In earlier lectures you saw experimental phasing methods like isomorphous replacement. Now we move to one of the MOST important and widely used methods today:
⭐ Molecular Replacement (MR) — solving the phase problem using an already known structure.
This lecture explains what MR is, how it works, why Patterson maps are critical, and what factors affect success.
Instead of experimentally determining phases:
➡️ Use a known homologous structure (search model) ➡️ Fit it into the crystal unit cell of the unknown protein ➡️ Calculate phases from this model
This is possible because:
Scientists often want multiple structures of the same protein:
Example enzyme states:
| State | Why study it |
|---|---|
| Apo state (no ligand) | Baseline conformation |
| Substrate-bound | Binding mechanism |
| Transition state | Catalytic mechanism |
| Product-bound | Reaction outcome |
👉 Once the first structure is solved, all later structures can often be solved very quickly using MR.
This makes MR:
✅ Fast ✅ Efficient ✅ Widely used ✅ Essential in modern crystallography
Not only identical proteins — many possibilities:
To place the search model correctly in the crystal:
You must determine:
Together → full placement in the asymmetric unit
Just like in heavy-atom phasing:
⭐ Patterson maps require NO phase information
They describe interatomic vectors.
This makes them perfect for MR.
Vectors between atoms within the same molecule
Properties:
👉 Used for the rotation function
Vectors between atoms in different molecules in the unit cell
👉 Used for the translation function
Procedure:
How comparison works:
🎯 Goal → Find 3 rotation parameters
If the cell box is too small:
❌ Intramolecular and intermolecular vectors overlap ❌ Hard to detect correct orientation
If box size is increased:
✅ Clear separation of vector peaks ✅ Better correlation ✅ Easier solution
👉 This is something crystallographers can tune computationally
Once orientation is known:
Now move the molecule around inside the unit cell.
Goal:
➡️ Match intermolecular Patterson peaks
When peaks overlap:
⭐ Correct position found
🎯 Goal → Find 3 translation parameters
Now:
Electron density can now be computed:
[
ho(x,y,z) = sum |F_| e^{iphi_} ]
Also includes:
MR works best when:
✅ High structural similarity ✅ Good resolution diffraction data ✅ Possible non-crystallographic symmetry averaging
Problems arise when:
❌ Very low homology ❌ Large conformational changes ❌ Low resolution data ❌ No NCS averaging ❌ Poor search model quality
Because:
It is now:
⭐ One of the most used methods in macromolecular crystallography
Think of MR as:
🧩 “Docking a known protein model into a crystal until the diffraction pattern makes sense.”
It is: