In this project, we conducted an evaluation of mass spectrometry methods using mouse heart and HCT116 cell samples. The results demonstrated that the data-independent acquisition (DIA) approach outperformed data-dependent acquisition (DDA) in the identification of Altprot and canonical proteins. Subsequently, we assessed several different DIA library building methods, including traditional DDA-based library building, gas-phase fractionation(GPF) library building, and machine learning-based prediction library building. Notably, the traditional DDA library building method exhibited a higher likelihood of false positive identifications. Furthermore, we applied the aforementioned mass spectrometry methods to investigate the process of mouse heart development. Through this analysis, we identified a subset of Altprots, including ASDURF, which may potentially play crucial roles in heart development. These findings serve as a fundamental basis for future exploration and investigation of Altprot in this context.