearest neighbours approach within expert‐derived structure–activity alert spaces , 2017, Journal of applied toxicology : JAT.

[13]  Anne Marie Api,et al.  Fragrances Categorized According to Relative Human Skin Sensitization Potency , 2017, Dermatitis.

[14]  Paola Gramatica,et al.  Principles of QSAR models validation: internal and external , 2007 .

[15]  Lidiya Stavitskaya,et al.  Establishing best practise in the application of expert review of mutagenicity under ICH M7. , 2015, Regulatory toxicology and pharmacology : RTP.

[16]  Ocspp,et al.  The Frank R. Lautenberg Chemical Safety for the 21st Century Act , 2016 .

[17]  G M Rusch,et al.  Variability of a mouse ear swelling test (MEST) in predicting weak and moderate contact sensitization. , 1990, Fundamental and applied toxicology : official journal of the Society of Toxicology.

[18]  Philip N. Judson,et al.  Assessing confidence in predictions made by knowledge-based systems , 2013 .

[19]  I Kimber,et al.  Skin sensitization , 2015, Human & experimental toxicology.

[20]  Thorsten Meinl,et al.  KNIME: The Konstanz Information Miner , 2007, GfKl.

[21]  Samuel J. Webb,et al.  Self organising hypothesis networks: a new approach for representing and structuring SAR knowledge , 2014, Journal of Cheminformatics.

[22]  I Kimber,et al.  Skin sensitization risk assessment: a comparative evaluation of 3 isothiazolinone biocides , 1999, Contact dermatitis.