Recep Adiyaman
Vision About Experience Education Skills Publications Tools Newsletter Updates Podcasts Blog Contact CV Subscribe
All Issues
Older Newer
Daily Signal April 07, 2026 · 8 min read

Issue #84: Salt Bridge Builder: Using Residue Distances to Predict Salt Bridge Formation.

Protein Design Digest #84: BA-Pred and RMSD-Pred: Integrated Graph Neural Network Models for Accura…

Share X LinkedIn
Protein Design Daily

Building something in Protein Design?

I love collaborating on new challenges. Let's build together.

LinkedIn Email

Subscribe to Protein Design Digest

Daily curated signals from arXiv, PubMed, and BioRxiv.

Signal of the Day

Salt Bridge Builder: Using Residue Distances to Predict Salt Bridge Formation.

Salt bridges contribute disproportionately to protein folding stability and protein-protein interaction energetics, yet systematic tools for engineering novel salt bridges remain limited. There are several approaches that can quantify the energetics of removing salt bridges between proteins, but no existing tools are available for adding salt bridges at protein interfaces. Here, we introduce Salt Bridge Builder (SBB), a software package that identifies candidate mutation sites for adding interprotein salt bridges using residue distance heuristics derived from large-scale structural data. Using the SKEMPI v2 database, we demonstrate that charged-to-uncharged mutations that disrupt interprotein salt bridges result in binding free energy penalties significantly larger than those of comparable mutations that do not, underscoring the stabilizing role of salt bridges at protein interfaces. We benchmark six residue distance metrics for their ability to predict salt bridge formation and show that the side-chain centroid distance (SCCD) provides the optimal balance between the predictive performance and computational efficiency. Based on these findings, we formulate an efficient algorithm that identifies putative salt bridge-forming mutations while avoiding disruption of existing electrostatic interactions. We apply SBB to the kinesin superfamily and identify kinesin-5 as uniquely enriched in potential salt-bridge-building sites at the microtubule interface. Molecular dynamics simulations of engineered kinesin-5 mutants reveal that only a subset of predicted salt bridges exhibits high occupancy, highlighting the role of local microenvironments in stabilizing engineered electrostatic interactions. Principal component analysis of the residue microenvironment distinguishes high-occupancy salt bridges, suggesting a path toward a priori stability prediction. Long-range electrostatic force calculations further show that selected mutations modulate kinesin-5-microtubule attraction. Together, this work establishes residue-distance-based salt bridge engineering as a viable protein-protein engineering strategy and provides a foundation for future extensions of SBB that incorporate microenvironment-aware stability prediction.

Why this matters:

Also Worth Reading

Enhancing CYP450-Ligand Binding Predictions: A Comparative Analysis of Ligand-Based and Hybrid Machine Learning Models.

Predicting cytochrome P450 (CYP450) ligand binding is critical in early-stage drug discovery as CYP450-mediated metabolism profoundly influences drug efficacy, safety, and adverse reaction risks. However, experimental determination of CYP450-ligand interactions remains resource- and time-intensive, underscoring the need for robust computational alternatives. While ligand-based methods are commonly employed, they often fail to fully account for structural intricacies governing protein-ligand interactions. To address this gap, we developed a hybrid machine learning framework integrating ligand descriptors, protein descriptors, and protein-ligand interaction descriptors that include molecular docking-derived parameters, rescoring function components from multiple algorithms, and structural interaction fingerprints (SIFt). Evaluated on CYP1A2 and CYP17A1 isoforms, our model demonstrated superior predictive accuracy in cross-validation compared with stand-alone molecular docking and ligand-based approaches. Furthermore, benchmarking against state-of-the-art tools (SwissADME and ADMETlab 3.0) revealed enhanced performance in binding prediction. This work establishes a versatile framework for advancing computational tools to prioritize CYP450 binding assessments during drug discovery.

Evaluating zero-shot prediction of monomeric protein design success by AlphaFold, ESMFold, and ProteinMPNN.

