[phenixbb] Merging reflections
Terwilliger, Thomas Charles
terwilliger at lanl.gov
Mon Jun 1 14:19:24 PDT 2015
If you have good data to 3 A and a search model, go ahead and use MR to solve the structure. Just delete the ligand from the search model and you won't have model bias and you can see if the ligand is there.
You might use the anomalous signal to check your solution by calculating an anomalous difference Fourier after you have obtained an MR solution. It should have a peak at the position of your Mo atom, perhaps visible even if the anomalous data are pretty weak.
The signal in your anomalous data may be much smaller than the errors in measurement in some of your datasets. In that case you wouldn't be able to see it.
All the best,
On Jun 1, 2015, at 3:09 PM, mohamed noor wrote:
> Dear all
> I have a few native datasets, one anomalous dataset with signal up to 10 A and two other datasets (also collected at the phasing wavelength but no anomalous signal seems to be present) and a sulfur SAD dataset with signal to about 7 A. All the datasets have a resolution of about 2-3 A and processed with xia2/XDS.
> I couldn't do fluorescence scan as the detector couldn't detect at such a high energy (20 keV).
> A few questions:
> 1. Why is there at least some signal from one dataset but not others? The signal should come from a covalently bound ligand, molybdopterin to be precise.
> 2. Since I(+) should not be equal to I(-) when there is anomalous signal, should I just merge the native dataset together with those collected at phasing wavelength that have no anomalous signal? Or do I merge everything together?
> 3. Will the anomalous signal to 10 A be useful to do MR-SAD? I could solve the structure with MR but I wanted to be sure my ligand is there.
> phenixbb mailing list
> phenixbb at phenix-online.org
> Unsubscribe: phenixbb-leave at phenix-online.org
More information about the phenixbb