<html> <head> </head> <body> QSAR for eye irritation (Draize test) </body> </html> <html> <head> </head> <body> </body> </html> <html> <head> </head> <body> 30.01.2009 </body> </html> <html> <head> </head> <body> </body> </html> <html> <head> </head> <body> </body> </html> <html> <head> </head> <body> </body> </html> <html> <head> </head> <body> Model is proprietary, but the training and test sets are available. </body> </html> <html> <head> </head> <body> None to date. </body> </html> <html> <head> </head> <body> Rabbit </body> </html> <html> <head> </head> <body> Modified maximum average score (MMAS) derived from Draize rabbit eye test scores. </body> </html> <html> <head> </head> <body> Modified maximum average score (MMAS) divided by the molarity of the pure liquid. </body> </html> <html> <head> </head> <body> log (MMAS/P°) and log (1/EIT) instead of MMAS/P° and EIT (logarithm of the Draize test scores adjusted by the liquid saturated vapor-pressure) </body> </html> <html> <head> </head> <body> Draize rabbit eye test. The in vivo rabbit eye irritation/corrosion data have been generated since 1981 in studies carried out according to OECD Test Guideline 405 (EU Test Method B.5) and following the principles of Good Laboratory Practice. In the Draize rabbit eye test (Draize et al., 1944), a 0.1ml (or weight equivalent) sample of test substance is placed into the eye. Eye irritation is defined as the production of changes in the eye that are fully reversible within 21 days of application, whereas eye corrosion is defined as production of tissue damage in the eye, or serious physical decay of vision, which is not fully reversible within 21 days of application. The tissue grades are combined into a weighted score; the highest average score across test animals is termed the maximum average score (MAS). The modified Draize scores were defined as modified maximum average score (MMAS) divided by the molarity of the pure liquid; the latter is given by 1000 times the density of the pure liquid divided by the liquid molecular weight. The MMAS refer to the effect of pure bulk liquids, whereas the EIT (in ppm) are established from the effect of the vapour of liquids at some particular partial pressure. </body> </html> <html> <head> </head> <body> The MMAS data were selected from European Centre for Ecotoxicology and Toxicology of Chemicals databank (ECETOC, 1998). MMAS scores for 68 pure bulk liquids were adjusted by the liquid-saturated vapor pressure P°. These 68 adjusted scores, as log (MMAS/P°), were shown to be equivalent to eye irritation thresholds (EIT), expressed as log (1/EIT), for 23 compounds in humans (Abraham et al, 2003). The EIT data were selected from Cometto-Muñiz, et al (2003). The Draize test scores and EIT can be compared as: (log(MMAS/P°)=log(1/EIT) +m’’). Statistics (the experimental results were obtained using Draize test scores for 68 compounds): max value: 2.37 min value: -5.24 standard deviation: 1.538 skewness: 0.886 </body> </html> <html> <head> </head> <body> QSAR </body> </html> <html> <head> </head> <body> log(MMAS/P0) = 0.005 * Gravitation index (all bonds) (AM1) + 6.816 * HASA-1/TMSA (AM1) - 3.586 * Lowest e-e repulsion (1-center) (AM1) - 30.864* Max nucleophilic reactivity index (AM1) for C atoms + 2.822 </body> </html> <html> <head> </head> <body> Initial pool of ~1000 descriptors. Stepwise descriptor selection based on a set of statistical selection rules: 1-parameter equations: Fisher criterion and R2 over threshold, variance and t-test value over threshold, intercorrelation with another descriptor not over threshold 2 parameter equations: intercorrelation coefficient below threshold, significant correlation with endpoint, in terms of correlation coefficient and t-test. Stepwise trial of additional descriptors not significantly correlated to any already in the model. </body> </html> <html> <head> </head> <body> 1D, 2D, and 3D theoretical calculations. Quantum chemical descriptors derived from Merck Molecular Force Field (MMFFs) (vacuum) AM1 calculation. Model developed by using multilinear regression. </body> </html> <html> <head> </head> <body> 18 (72 chemicals/4 descriptors) </body> </html> <html> <head> </head> <body> Applicability domain based on training set: a) by chemical identity: organic liquids (diverse set of aromatic, cyclic and aliphatic alcohols, esters, halogen compounds, ketones). b) by descriptor value range: the model is suitable for compounds that have the descriptors in the following ranges: Gravitation index (all bonds)(AM1): min 213.606, max 2497.675 HASA- 1/TMSA(AM1): min 0, max 0.318 Lowest e-e repulsion (1-center) (AM1): min 1.296, max 3.543) Max nucleophilic reactivity index (AM1) for C atoms: min 0.002, max 0.052) </body> </html> <html> <head> </head> <body> Presence of functional groups in structures Range of descriptor values in training set with ±30% confidence Descriptor values must fall between maximal and minimal descriptor values of training set ±30%. </body> </html> <html> <head> </head> <body> See 5.1 </body> </html> <html> <head> </head> <body> 72 data points: 69 negative values, 3 positive values </body> </html> <html> <head> </head> <body> </body> </html> <html> <head> </head> <body> R2= 0.893 (Correlation coefficient); S2= 0.267 (Standard error of the estimate); F= 140.539 (Fisher statistics) </body> </html> <html> <head> </head> <body> R2cv= 0.877 </body> </html> <html> <head> </head> <body> R2cv= 0.874 </body> </html> <html> <head> </head> <body> ABC analysis (2:1 training : prediction) on sorted data (in increasing order of endpoint value) divided into 3 subsets (A;B;C). Training set formed with 2/3 of the compounds (set A+B, A+C, B+C) and validation set consisted of 1/3 of the compounds (C, B, A) Average R2 (fitting) = 0.899 Average R2 (prediction) = 0.860 </body> </html> <html> <head> </head> <body> 8 data points: 7 negative values, 1 positive value </body> </html> <html> <head> </head> <body> The experimental dataset was sorted according to increasing values of the endpoint value and each tenth compound was assigned to the test set. </body> </html> <html> <head> </head> <body> R2 = 0.802 </body> </html> <html> <head> </head> <body> The descriptor values of the test set are within the limits of applicability. </body> </html> <html> <head> </head> <body> </body> </html> <html> <head> </head> <body> According to the model equation, eye irritation depends on the hydrogen bond donor and acceptor capabilities of a liquid as well as on the overall shape and bulkiness of the molecules. The key issue is the transport from the eye surface into the biophase, binding to the phospholipid membrane and possible binding to the receptor. </body> </html> <html> <head> </head> <body> A posteriori mechanistic interpretation, consistent with published scientific interpretations of experimental data. </body> </html> <html> <head> </head> <body> The descriptor HASA-1/TMSA (AM1) reflects transfer of the compounds to a phase characterized by hydrogen bonding whereas the Lowest e-e repulsion (1-center) (AM1) for C atoms reflects the transfer of the compounds to a phase that is quite polar and hydrophobic. The proposed mechanism based on the model agrees well with literature (Abraham et al, 2003). </body> </html> <html> <head> </head> <body> </body> </html> Q2-22-1-135 2009/12/10 eye irritation, Draize eye test, MMAS, Molcode