This study investigates the interaction between proteins and ligands using multiple trajectory analysis methods. The goal is to identify binding positions and strengths, which are crucial for drug discovery. Molecular docking and molecular dynamics (MD) techniques are widely used to predict binding positions and affinities. However, few studies have systematically explored MD trajectory evolution in this context, potentially missing important information. To address this, a framework called Moira (molecular dynamics trajectory analysis) was developed to automate the entire process, including docking, MD simulations, and various analyses and visualizations. Moira was used to analyze 400 MD simulations based on geometric features (root mean square deviation and protein-ligand interaction profiler) and energetics (molecular mechanics Poisson-Boltzmann surface area) to distinguish native poses among four poses. The study first evaluates the performance of AutoDock Vina in predicting binding positions by analyzing 13,450 complexes in the PDBbind database. It then compares the performance of various scoring strategies for ranking different poses, including both static and dynamic methods. The results show that AutoDock Vina has limited accuracy in evaluating static poses. Next, the study investigates the structural evolution of protein-ligand complexes using MD simulations for 100 complexes with four poses over a 25 ns time span. The results show that the c_native pose consistently exhibits higher stability compared to c_5a and c_10a. The RMSD values for the c_native and c_2a initial structures remain consistently low, while the RMSD values for the c_5a and c_10a pose initial structures are significantly larger and exhibit substantial fluctuations. The study also analyzes the interactions between the ligand and the protein using PLIP. The results show that hydrophobic interactions, hydrogen bonds, and water bridges are the most prominent driving forces in ligand-protein binding. The study further evaluates the performance of MM/PBSA in predicting binding affinities. The results show that the c_native pose consistently outperforms the other poses in terms of Pearson correlation with experimental values. The study concludes that the Moira workflow provides a comprehensive paradigm for the exploration of ligand-protein interactions based on classical molecular dynamics, including both the simulations and their analysis. The workflow facilitates convenient investigation of ligand-protein interactions on a general basis, important for the further advancement of drug design methodologies.This study investigates the interaction between proteins and ligands using multiple trajectory analysis methods. The goal is to identify binding positions and strengths, which are crucial for drug discovery. Molecular docking and molecular dynamics (MD) techniques are widely used to predict binding positions and affinities. However, few studies have systematically explored MD trajectory evolution in this context, potentially missing important information. To address this, a framework called Moira (molecular dynamics trajectory analysis) was developed to automate the entire process, including docking, MD simulations, and various analyses and visualizations. Moira was used to analyze 400 MD simulations based on geometric features (root mean square deviation and protein-ligand interaction profiler) and energetics (molecular mechanics Poisson-Boltzmann surface area) to distinguish native poses among four poses. The study first evaluates the performance of AutoDock Vina in predicting binding positions by analyzing 13,450 complexes in the PDBbind database. It then compares the performance of various scoring strategies for ranking different poses, including both static and dynamic methods. The results show that AutoDock Vina has limited accuracy in evaluating static poses. Next, the study investigates the structural evolution of protein-ligand complexes using MD simulations for 100 complexes with four poses over a 25 ns time span. The results show that the c_native pose consistently exhibits higher stability compared to c_5a and c_10a. The RMSD values for the c_native and c_2a initial structures remain consistently low, while the RMSD values for the c_5a and c_10a pose initial structures are significantly larger and exhibit substantial fluctuations. The study also analyzes the interactions between the ligand and the protein using PLIP. The results show that hydrophobic interactions, hydrogen bonds, and water bridges are the most prominent driving forces in ligand-protein binding. The study further evaluates the performance of MM/PBSA in predicting binding affinities. The results show that the c_native pose consistently outperforms the other poses in terms of Pearson correlation with experimental values. The study concludes that the Moira workflow provides a comprehensive paradigm for the exploration of ligand-protein interactions based on classical molecular dynamics, including both the simulations and their analysis. The workflow facilitates convenient investigation of ligand-protein interactions on a general basis, important for the further advancement of drug design methodologies.