Most environmental matrices contain a diverse range of synthetic and natural compounds, some of which exhibit toxic effects. Nevertheless, current methods often do not provide sufficient resolution and power to specifically identify compounds responsible for the biological effects of the complex environmental mixtures. Here, we present the development of the EnDiTrap pipeline that facilitates the identification of compounds that specifically interact with targets involved in the regulation of the endocrine system. We used heterologously expressed ligand binding domain (LBD) of retinoic acid receptor alpha (RARα) as a model system for pull-down-based pipeline development, consequent optimization, and standardization. The applicability of the EnDiTrap pipeline was validated using standard ligand and tested through a case study with environmental samples of freshwater blooms. Results showed that the EnDiTrap pipeline significantly helps to reduce the number of putative features and facilitates the identification of suspect compounds responsible for observed biological effects. We also compared the performance of the software tools Compound Discoverer and MSDial commonly used for processing and analysis of mass spectrometry data. This comparison provided insight into the impact of different software processing on the outcome that brought interestingly contrasting results. This study enhances our ability to specifically identify effect drivers in environmental mixtures of chemicals. Moreover, the developed EnDiTrap pipeline can be applied to various protein targets thus presenting broad applicability.