836.3/KK4

T-pattern sequential analysis to quantify ''route tracing'' locomotor stereotypic behavior

Georgia World Congress Center: Halls B3-B5

Wednesday, Oct 18, 2006, 3:00 PM - 4:00 PM

*S. J. BONASERA¹, K. A. SCHENK², L. H. TECOTT²;
¹Geriatrics, University of California, San Francisco, San Francisco, CA, ²Psychiatry, University of California, San Francisco, San Francisco, CA.

The advent of large-scale genomic and drug discovery technologies has highlighted the need for automated, high throughput, yet sensitive assays to quantify mouse behavior. To this end, we assessed t-pattern sequential analysis to quantify mouse locomotor stereotypies. This method reveals patterns by testing whether members of a set of defined states follow one another within a specific time interval at a probability greater than that of chance. Mouse home cage activity following treatment with the dopamine transporter inhibitor GBR-12909 (known to evoke stereotypic behaviors in mice) were analyzed. Dose-dependent increases in the total number of observed t-patterns, and overall t-pattern complexity were observed. Small sets of complex t-patterns accounted for a greater percentage of overall trial duration as GBR-12909 dosage increased, suggesting that t-pattern analysis yields a measure akin to a stereotypy score. In fact, this t-pattern derived stereotypy score showed the same overall trend with less within-group variability compared to scores derived from manual observation. These findings remained similar after reanalysis with removal of patterns unmatched after human scoring, 10 and 16% downsampling of original locomotor data, and normalization of locomotor speeds at both low and high ranges. We conclude that t-pattern analysis is a versatile and robust pattern detection and quantification algorithm that may augment currently available behavioral phenotyping strategies.

 S.J. Bonasera , None; K.A. Schenk, None; L.H. Tecott, None.

NIH 5 K08 MH065983-03

Stephen D. Bechtel Foundation