Data Quality Issues in Toxicological Knowledge Discovery
Christoph Helma, Eva Gottmann, Stefan Daniel Krämer, Bernhard Pfahringer · 1999
Every SAR technique for toxicity prediction relies on the exact estimation and representation of chemical and toxicological properties. We will present poten-tial sources of errors associated with the utilization of l~trge, noncongeneric datasets and complex toxicologi-(:al endpoints (e.g. carcinogenicity). According to (~xperience we have identified the major problems in the areas of compound i entification, descriptor calcu-lation and toxicity data. Generally, we consider the chemical data as more reliable than the results from toxicity experiments. As it is impossible to tackle the data quality problem on a case by case basis for a large mlmber of compounds, we will propose some possibili-tie. ~ for routine quality control of large datasets.