| The United States -- Environmental Protection Agency (U.S.- E.P.A.) gives out a list of chemicals, known as Contaminant Candidate List (CCL), periodically. This list includes those chemicals that are found in drinking water even after necessary water treatment is carried out. In this work, we generate compounds using inverse quantitative structure-activity relationship (I-QSAR) methodology and the molecular descriptor called the Signature. The properties considered are the log D values at seven pH, LD50 values for rat and mouse. We derive the height-1 (H-1) signatures for the training set that consists of seven compounds recognized by the US-EPA. Then using the knowledge of the signature database and the descriptor matrix, we derive the constraint equations and solve them. We generated around 300 million solutions and after screening through various screens and regenerating those solutions, we arrive at a focused database of potential candidates. We categorized those compounds into different scaffolds based on their structure for further studies... |
