Link to bioRxiv paper:
http://biorxiv.org/cgi/content/short/2023.04.23.538002v1?rss=1
Authors: Inaba, S., Uesaka, N., Tanaka, D. H.
Abstract:
Disgust represents a quintessential manifestation of negative affect. Prototypical sensory expressions of disgust are triggered by bitter and other unappetizing tastes in human infants, non-human primates, and rodents. Disgust in mice has been quantified through the taste reactivity (TR) test. TR has been video recorded and counted manually to be quantified, requiring a significant amount of time and effort, however. Here we constructed the method to automatically count TR to assess both innate and learned disgust in mice using machine learning. We automatically tracked TR using DeepLabCut as the coordinates of the nose and both front and rear paws. The automated tracking data was split into test and training data, and the latter was combined with manually labeled data on whether or not a TR was present, and if so, which type of the TR was present. Then, a random forest classifier was constructed, and the performance of the classifier was evaluated in the test dataset. Throughout, the total numbers of disgust TRs predicted by the classifier were highly correlated with those counted manually. The present method will facilitate large-scale screening and long-term experiments that require counting numerous TR, which are challenging to conduct manually.
Copy rights belong to original authors. Visit the link for more info
Podcast created by Paper Player, LLC
