Skip to content

Latest commit

 

History

History
107 lines (61 loc) · 5.49 KB

README.md

File metadata and controls

107 lines (61 loc) · 5.49 KB

DOI

mdx-examples

Workflows for diffuse scattering data reduction and lattice disorder modeling with MATLAB and mdx-lib.

The lysozyme examples reproduce the workflow described in Meisburger, Case, & Ando, 2022.

Version history

Version 0.1.1

Version to accompany the publication of our manuscript in Nature Communications.

Version 0.1.0

Current development version, accompanies our preprint on bioRxiv.

Repository Contents

Each example consists of a series of MATLAB scripts (starting with job##) that are run in numerical order. A "setup" job downloads source data from the SBGrid databank, and all other input files are included in the repository (such as XDS input files, structure factor files, and atomic coordinate files).

In each example, the workflow is divided into two parts organized within subfolders of this repository. In subfolders ending in _map, the diffraction images are processed to produce a three-dimensional map of diffuse scattering. In subfolders ending in _model, a lattice disorder model is fit to the diffuse scattering, and the data are further analyzed following the workflow described in Meisburger, Case, & Ando, 2022.

Example 1: Triclinic lysozyme

The dataset for lysozyme in the triclinic space group (P1) is described in Meisburger, Case, & Ando, 2020 and in Meisburger, Case, & Ando, 2022. Diffraction images are in the SBGrid databank (SBGrid 747), and the structure has been deposited in the Protein Data Bank (PDB 6o2h).

Example 2: Orthorhombic lysozyme

The dataset for lysozyme in orthorhombic space group (P212121) is described in Meisburger, Case, & Ando, 2022. Diffraction images are in the SBGrid databank (SBGrid 958), and the structure has been deposited in the Protein Data Bank (PDB 8dz7).

Example 3: Tetragonal lysozyme

This dataset for lysozyme in tetragonal space group (P43212) is described in Meisburger, Case, & Ando, 2022. Diffraction images are in the SBGrid databank (SBGrid 957), and the structure has been deposited in the Protein Data Bank (PDB 8dyz).

Requirements

Software

  • Unix-like operating system (tested on Mac)
  • MATLAB and the following toolboxes installed:
    • Parallel Processing Toolbox is used, but can be disabled by setting useParallel to false in individual scripts
    • Statistics and Machine Learning Toolbox
  • mdx-lib, version 1.2

The examples have been tested using MATLAB version R2021a on Mac OS 10.14

Hardware

The _map examples can make use of multi-core processing. The required RAM can be very large, especially when building the fine maps. The examples were tested on a desktop computer with 64 Gb of RAM and a 10-core CPU. On the test system, each map example required several hours to run.

The _model examples are less resource intensive, however sufficient RAM is needed for storing the diffuse maps. On the test system, each example required about 15 minutes.

Instructions

After downloading or cloning this repository, open MATLAB and navigate to the example directory (such as lys_tri_map). Then,

  1. Add mdx-lib directory to the MATLAB path

  2. Run the jobs in numerical order.

Currently, the _model jobs require that you have run the corresponding _map job first, since the experimental maps are needed for modeling. If you want to skip the _map step, the processed maps can be downloaded from a public repository (not yet: coming soon).

Notes:

  • In general, the job m-files should be inspected before they are run, in case there are variables that need to be modified (such as useParallel).
  • Additional information can be found in the m-file comments
  • Jobs can be run all at once, or step-wise using cell mode in the editor.
  • The workspace can be cleared between jobs. Information that needs to be carried over is saved in files, typically in the proc or export subdirectories.

Expected Results

The final modeling job generates a report summarizing the results. Reports are included here:

References

Meisburger, SP, Case, DA & Ando, N. (2020). Diffuse X-ray scattering from correlated motions in a protein crystal. Nat Commun 11, 1271. https://doi.org/10.1038/s41467-020-14933-6

Meisburger, SP, Case, DA & Ando, N. (2022). Robust total X-ray scattering workflow to study correlated motion of proteins in crystals. bioRxiv 2022.08.22.504832; doi: https://doi.org/10.1101/2022.08.22.504832