Table 2 Associations between clusters and Cosinor model parameters.

From: Data-driven modelling approach to circadian temperature rhythm profiles in free-living conditions

 

Cluster

Mean

95% lower CI

95% upper CI

p value

FDR corrected p value

Period (h)

1

25.75

25.15

26.35

0.679

1.00

2

25.76

25.25

26.27

3

25.33

24.49

26.17

Mesor (°C)

1

34.29

34.22

34.35

< 0.001

< 0.001

2

33.46

33.40

33.51

3

32.80

32.71

32.89

Amplitude

1

1.02

0.93

1.10

< 0.001

< 0.001

2

1.38

1.30

1.45

3

2.09

1.96

2.21

Bathyphase based on cosinor model (hh:mm)

1

14:52

14:27

15:16

0.101

0.149

2

14:43

14:22

15:04

3

14:07

13:33

14:41

Bathyphase based on smoothed data (hh:mm)

1

13:35

13:06

14:05

0.492

0.528

2

13:59

13:34

14:24

3

14:17

13:36

14:58

Acrophase based on cosinor model (hh:mm)

1

2:16:15

1:47:12

2:45:18

0.154

0.193

2

1:50:36

1:25:43

2:15:29

3

1:28:44

0:47:53

2:09:36

Acrophase based on smoothed data (hh:mm)

1

2:40:23

2:14:35

3:06:11

0.017

0.038

2

2:34:30

2:12:24

2:56:37

3

2:02:20

1:26:02

2:38:37

  1. Bathyphase and acrophase are presented in time units. Both cosinor model estimates and observed times from loess smoothed data are presented. Confidence intervals calculated using bias corrected bootstrapping with 1000 bootstrapped resamples. False discovery rate corrected p-values take account all omnibus analysis presented in article.
  2. CI  confidence interval, FDR false discovery rate.
Back to article page