Various (all depressing) possibilities..
You don’t say what the sequence ID to the models us
Your protein is flexible with poor fit to the available models. How well do
those overlap?
The bioinformatics tools in i2 give you pictures showing this. Look at them
and see if there are sensible places t
Hi Marco
Further to Pedro’s comment, one way to avoid missing data in cusps is to use a
multi-sweep strategy for data collection (as championed by the good folk at
Global Phasing over many years). Many of the problems encountered in structure
solution can be traced back to a sub-optimal data co
Hi Marco,
Further to Zhen's comment, that can also happen if you have a cusp
containing one of the reciprocal lattice axes. The SG may be P212121 but
one of the 21 is never located due to the missing data.
You will always sort of get a solution from MR but you should look at
the Z-score of t
Hello Marco,
I also feel that it might be due to the wrong space group because all the
homologous models and alphafold model did not improve the R values. Make sure
that you turn on “All possible in same pointgroup” when you run
Phenix.Phaser-MR. I work with P212121 crystals a lot. Sometimes, i
