As any music lover knows, the human ear is adept at picking out subtle patterns. And the growing power of computers to translate almost any kind of information into variations in pitch, rhythm, or volume is boosting the field of sonification, the representation of data as sound. From sounding out variations on a pathologist's tissue section slide to flagging suspicious travel activity, sonification has the potential to help scientists, doctors, and analysts spot trends and trouble spots.

Ronald Coifman, a mathematician at Yale University, and Jonathan Berger, a composer at Stanford University, have developed software that transforms light reflected off colon cells under a microscope into pulsating sounds. Under one setting, cancerous cells are louder than healthy ones. Coifman and Berger's study is mainly aimed at discovering which sound patterns are most effective at conveying complex data, which Coifman says they and other researchers will achieve in two years.