Skip to main content

Table 2 Result of the models for gut bacterial composition personalization. Personalization expressed as minimum dissimilarity Bray-Curtis (BC) and weighted unifrac (Wunifrac) (n = 543 samples). Season is marginally significant in one model and significant in interaction with age in the other

From: Aging gut microbiota of wild macaques are equally diverse, less stable, but progressively personalized

 

Fixed effects

Levels

Estimate

SE

CIlower

CIupper

p

BC

Intercept

 

−0.04

0.04

−0.11

0.03

-

 

Age1

 

0.03

0.01

0.01

0.06

0.02

 

Season2

     

0.06

  

M (ref L)

0.03

0.04

−0.05

0.09

 
  

R (ref L)

−0.06

0.04

−0.12

0.01

 
 

Group3

     

0.5

  

MST (ref MOT)

0.03

0.03

−0.02

0.09

 
  

SST (ref MOT)

0.04

0.04

−0.04

0.11

 
 

Gestation

 

−0.08

0.04

−0.15

−0.00

0.05

Dispersion parameter

Intercept

 

3.98

0.07

3.87

4.12

-

 

Age1

 

0.38

0.07

0.25

0.50

< 0.001

Wunifrac

Intercept

 

−1.68

0.03

−1.74

−1.62

-

 

Age1

 

−0.02

0.02

−0.05

0.02

-

 

Season2

     

-

  

M (ref L)

0.11

0.03

0.05

0.18

 
  

R (ref L)

0.06

0.03

0.00

0.11

 
 

Group3

     

0.6

  

MST (ref MOT)

0.01

0.02

−0.03

0.05

 
  

SST (ref MOT)

0.03

0.03

−0.02

0.08

 
 

Gestation4

 

−0.05

0.03

−0.11

0.01

0.08

 

Age1 × season2

     

0.04

  

Age × M (ref L)

0.01

0.03

−0.04

0.07

 
  

Age × R (ref L)

0.06

0.02

0.02

0.10

 

Dispersion parameter

Intercept

 

4.99

0.07

4.89

5.15

 
 

Age1

 

0.28

0.07

0.17

0.43

< 0.001

  1. 1Z-transformed to a mean of 0 and sd of 1; mean and sd of age on the original scale are 12.44 and 5.79 years, respectively. 2Categorical predictor with 3 levels: lean, intermediate, and rich, with the reference level set to the lean season. 3Categorical predictor with 3 levels MOT, MST, SST, with the reference level sets to group MOT. 4Binary predictor with reference sets to non-gestating