As an important part of the data-independent acquisition-mass spectrometry (DIA-MS) analysis approach, spectral library generation is closely related to the final protein identification and quantitation. In this study, an attempt is made to evaluate the influence of different sample separation methods (reversed-phase liquid chromatography (RPLC) linear gradient elution and spin column step gradient fractionation) and data acquisition modes (data-dependent acquisition (DDA) and DIA) on library generation of the human urine proteome. The RPLC separation method provides more proteins for generating the library than the spin column; however, the spin column library leads to identification of more proteins when used for the urine proteome analysis. The DDA mode can provide a larger spectral library than the DIA mode and the DDA-library leads to more protein identifications in the study. Use of a combined urinary protein library provides the most complete protein and peptide information, and the DIA approach can provide accurate protein quantitative information even for low-abundance proteins. When the combined library is used to analyze the normal individual urine proteome (18 males and 18 females), gender-related and age-related proteins are discovered and functionally analyzed. This strategy may benefit urine proteome DIA analysis.