Skip to main content

Table 2 A list of covariates included in multiple kernel learning, multilayer BayesB, and partial least squares (PLS)

From: Integrating genomic and infrared spectral data improves the prediction of milk protein composition in dairy cattle

Model

Sub-model

Effecta

Herd

DIM

Parity

FTIR

Genomics

Top markers

Pedigree

Kernel

M1

   

\(\checkmark\)

   

M2

\(\checkmark\)

  

\(\checkmark\)

   

M3

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

   

M4

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

  

M5

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

 

\(\checkmark\)

 

M6

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

 

M7

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

  

\(\checkmark\)

BayesB

M1

   

\(\checkmark\)

   

M2

\(\checkmark\)

  

\(\checkmark\)

   

M3

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

   

M4

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

  

PLS

M1

   

\(\checkmark\)

   

M2

\(\checkmark\)

  

\(\checkmark\)

   

M3

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

   

M4

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

  

M7

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

\(\checkmark\)

  

\(\checkmark\)

  1. aDIM: days in milk; FTIR: milk Fourier transform infrared spectroscopy; Top markers: top three markers with the largest effects; Genomics: genomic relationship matrix in kernel methods, markers in BayesB, and principal components of genomic relationship matrix in PLS; Pedigree: numerator relationship matrix in kernel methods and principal components of numerator relationship matrix in PLS