Print

---

I would like to hear your opinion on how to model ion-binding sites at low resolution. 

I am currently working on a structure at 3.5 A resolution, which has several Mg2+ atoms bound, coordinated by both protein atoms and waters. The electron density nicely accommodates the coordinating water molecules around the Mg2+ (at positions already known from higher resolution structures), and not modelling the waters results in small positive Fo-Fc map peaks at some of the respective positions after refinement.

However, as far as I know, there is a general consensus in the field to not include waters into models at a resolution worse than about 2.8A. Would you say that this also holds for waters involved in ion coordination? 

I’m looking forward to hear your thoughts on this.

With best regards,

Maike