Unraveling the Role of Hydrogen Bonds in Thrombin via Two Machine Learning Methods

Abstract

Hydrogen bonds play a critical role in the folding and stability of proteins, such as proteins and nucleic acids, by providing strong and directional interactions. They help to maintain the secondary and 3D structure of proteins, and structural changes in these molecules often result from the formation or breaking of hydrogen bonds. To gain insights into these hydrogen bonding networks, we applied two machine learning models - a logistic regression model and a decision tree model - to study four variants of thrombin: wild-type, ΔK9, E8K, and R4A. Our results showed that both models have their unique advantages. The logistic regression model highlighted potential key residues (GLU295) in thrombin’s allosteric pathways, while the decision tree model identified important hydrogen bonding motifs. This information can aid in understanding the mechanisms of folding in proteins and has potential applications in drug design and other therapies. The use of these two models highlights their usefulness in studying hydrogen bonding networks in proteins.

Allosteric Modulation of Thrombin by Thrombomodulin: Insights from Logistic Regression and Statistical Analysis of Molecular Dynamics Simulations

“…62 These parameters are as used in our prior research. 24 To optimize the computational efficiency of our analysis, we implemented a targeted filtration of hydrogen bonds based on their frequency of occurrence within the concatenated trajectory frames. Our strategy was to exclude hydrogen bonds that were either too sporadic or too persistent across all systems, as such bonds are unlikely to be crucial for distinguishing between different systems.…”

“…This reduction in computational demand is consistent with the O(n) complexity of the logistic regression model, 63,64 thereby validating our approach and corroborating the methods used in our previous research. 24 After filtering, 312 hydrogen bonds served as explanatory variables in a two-dimensional matrix. Each column represented a hydrogen bond, each row represented a frame of the concatenated trajectory, and each entry was either 1 or 0, indicating the presence or absence of the hydrogen bond within that frame.…”

“…For a detailed discussion of this method and its applications in our research, we direct the readers to our recent publication. 24 ■ RESULTS AND DISCUSSION RMSF Analysis Revealing TM56 and TM456 Induced Atomic Flexibility Changes in Thrombin. Figure 1 shows the RMSFs, highlighting the atomic flexibility changes in thrombin as a result of binding with TM56 and TM456.…”

“…The use of GPU-enabled simulations conferred a significant advantage, enabling us to extend each simulation to 50 times the duration of prior computational studies on TM456 . We have used such simulations before to study the dynamic allostery of thrombin due to ion binding, mutation, and aptamer binding as well as to study dynamic allostery in other biomedically relevant protein complexes. − …”

Unraveling the Impact of W215A/E217A Mutations on Thrombin’s Dynamics and Thrombomodulin Binding through Molecular Dynamics Simulations

“…For the analysis, logistic regression was employed as the statistical model, consistent with our prior methodology [22]. This framework uses a logistic function to model the relationship between a binary response variable and one or more predictor variables [65,66].…”