De novo protein design has enabled the creation of proteins with diverse functionalities that are not found in nature. Despite recent advances, experimental success rates remain inconsistent and context-dependent, posing a bottleneck for broader applications of de novo design. To overcome this, structure and sequence prediction models have been applied to assess design quality prior to experimental testing to save time and resources. In this study, we examined the extent to which AlphaFold, Protein MPNN, and ESMFold can discriminate between experimentally successful and unsuccessful designs. We first curated a benchmark dataset of 614 experimentally characterized de novo designed monomers from 11 different design studies between 2012 and 2021. All predictive models demonstrated moderate ability to discriminate experimental successes (expressed, soluble, monomeric, and fold with the correct secondary structure) from failures. Still, many failed designs have better confidence metrics than successful designs, and confidence metrics were topology-dependent. Among all computational models evaluated, ESMFold average predicted local-distance difference test (pLDDT) yielded the best individual performance at distinguishing between successful and unsuccessful designs. A logistic regression model combining all confidence metrics provided only modest improvement over ESMFold pLDDT alone. Overall, these results show that these models can serve as an initial filtering strategy prior to experimental validation; however, their utility at accurately predicting experimentally successful designs remains limited without task-specific training.

Molecular docking, MM/GBSA, FEP/MD, and DFT/MM MD studies on predicting binding affinity of carbonic anhydrase II inhibitors.

Human carbonic anhydrase II (hCAII) is one of the zinc-containing metalloenzymes that catalyzes various hydration reactions. We report an investigation of whether modern computational tools can predict inhibitory potency of a set of sulfonamides against hCAII. The methods used are molecular docking, molecular dynamics simulations coupled with the free energy perturbation theory (FEP/MD), molecular mechanics combined with the generalized Born and surface area continuum solvation (MM/GBSA), and quantum mechanics/molecular mechanics (QM/MM) metadynamics simulations. A comparison is presented between experimental and computed binding free energy properties. All MD-based approaches demonstrate robust performance with R 2 values in the range of 0.89 to 0.99 for the subset of structurally simple sulfonamides, underscoring the importance of accounting for dynamic protein-ligand interactions. With regard to the other subset comprised of structurally more diverse sulfonamides, for which the QM(B3LYP)/MM(CHARMM) metadynamics approach is less affordable, the FEP/MD method yields an R 2 value of 0.70. Notably, the R 2 value increases to 0.80 after the removal of one outlier (chlorzolamide).

Research & AI Updates

From the Industry

Quick Reads

Computer-aided discovery of CDK16 inhibitors: a docking-augmented machine learning regression modelling approach.

Cyclin-dependent kinase 16 (CDK16), a serine/threonine protein kinase, is a critical regulator of cell cycle progression, vesicle trafficking, and apoptosis. Read more →

Study on the Mechanism of Ku Diding in the Treatment of Diabetes based on Network Pharmacology, Molecular Docking Technology, and Molecular Dynamics.

Introduction To explore how Ku Diding (KDD) works in managing Diabetes Mellitus (DM), researchers utilized network pharmacology, molecular docking, and molecular dynamics methodologies. Read more →

Ferulic Acid- and Quercetin-based Extracts Enhanced Neuropharmacological Potential in Swiss Albino Mice: Molecular Dynamics Simulation,mMolecular Docking, ADMET, and Biological Investigation.

Neurological disorders impair cognitive, emotional, and behavioural functions, leading to a reduced quality of life. Read more →

An integrative pharmacovigilance, network toxicology and molecular docking study on drug-induced cheilitis.

Drug-induced cheilitis represents an inadequately comprehended adverse reaction, with limited understanding of its underlying mechanisms. Read more →

Salt Bridge Builder: Using Residue Distances to Predict Salt Bridge Formation.

Salt bridges contribute disproportionately to protein folding stability and protein-protein interaction energetics, yet systematic tools for engineering novel salt bridges remain limited. Read more →

RinPy, a Python Package for Residue Interaction Network Model to Analyze Protein Structures and Predict Ligand Binding Sites.

The residue interaction network (RIN) model has been a useful approach for identifying functional residues and ligand-binding sites in different protein structures based on their contact topology. Read more →

Modulation of nutritional and aroma qualities of rapeseed oil by microwave and dehulling treatments: Based on insights from metabolomics, volatile compound analysis, and molecular docking studies.

This study compared the differences in compositional profiles and aroma characteristics between unde-hulled rapeseed oil (U-RO) and dehulled rapeseed oil (P-RO) after microwave pretreatment. Read more →

Computational discovery of SARS-CoV-2 viral entry inhibitory peptides from Androctonus mauretanicus scorpion venom: molecular docking and molecular dynamics simulations targeting the spike protein

Pipeline Tip

Normalise thermal B-factors when comparing different crystal structures.

Resources & Tools

Deep learning is not a magic wand, but a powerful lens for structural biology. — Recep Adiyaman

BS HF DK