G-quadruplexes are secondary, non-canonical RNA/DNA structures formed by guanine-rich sequences assembled into four-stranded helical structures by the progressive stacking of G-Tetrads, planar arrangements of guanines stabilised by monovalent ions such as \(\text {K}^{+}\) or \(\text {Na}^{+}\) . Their stability plays a very important role in the prevention of DNA degradation, leading to the promotion or inhibition of specific biological pathways upon formation. In this work, we explore the occurrences of intermediates originating from the unfolding of these structures by using all-atom simulations, analyzing a small number of significant reaction coordinates to follow the evolution of the system by applying a mesoscopic simplification of the structures followed by two different dimensionality reduction techniques: Principal Component Analysis (PCA) and time-Independent Component Analysis (tICA). The data of the reduced trajectories are then encoded into a Complex Markov Network which, in conjunction with an Stochastic Steepest Descent, provides a hierarchical organization of the different nodes into basins of attraction. This procedure is able to reveal the main intermediates and the most relevant transitions the system undergoes in its denaturation path.
