Python-based
Hierarchical
ENvironment
for
Integrated
Xtallography
Roman Zubatyuk, Malgorzata Biczysko, Kavindri Ranasinghe, Nigel W. Moriarty, Hatice Gokcan, Holger Kruse, Billy K. Poon, Paul D. Adams, Mark P. Waller, Adrian E. Roitberg, Olexandr Isayev, Pavel V. Afonine
The purpose of AQuaRef is to refine atomic model coordinates using geometric restraints derived from AI-accelerated quantum-mechanical calculations. Real- and reciprocal-space refinements are supported using crystallographic or cryo-EM experimental data.
Atomic models resulting from AQuaRef refinement are expected to have improved geometry, as evaluated by standard validation tools such as MolProbity, Rama-Z, and hydrogen bond skew-kurtosis plots. They should also maintain a similar or improved fit to the experimental data compared to standard classic refinement.
Protein-only (plus water) models are supported at the moment. Static disorder (i.e., alternative conformations) is not supported, except for exchangeable H/D sites in case of refinement against neutron diffraction data.
If a model contains unsupported entity (e.g., a metal ion) or alternative conformations, it is still possible to refine it but unsupported entities need to be excluded using 'exclude' keyword, as shown below in Examples.
The atomic model needs to be nearly final (in the later stages of standard classic refinement) and atom-complete (e.g., all hydrogen atoms are present explicitly, no missing side chains). The model is also expected to have meaningful geometry, such as being devoid of overlapping (clashing) atoms or severely distorted covalent bonds.
Hardware: While this can possibly work on both CPU and GPU computers, currently available builds are specific to Linux machines with CUDA 11 or 12. Additionally, runtimes on CPU-based computers are expected to be much slower compared to GPUs.
No GUI: Command line only.
AQuaRef is available as part of Phenix or separately from Phenix as part of Q|R refinement project (www.qrefine.com or https://github.com/qrefine/qrefine).
A self-consistent, quck to run, simple example is distributed with the software (Phenix, Q|R) and located in phenix/modules/qrefine/examples/AQuaRef folder. Check README file for details. If using AQuaRef for the first time, we suggest to run this example first in order to make sure all works as expected.
Reciprocal space refinement:
qr.aquaref model.pdb data.mtz
Real space refinement:
qr.aquaref model.pdb map.mrc
Optimization of model geometry (no experiemntal data used):
qr.aquaref model.pdb
Reciprocal space refinement exclusing Mg and ligand ATP:
qr.aquaref model.pdb data.mtz exclude="element MG or name ATP"
AQuaRef: Machine learning accelerated quantum refinement of protein structures https://doi.org/10.1101/2024.07.21.604493