Automated multi-day tracking of marked mice for the analysis of social behaviour

Highlights

• A fully automated system to track multiple animals in a large arena without losing their identities is presented.

• The system learns unique bleach patterns on the mice's fur and tracks them during both dark and light cycles.

• Identification of six mice in the experimental setup was 97% correct during non-sleep intervals.

• As a proof of principle, we tracked groups of four mice and report social trends that develop across hours and days.

Abstract

A quantitative description of animal social behaviour is informative for behavioural biologists and clinicians developing drugs to treat social disorders. Social interaction in a group of animals has been difficult to measure because behaviour develops over long periods of time and requires tedious manual scoring, which is subjective and often non-reproducible. Computer-vision systems with the ability to measure complex social behaviour automatically would have a transformative impact on biology. Here, we present a method for tracking group-housed mice individually as they freely interact over multiple days. Each mouse is bleach-marked with a unique fur pattern. The patterns are automatically learned by the tracking software and used to infer identities. Trajectories are analysed to measure behaviour as it develops over days, beyond the range of acute experiments. We demonstrate how our system may be used to study the development of place preferences, associations and social relationships by tracking four mice continuously for five days. Our system enables accurate and reproducible characterisation of wild-type mouse social behaviour and paves the way for high-throughput long-term observation of the effects of genetic, pharmacological and environmental manipulations.

Introduction

Mouse models have been recently developed to study the cognitive and social deficits observed in autism (Jamain et al., 2008, Penagarikano et al., 2011), schizophrenia (Hikida et al., 2007, Tremolizzo et al., 2002), Down syndrome (Olson et al., 2004, Reeves et al., 1995) and fragile X syndrome (Kooy et al., 1996, Zang et al., 2009). Social relationships in mice develop and evolve over the course of many days (Hurst et al., 1993, Poole and Morgan, 1975). The ability to carry out thorough, quantitative, long-term observations would likely have transformative effects on understanding and measuring social behaviour and its pathologies. However, widely used assays are often performed for short durations that can miss persistent durable traits (Fonio et al., 2012). A key challenge in performing long-term assays is the ability to obtain reliable annotation. However, it is not practical to have these assays done by human experts because they are tedious, expensive and not easily reproducible (de Chaumont et al., 2012, Spencer et al., 2008). Computer vision systems that are able to analyse animal behaviour automatically hold much promise (Reiser, 2009). Despite recent progress, state-of-the art computer vision systems are limited to the observation of two mice sharing an unfamiliar enclosure for a period of 10–20 min, often in partition cages, which limit social interaction (de Chaumont et al., 2012, Spencer et al., 2008). Significant progress in the classification of actions, once animal trajectories have been computed, has recently been reported (Burgos-Artizzu et al., 2012, de Chaumont et al., 2012, Jhuang et al., 2010). However, reliable tracking and the identification of individual mice when multiple mice share the same enclosure for several days remains an open problem.

Automatically tracking the identities of multiple animals in a video sequence is difficult. Current approaches are based on the assumptions that the animals are always visible, do not overlap, and do not move too quickly, or employ heuristics, such as size differences across animals (Dankert et al., 2009), constrained environments (Branson et al., 2009) or artificially coloured markers (EthoVision, Noldus) to resolve animal identities. Attached coloured markers are easily groomed out and are not discriminable in infrared lighting, which is required for observation during dark cycles. All of the above approaches can fail and require human verification and correction of the results (de Chaumont et al., 2012). Furthermore, mice have flexible bodies, are highly interactive (cuddling, chasing, jumping on top of each other, mounting, etc.), and live in fairly complex environments (e.g., environments involving nests and bedding into which the mice burrow, which makes them invisible to the camera for periods of time). These factors make tracking and identification challenging, particularly when prolonged observation of social behaviour is desired.

We present a method that is capable of tracking individual mice interacting socially in a group over days without confusing identities; identities are maintained even when individuals hide and burrow in the bedding. The method consists of a single-camera computer vision system that automatically learns the appearance of each mouse and uses that appearance to infer each animal's identity throughout the experiment. We developed a set of uniquely discriminable patterns for marking the back of each animal. These patterns are produced by applying harmless hair bleach to the fur, cannot be groomed out, and can be tracked under infrared illumination during both dark and light cycles. The trajectories computed by our system may be used to detect and quantify mouse social behaviour (courtship, aggression, dominance, etc.) and to study its evolution over days. The system is easily reproducible, inexpensive, does not use any specialized hardware, user-friendly, and scalable to allow high throughput (the system and installation instructions are available at http://motr.janelia.org). Using our system, we characterised how social interaction developed in groups of four wild-type mice (two males and two females) over a five-day period.