Download PDFOpen PDF in browser

A Machine-Learning Based Real-Time Analysis to Quantify the Effects of Gut Microbes in Social Insect Collective Behavior

EasyChair Preprint no. 1669

2 pagesDate: October 16, 2019

Abstract

Researchers in gastroenterology, microbiology and neuroscience employ animal models to investigate the role of gut microbes in socially relevant behavior. Surprisingly, despite being lab-friendly models for sociality-related questions, group-living insects receive little attention from this perspective. Nonetheless, insect social behavior research is currently undergoing a radical transition, thanks to the increasing availability of high-throughput video tracking systems. However, these video tracking systems have minimal behavioral pattern identification and perform data analysis only a posteriori on previously recorded videos. Our goal is to study how gut microbes of individual ants affect the emerging properties of the whole colony, and develop a machine-learning based behavioral pattern classifier, to help us recognizes individual behavior in real-time. This intelligent tracking system will be connected with a processing backend for simultaneous data harvesting and analysis. The device will be used to monitor laboratory ant colonies including artificially gut microbe-remodelled workers in differential proportions. Generated datasets will include individual behavior, social interactions and colony-level information. There are two main predictable scenarios: 1) microbe remodelling-mediated individual behavioral changes do not affect the emerging properties of the group; 2) colony-level properties vary based on the presence/proportion of microbe-suppressed individuals. This study would be the first attempt to investigate the role of gut microbes on whole animal societies, and the first to provide these results in real-time.

Keyphrases: animal behavior, machine learning, social evolution, social insects, video tracking

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@Booklet{EasyChair:1669,
  author = {Alessio Sclocco and Serafino Teseo},
  title = {A Machine-Learning Based Real-Time Analysis to Quantify the Effects of Gut Microbes in Social Insect Collective Behavior},
  howpublished = {EasyChair Preprint no. 1669},

  year = {EasyChair, 2019}}
Download PDFOpen PDF in browser