Thursday, 15 March 2012

Experimental Phasing with Pseudo-symmetry in an unusual setting P 21 2 21

From: Harm Otten
Date: 15 March 2012 15:14


Dear colleagues
do you have any professional thoughts on my case here?

Available SeMet data to 3.6 Å (where only the redundant Se-Edge peak to can be used), a "high resolution" native data to 2.6 Å, a
Zinc-soak, collected at the Zinc-Edge to 3.0 Å.

Considerable Patterson peak indicates pseudo translational symmetry, as of 42% of the origin peak. The fractional vector (x,y,z) is 0, 0.22, 0.5. The unit cell at 65 Å, 104 Å, 130 Å (all angles 90 deg) is expected to contain 2 copies at 51% solvent content.

Best solutions for the substructure with 6 of the expected 12 Selenium sites found in P 21 2 21 (sometimes called 2018). Low CC of ~0.4 and problematic to refine and or find new sites after density modification.

Molecular Replacement (with and without assisted SAD) did not work, i.e. R-factors >50% and CC <0.4. With the very best slightly manually trimmed and refined solution I see R/Rfree 0.47/0.51.

Programs in use are shelx_cde, autoSHARP, ARP/wARP, crank, oasis, rantan, phaser, mlphare, as well as phenix.autosol, phase_and_build, refine, sculptor, enseml, automr.

1) Can I reassure myself of the right space group? (tried visual absences inspection in HKLVIEW and pointless, phenix.xtriage)

2) What is the best phasing program at these resolutions in your experience?

Thanks in advance for your feedback!

Harm
---
Harm Otten, PhD

No comments:

Post a Comment