Dhibar, S. | bioRxiv Science

bioRxiv2025

Correlations among hydrogen bond fluctuations in the apo state are enough to reveal allosteric networks in proteins

Allostery is essential for regulating biochemical pathways, where ligand binding at one site influences enzymatic activity at a distant functional site. Identifying allosteric sites and mapping signal transduction pathways in biomolecules remain a significant challenge. Existing experimental or computational methods, in general, can identify a subset of total allosteric network. Here, we consider that in addition to providing structural stability and flexibility of proteins, hydrogen bonds inside the protein may act as conduits for long-range communication. To this end, we develop a computational framework that predicts allosteric sites and total allosteric network by analyzing correlations among hydrogen bond fluctuation from equilibrium molecular dynamics trajectory of apo state of proteins. We demonstrated that this method can accurately captures experimentally verified allosteric sites and suggest allosteric signal transduction pathways across three different proteins. Furthermore, since our predictions are derived solely from the simulation trajectory of the apo state, these findings reinforce the idea that the signature of allostery is inherently encoded in the apo state of the protein. This approach offers a useful strategy to decode allosteric network and pockets, with broad implications for drug discovery and the targeted modulation of allostery in proteins. SignificanceAllostery, distal ligand-induced functional modulation of biomolecules, is central to biological regulation, yet its intricate mechanisms remain elusive. Precise identification of allosteric sites and pathways is essential for targeted drug development, minimizing off-target effect. Here, we introduce a novel computational approach, Hydrogen Bond Allosteric Map (HBAlloMap), which leverages the dynamic fluctuations of hydrogen bond networks derived from microsecond molecular dynamics simulations to comprehensively map allosteric modules within proteins. By analyzing the fluctuation of correlated hydrogen bonds, our method effectively reveals key allosteric hotspots and signal transduction pathways in PDZ3, PDZ2, and Pin1. By providing a robust and efficient tool for deciphering allosteric mechanisms, this method has the potential to accelerate drug discovery and deepen our understanding of allostery.

biophysics↗

bioRxiv2025

Prediction of Enzyme function using interpretable optimized Ensemble learning framework

Accurate prediction of enzyme function, particularly for newly discovered uncharacterized sequences, is immensely important for modern biology research. Recently machine learning (ML) based methods have shown promises. However, such tools often suffer from complexity in feature extraction, interpretability, and generalization ability. Here we present an interpretable ML method, SOLVE that addresses these issues by using only combination of tokenized subsequences from the proteins primary sequence for classification. Its optimized ensemble learning framework improves prediction accuracy, distinguishes enzymes from non-enzymes, and predicts enzyme commission (EC) numbers for mono- and multi-functional enzymes. Additionally, SOLVE provides interpretability through Shapley analyses, identifying functional motifs at catalytic and allosteric sites of enzymes. By focusing only on primary sequence data, SOLVE simplifies high-throughput enzyme function prediction for functionally uncharacterized sequences and outperforms existing tools. With high prediction accuracy and its identification ability of functional regions, SOLVE can become a promising tool in different fields of biology and therapeutic drug design.

biophysics↗