摘要: High-throughput and high-content pharmacological screening methods generate such a flood of chemical biological data that their analysis would barely be feasible without the use computational tools (chapters 6 15). As is case with forward genetics (chapter 8), for which target itself unknown information incomplete, small molecule represents best defined piece data. Chemists, biologists informaticians have to pull together as they can in order create from joint observations ingenious hypotheses about reasons moleculeO s bioactivity, or lack thereof. An apparently simple question ask is: what molecule? And sensible representation make it? Example 11.1 illustrates range possible answers.