Traits such as feed efficiency in dairy cattle are likely to be influenced by the genome of the host and the composition and abundance of microbiomes in the rumen. Here we describe an integrative approach that utilizes both genomic (SNP) and rumen microbiome data to predict future residual feed intake (RFI). The approach was tested in a small sample, of 28 Australian Holstein-Friesian dairy cattle that had 30K SNP genomic predictions for RFI and rumen microbiome profiles. The genomic and microbiome profile predictions were combined using a linear regression model. Results are very preliminary due to the small size of the data set, however the prediction accuracy in cross validation was maximized when both SNP and rumen microbiome profiles were used (r=0.57; 95% CI: 0.33:0.72). These results, while promising, should be repeated in a larger data set.