Figure 3
Defining daily torpor as an outlying low metabolism.
(a) The daily torpor-detection pipeline. The first 24-hour data set is used to estimate the baseline metabolism of the individual animal (first panel). The estimated baseline is then applied to the rest of the recordings (second panel). The baseline estimation provides the CI for the prediction from the distribution of the posterior estimates (third panel). Torpor, which is defined as a lower outlier from the CI, is marked in red dots (fourth panel). The filled and unfilled triangles denote food removal and return, respectively. (b) Multiple torpor definitions were compared in four mice (mice 13 to 16). The animals were placed in a constant TA of 12 °C for three days, and food was restricted during the second day. Results are shown for mouse 13. The two leftmost panels show daily torpor defined by a fixed threshold TB of 31 °C or 34 °C. The third panel shows daily torpor defined by a lower outlier of the 99.9% CI of the estimated TB. The fourth panel includes the TB-based definition further narrowed down by adding the condition of lower outliers from the 99.9% CI of the estimated VO2. The filled and unfilled triangles denote food removal and return, respectively. The data for the remaining three animals are available in Supplementary Fig. 3a. (c) Boxplots for various torpor statistics according to the different torpor definitions listed in Fig. 3b. The band inside the box, the bottom of the box, and the top of the box represent the median, the first quartile, and the third quartile, respectively. The end of the upper whisker is the highest value that is within 1.5 times the inter-quartile range (IQR). The end of the lower whisker is the lowest value that is within 1.5 times the IQR. All data points are shown as grey dots.
