Automated analysis of long-term grooming behavior in Drosophila using a k-nearest neighbors classifier.

摘要: From birds that preen their feathers to dogs lick fur, many animals groom themselves. They do so stay clean, but routine grooming also has a range of other uses, such as social communication or controlling body temperature. Despite its importance, remains poorly understood; it is especially unclear how this behavior regulated. Fruit flies could be good model study because they are often used in laboratories look into the genetic and brain mechanisms control behavior. Flies clean themselves by sweeping legs over wings body, little known about insects ‘naturally’ long periods time. This partly scientists have had recognize classify eye, which highly time-consuming. Here, Qiao, Li et al. created system automatically detect fruit First, camera records movement an individual insect. A computer then analyzes images picks out general features fly’s can help work what insect doing. For example, if fly moving limbs, not main part probably itself. borrowed algorithm from area science ‘machine learning’ teach each automatically. The new successfully recognized 90% cases, revealed spend 13% waking life grooming. It showed seems controlled two potentially independent internal programs. One program tied clock fly, regulates when grooms during day. commands cleans itself, balances amount time spent on with behaviors. Cleaning oneself just important for disease-free: reflects health state individual. loss associated sickness, old age, and, humans, mental illness. If understand at molecular levels, may give insight these relate diseases. make studies possible.