search

isblab / desmosome

Code repository for the Desmosome project
Creative Commons Attribution Share Alike 4.0 International
0 stars 0 forks source link

readme

DOI

Integrative structure of the desmosome outer dense plaque

This repository is of the integrative model of the desmosome ODP based on data from X-ray crystallography, electron cryo-tomography, immuno-electron microscopy, yeast two-hybrid experiments, co-immunoprecipitation, in vitro overlay, in vivo co-localization assays, in-silico sequence-based predictions of transmembrane and disordered regions, homology modeling, and stereochemistry information. It contains input data, scripts for modeling and results including bead models and localization probability density maps. The modeling was performed using IMP (Integrative Modeling Platform).

Depositions in PDB and ModelArchive

PDB entry for the integrative model : 9A8U

Model archive entries for Alphafold3 models: Full ODP, PG-DSC1,PG-DSC1, and DP-PKP1

Image

Layout

There are 3 main folders: input, scripts, results. There is an explanation README for the folder-specific contents in each of the folders (input, scripts, results) and the general protocol is explained below.

Additional data is uploaded in Zenodo (the set of major cluster models corresponding to the main modeling run presented in the paper).

Protocol Summary

The below description applies to the main run (epithelial ODP) as well as two auxillary runs (epithelial ODP with PKP3, basal ODP). For preprocessing of the data (homology modeling, GMM generation, etc) see inputs README.

Sampling

This step involves employing MCMC sampling to generate the IMP models of ODP based in input information. The steps are:

  1. Compile IMP with the custom Desmosome module (scripts/modeling/custom_IMP_module). See the instructions for compilation in the official guide
  2. The next step requires the data to be present in the working directory. Simply create a data directory containing all the FASTA and PDB files as well as the GMM text files and the original .mrc files in a subdirectory data/gmm. (see scripts README for details on this setup).
  3. To run the sampling of each of the three aforementioned ODPs, run scripts/modeling/main_run/all_run.sh (or other_runs/{specific_run}/all_run.sh) which uses GNU Parallel to run the corresponding server_run.sh across a set of servers listed as a server list. For a single server run, a possible commands is seq 0 44 | parallel '{path_to_script}/server_run.sh {}'.
  4. The output is generated in separate folders named output{number} which are all collected for the next steps (in a multi-server setting).

Analysis

This step involves getting the good scoring models, running Sampcon (sampling exhaustiveness), and analyzing the major clusters (generate contact maps, plots etc). These are accomplished by the following steps:

  1. Collect all the output{number} folders generated above in a single directory
  2. Create analysis_output directory
  3. Run scripts/analysis/main_run/extract_all.sh. Ensure that the paths used in the script (for example, the path to the IMP-installation setup script) are correct.
  4. All the output is generated in analysis_output directory.

Alphafold results

Finally, we also run Alphafold3 on all pairs of desmosomal proteins as well as on the entire ODP. The best model and corresponding stats file are stored for each job. See scripts README for the analysis of confidently predicted interface residues.

Versions and Requirements

All of the modeling and analysis was done in a multi-server setup with Linux Fedora using Bash scripts. Pre-processing (homology modeling, tomogram processing) was done in Windows 10. The python libraries (and their versions) used in the project are as follows:

  1. matplotlib (3.6.2)
  2. numpy (1.21.5)
  3. scipy (1.9.3)
  4. IMP (2.17)
  5. mrcfile (1.4.3)
  6. tqdm (4.62.0)
  7. biopython (1.79)

Additional Linux software needed to run all the scripts in the repository:

  1. GNU Parallel
  2. OpenMPI

Information

Author(s): Satwik Pasani, Kavya S Menon, Shruthi Viswanath\ Date: August 12th, 2024\ License: CC BY-SA 4.0 This work is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License. Last known good IMP version: Not tested
Testable: Yes Parallelizeable: Yes\ Publications: Pasani S, Menon KS, Viswanath S (2023). The molecular architecture of the desmosomal outer dense plaque by integrative structural modeling. DOI.