Glycobiological chemistry, which focuses on the study of the structure, function, and biosynthesis of glycans, has emerged as a critical field in understanding complex biological processes and developing novel therapeutic interventions. The intricate nature of glycans, coupled with their extensive roles in cellular communication, immune response, and disease pathology, necessitates advanced methodologies for their analysis and manipulation. In this context, the integration of conventional molecular dynamics (cMD) simulations, enhanced sampling methods and genetic algorithms have become indispensable, providing deep insights into the dynamic behavior and conformational landscapes of glycan-related enzymes and complexes. Recent advances in glycobiological chemistry have significantly deepened our understanding of glycan structures and their roles in biological systems, driven by innovations in both experimental and computational techniques. Experimentally, cutting-edge methods such as cryo-electron microscopy (cryo-EM), advanced mass spectrometry, and high-resolution NMR spectroscopy have enabled the precise characterization of glycan structures and their interactions with proteins, providing insights into the mechanisms of glycosylation and its implications in disease. Concurrently, the above mentioned computational approaches have advanced the ability to model the dynamic behavior of glycans and their associated enzymes at an atomic level, revealing key conformational states and energy landscapes that underlie glycan function and enzyme catalysis. These developments have been further augmented by machine learning algorithms and bioinformatics tools that facilitate the prediction of glycan structures and their biological roles. Such computational approaches offer a powerful tool to explore the time-dependent behavior of biomolecules at the atomic level, allowing for the elucidation of structural and functional relationships that are often inaccessible through experimental techniques alone. In addition, scientific programming plays a pivotal role in the implementation and optimization of these computational techniques, enabling the development of custom algorithms and workflows tailored to the specific challenges posed by glycan-related systems. Together, these experimental and computational advancements are pushing the boundaries of glycobiological research, offering new avenues for therapeutic intervention and biomolecular engineering. The present Ph.D. thesis has focused on applying and developing computational techniques, protocols and analysis to complex glycan-protein systems. The first part of this work reports the application of a multi-scale computational protocol to study egression pathways and water dynamics of glycan-based products in the glycohydrolase HvExoI, the formation and characterization of pre-catalytic complexes in glycotransferase GGCGT, the characterization of loop conformational states in glycoside oxidase PsG3Ox and the conformational exploration of the peptide beta-endorphin. HvExoI is crucial for the breakdown of complex carbohydrates, influencing plant defense mechanisms and energy metabolism, with potential implications for biofuel production and agricultural biotechnology. GGCGT is key in the biosynthesis of complex glycans with C-glycosydic linkages, which are essential for cell signaling, immune responses, and protein stability, highlighting its importance in developing novel glycoengineering strategies and therapeutic glycoproteins. PsG3Ox is involved in the oxidative degradation of glycosidic bonds, a process that has significant applications in biocatalysis and the development of biotechnological processes for carbohydrate conversion. Finally, beta-endorphin, an endogenous opioid peptide, plays a vital role in pain modulation, immune function, and stress response, making it a crucial target for research into pain management and neuropharmacology. The second part of this work has focused on the improvement and development of a new multi-objective genetic algorithm for complex molecular modeling, GaudiM2. GaudiM2 is an open-source application developed as an extension of its predecessor, GaudiMM, with a focus on employing multi-objective search strategies to explore complex systems with a large number of potential solutions.