Data mining of toxic chemicals: Structure patterns and QSAR

We take a two-step strategy to explore noncongeneric toxic chemicals from the database RTECS: the screening of structure patterns and the generation of a detailed relationship between structure and activity. An efficient similarity comparison is proposed to screen chemical patterns for further QSAR analysis. Then CoMFA study is carried out on one structure pattern as an example of the implementation, and the result shows that QSAR studies of structure patterns can provide an estimate of the activity as well as a detailed relationship between activity and structure. From the performance of overall procedure, such a stepwise scheme is demonstrated to be feasible and effective to mine a database of toxic chemicals.