Milk samples were collected from Holstein and Jersey breeds of dairy cattle maintained under the same management practices and environmental conditions over a seven-day period. Milk samples were collected twice daily from six cows of each breed as described in Tacoma et al., 2016. Samples were composited within individual cow over the experimental period and skimmed to remove the fat layer. Skimmed milk samples were fractionated using CaCl2 precipitation, ultracentrifugation and ProteoMiner treatment to remove the high abundance milk proteins. Separation of the low abundance proteins was achieved using SDS-PAGE. Differential protein abundances were analyzed by mass spectrometry-based proteomic approaches followed by statistical analyses of the peptide count data. The complete list of low-abundance proteins identified in both breeds is provided in this dataset as well as the total number of distinct sequenced peptides and gene ontology functions for each protein. The relative abundance of a select few proteins is depicted using the SIEVE software. Data are presented in Excel format (zipped for download): Supplemental Table 1: Identification of gene ontology functions of the full protein profile between different breeds; Supplemental Table 2: The complete list of the low-abundance proteins identified as well as peptide count data from both breeds.