Usually, spectral data are used as predictors to predict milk components, animal
characteristics, and even reproductive status. Another innovative way to use spectral data involves
considering spectral wavenumbers as traits and then analyzing from the genetic perspective. In this
study, we considered milk spectral data directly as traits, then detected the influence of some
non-genetic factors on spectral wavenumbers and estimated the genetic parameters of spectral points.
The result of the present study could be used as a management tool for dairy farm and also provides
a further understanding of genetic background of milk mid-infrared (MIR) spectra. In future, milk
spectral data could be applied more effective. For example, some sub-clinical diseases might be
detected based on the difference between the expected and observed values of the spectral traits.
In addition, we could also use genetic correlation between wavenumbers and a trait of interest, which
are difficult and expensive to measure, to apply for the genetic improvement of dairy species.