HRaDeX: high-resolution HDX–MS analysis made simple ⚙️🧬

HaDeX
publications
A new R package and web server (HRaDeX) enabling high-resolution analysis of HDX–MS data by reconstructing residue-level deuterium uptake rates from peptide-level measurements.
Author

BioGenies Lab

Published

May 19, 2025

Keywords

HDX-MS, protein dynamics, structural proteomics, HRaDeX, mass spectrometry, protein folding, bioinformatics tools

📌 Project highlights

🎉 New paper out! This one is all about turning messy HDX–MS data into something actually interpretable 😄

👉 HRaDeX: R Package and Web Server for Computing High-Resolution Deuterium Uptake Rates for HDX–MS Data

🔗 Try it yourself (no excuses 😉)

HRaDeX is available as both a web application and an R package, allowing flexible use across workflows, from quick exploration to large-scale analysis.

🎧 Audio summary

HDX–MS + peptide overlap + kinetic modeling = 😵‍💫

👉 So yes, this one also deserves a quick audio breakdown 🎧

🔬 What is this about?

HRaDeX is a bioinformatics tool for analyzing hydrogen–deuterium exchange mass spectrometry (HDX–MS) data, enabling reconstruction of protein dynamics at near residue-level resolution.

HDX–MS measures how proteins exchange hydrogen with deuterium over time, revealing:

  • structural stability
  • folding dynamics
  • solvent accessibility

👉 But standard analysis is limited to peptide-level resolution, averaging signals across multiple residues.

🚨 The core limitation

Classic HDX–MS:

  • ❌ loses residue-level detail
  • ❌ averages overlapping signals
  • ❌ makes precise interpretation difficult

👉 This is where HRaDeX comes in.

🧠 The idea behind HRaDeX

HRaDeX reconstructs high-resolution information using:

🧩 Overlapping peptides

  • combines multiple fragments covering the same region

⚙️ Kinetic modeling

  • fits uptake curves using extended ZS model
  • separates:
    • fast
    • intermediate
    • slow exchange components

👉 Each peptide becomes a mixture of dynamic populations.

🎨 Visualizing protein dynamics

One of the nicest features:

👉 RGB color encoding of exchange rates

  • 🔴 fast (flexible)
  • 🟢 intermediate
  • 🔵 slow (structured)

➡️ giving a single-view map of dynamics across the sequence

⚙️ How the workflow works

According to the workflow:

  1. ✅ Filter invalid data
  2. 🧊 Detect “no exchange” regions
  3. 📈 Fit kinetic models
  4. 🎯 Select best model (BIC)
  5. 🔗 Aggregate overlapping peptides
  6. 🎨 Map results to sequence/structure

⚙️ Two resolution strategies

HRaDeX provides:

1️⃣ Shortest peptide

  • assigns values from smallest fragment

2️⃣ Weighted average

  • integrates all overlapping peptides

👉 The combination enables higher resolution than standard HDX–MS

📊 Performance & validation

Tested extensively:

  • ✔️ >4000 peptides
  • 📉 very low fitting error (RSS)
  • 🎯 ~4–7% RMSE in reconstructed uptake

👉 Meaning: HRaDeX reliably reproduces experimental HDX behavior

🧬 What you can actually do with it

  • map flexible vs stable regions
  • detect binding interfaces
  • study allosteric effects
  • compare protein states

👉 Example: binding-induced conformational changes clearly visible in case studies

🚀 Why this matters

This work bridges a major gap:

👉 from peptide-level data → residue-level interpretation

It enables:

  • better structural insight
  • more precise protein dynamics analysis
  • improved reproducibility

⚠️ Limitations

  • depends on peptide coverage
  • requires good experimental design
  • resolution limited if overlap is low

👉 Still a major step forward compared to standard pipelines