Lazar Toxicity Predictions

EPA Fathead Minnow Acute Toxicity (EPAFHM) - 96 hr LC50

• Endpoint Definition
• Algorithm
• Applicability Domain
• Predictivity
• Mechanistic Interpretation

Endpoint Definition

96 hr LC50 (concentration producing lethality in 50% of test animals after 96 hours exposure) in mg/l. Calculated using Spearman-Karber method.
Geometric mean of LC50s presented if more than one bioassay conducted for the chemical
Conversion of LC50_mg to mmol units: LC50_mmol = LC50_mg / STRUCTURE_MolecularWeight. “blank” or null entry indicates no mortality, or less than 50% mortality observed at 96hr. [ Details ] [ Original data ]

Algorithm Definition

lazar obtains predictions from the experimental results of compounds with similar structures ( neighbors ). For differentiated predictions chemical similarities are always determined in respect to the endpoint under investigation . A detailled description and formal definition of the lazar regression algorithm has been published in:

• A. Maunz and C. Helma: Prediction of chemical toxicity with local support vector regression and activity-specific kernels, SAR QSAR Environ. Res., in press [preprint]

You can donwload the source code for lazar ( GNU General Public License ) with git :
git clone git://github.com/helma/lazar-core.git

Applicability Domain Definition

The applicability domain (AD) of the training set is characterized by the confidence index of a prediction (high confidence index: close to the applicability domain of the training set/reliable prediction, low confidence: far from the applicability domain of the trainingset/unreliable prediction). The confidence index considers (i) the similarity and number of neighbors and (ii) contradictory examples within the neighbors. A formal definition can be found in:

• A. Maunz and C. Helma: Prediction of chemical toxicity with local support vector regression and activity-specific kernels, SAR QSAR Environ. Res., in press

The reliability of predictions decreases gradually with increasing distance from the applicability domain (i.e. decreasing confidence index). Figure 1 shows this dependency visually.

For simplicity we provide also results for an applicability domain definition with a sharp border at a confidence index of 0.2 . These results are summarized in Table 1 and indicated by the grey area in Figure 1 and by the black triangles in Figure 2 . Misclassifications within the applicability domain are summarized in the table of misclassifications .

The presence of substructures that are unknown to the training set ( unknown fragments ) is another factor that limits the applicability domain. As the training data cannot provide any information about unknown fragments, their relevance has to be evaluated by an expert user (as a rule of thumb large fragments are of less concern, because all shorter subfragments have been evaluated by the system). For this reason the presence/absence of unknown fragments is not considered in the formal applicability domain definition, but their presence is indicated in the table of misclassifications .

Validation Results (leave-one-out crossvalidation)

Definition and experimental comparison with external validation procedures:

• R. Benigni, T. I. Netzeva, E. Benfenati, C. Bossa andR. Franke, C. Helma, E. Hulzebos, C. Marchant, A. Richard, Y.-T. Woo, and C. Yang. The expanding role of predictive toxicology: an update on the (Q)SAR models for mutagens and carcinogens. J Environ Sci Health C Environ Carcinog Ecotoxicol Rev., 25:53-97, 2007.
• C. Helma: Lazy Structure-Activity Relationships (lazar) for the Prediction of Rodent Carcinogenicity and Salmonella Mutagenicity, Molecular Diversity 10, 147-158 (2006) [ preprint ]
• C. Helma and J. Kazius: Artificial Intelligence and Data Mining for Toxicity Prediction, Current Computer-Aided Drug Design 2, 1-19 (2006) [ preprint ]
• Presentation at Workshop on Evaluating Prediction Models in Mutagenicity and Carcinogenicity, Rome, Italy (2006) [ presentation ]

Table 1: Predictions within applicability domain

r 2	0.75
RMSE	0.71
Mean error	0.51
Predictions within 1 log unit	98 from 110 ( 89.1 %)

Predictions with a confidence > 0.2 are considered to be within the applicability domain (see Applicability Domain Definition )

Table 2: All predictions

r 2	0.33
RMSE	1.06
Mean error	0.76
Predictions within 1 log unit	395 from 529 ( 74.7 %)

Poor indication of the overall performance. Depends predominatly on the fraction of compounds beyond the applicability domain, which are by definition poorly predictable (see Applicability Domain Definition )

Figure 1: Predictions within 1 log unit vs. prediction confidence

Depicts the dependency of predictive accuracy on the confidence index (i.e. the distance to the applicability domain, see Applicability Domain Definition ). Fluctuations at the left hand side of the figure are statistical artefacts (small sample sizes) and therefore irrelevant.
The table of misclassifications shows all misclassified instances within the applicability domain.

Figure 2: Correlation of predicted vs meaѕured values

Predictions within the applicability domain are marked in black.

Mechanistic Interpretation

Neighbors

Neighbors are compounds that are similar in respect to 96 hr lc50 . It is likely that compounds with high similarity act by similar mechanisms as the query compound. You can retrieve additional experimental data and literature citations for the neighbors and the query structure by following the "Search PubChem" links on the prediction page.

Fragments

Activating and deactivating parts of the query compound are highlighted in red and green. Fragments that are unknown (or too infrequent for statistical evaluation are marked in yellow. You can retrieve additional statistical information about the individual fragments by following the "Relevant Fragments" link. Please note that lazar predictions are based on neighbors and not on fragments. Fragments and their statistical significance are used for the calculation of activity specific similarities.