# Shall i use molecular replacement in this case?

**URL:** <https://scientistsolutions.discourse.group/t/shall-i-use-molecular-replacement-in-this-case/618>\
**Category:** Protein Crystallography\
**Created:** [January 6, 2025, 12:17pm UTC](https://scientistsolutions.discourse.group/t/shall-i-use-molecular-replacement-in-this-case/618 "2025-01-06T12:17:47Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Nishan1618](https://avatars.discourse-cdn.com/v4/letter/n/f07891/32.png) [@Nishan1618](https://scientistsolutions.discourse.group/u/Nishan1618)\
**Post date:** [January 6, 2025, 12:17pm UTC](https://scientistsolutions.discourse.group/t/shall-i-use-molecular-replacement-in-this-case/618/1 "2025-01-06T12:17:47Z")

</div>

Hi I am working on a multidomain protein (single polypeptide) in which only one domain (say domain B) has related structures in the PDB. Is it possible to use molecular replacement using the available PDBs related to my domain B as a search model to find a successful solution to my whole multidomain protein within the Xtal? Is the subsequent manual building and refinement possible to solve the structure of the other domain?

Thanks

---

<div class="post-metadata">

**Author:** ![ARGERINE](https://avatars.discourse-cdn.com/v4/letter/a/8491ac/32.png) [@ARGERINE](https://scientistsolutions.discourse.group/u/ARGERINE)\
**Post date:** [June 16, 2025, 4:27pm UTC](https://scientistsolutions.discourse.group/t/shall-i-use-molecular-replacement-in-this-case/618/2 "2025-06-16T16:27:11Z")

</div>

Nishan1618

Its possible to use molecular replacement (MR) with a known structure of **domain B** to solve the structure of a **multidomain protein** if domain B is well-diffracting and structurally conserved. MR will locate domain B within the crystal, and this initial solution can often yield sufficient phase information to begin model building. After successful placement of domain B, you can perform **manual building and refinement** to model the remaining unknown domains (e.g., domain A), provided that the electron density map is of good quality. Iterative refinement (with tools like **Phenix** or **Refmac** ) and manual model building (using **Coot** ) can progressively complete the full protein structure. Challenges may arise if the unknown domain is flexible or poorly ordered, but with high-resolution data and a good MR solution for domain B, solving the entire multidomain structure is feasible. Sequence alignment and secondary structure predictions can guide model building of the unknown domains.