Table 6.

Coefficients and the associated standard errors (SE) of the predictor variables of the final DMI-B (CH₄ yield; g/kg DMI) model.

	Predictor variables	SCAD1	Conventional2
Animal	Intercept	15.13 (0.13)	15.12 (0.17)
	Body weight (kg)	–	0.89 (0.13)
	Acetate to propionate (A:P) ratio	0.93 (0.12)	0.25 (0.16)
Bacteria**	Uncl_Family of the order Gastranaerophilales	0.50 (0.12)
	Genus Pseudoramibacter	− 0.24 (0.12)
	Genus Megasphaera	− 0.33 (0.12)
	Genus Selenomonas	− 0.39(0.13)
	Genus Oribacterium	− 0.40 (0.12)
Number of predictors		6	2
${\widehat{\sigma }}_{\alpha }$		0.98	1.47

The bolded genera were shared between the ANIM-B and the DMI-B models.

Uncl unclassified.

For all models, the predictor variables were centered and scaled to have a mean of 0 and variance of 1.

**Log-transformed relative sequence abundance.

¹Developed using SCAD and all the available data after log-transformation of the microbial data.

²Developed using conventional model and only animal-related data except DMI.