Find optimal docking parameters using Hex

[rvanasa]

Docking simulators give a useful although rather janky metric for how an antibody would potentially bind to the target antigen. Most of these tools generate thousands of possible configurations and then leave the user to perform some sort of post-processing such as energy minimization to retrieve useful information from the simulation.

We are focusing on three different docking tools:

• Hex: http://hex.loria.fr/dist/index.php
• Frodock: http://frodock.chaconlab.org/
• ZDock: http://zdock.umassmed.edu/software/

Each of these has its own benefits and drawbacks. Hex is very fast and relatively easy to use; Frodock is state-of-the art but somewhat slower; and ZDock is well-established but relatively inaccurate and slow.

The goal is to create a pipeline where we can automatically dock antibodies to our target and evaluate how much the three simulators agree with each other. Configurations with the most agreement across simulations have been demonstrated to be much more reliable in their contact predictions.

The idea is that we feed the contact points from this pipeline into the deep learning model, which returns a score based on how strong the docking configuration would be in real life.

Fine-tuning the docking simulations is going to require learning how proteins and ligands interact with each other, so this is a perfect entry point for anyone who wants to dig into the microbiological side of the project.

How to start:

• Download Hex (http://hex.loria.fr/dist/index.php)
  • Available for Windows/Linux/Mac
• Acquire some experimental data
  • I would recommend downloading 6W41.pdb (https://www.rcsb.org/structure/6W41) because this is an actual antibody that binds to COVID-19
  • Split the antibody-antigen pair so you can re-dock them
  • One way of doing this is using pdb-tools (https://pypi.org/project/pdb-tools/)
    • $ pip install pdb-tools
    • $ pdb_selchain -H,L 6w41.pdb > receptor.pdb
    • $ pdb_selchain -C 6w41.pdb > ligand.pdb
  • If you just want the pre-split coronavirus PDB files, I'll drop them into the Slack chat for you
• Run a docking simulation
• Save and open the docked PDB file online (https://www.ncbi.nlm.nih.gov/Structure/icn3d/full.html)
• Experiment with the docking parameters to get the simulation to match the real-life configuration

Common terminology:

• Receptor: antibody
• Ligand: antigen (in the context of the docking simulators)
• Light / Heavy chains: the two protein sequences which comprise the antibody (they sort of twist around each other)
• CDR: specific parts of the antibody which bind to the antigen (the tips of the light/heavy chains)
  • Docked solutions should pretty much always bind to these regions
• Epitope: contact point on the antibody
• Paratope: contact point on the antigen

Note that in most simulators, the antibody is called the "receptor" while the antigen is called the "ligand." If you need any other clarification on terminology, be sure to let me know since other people will probably run into it as well.

Here's a bit of documentation for Hex: http://hex.loria.fr/manual800/hex_manual.pdf

[imcsherry]

Currently learning to use Hex. I've done a little bit of messing around and didn't get super far on anything productive - hopefully I can have some useful data in the next couple days

[rvanasa]

Awesome; thanks for having a go at it! Definitely keep me posted.

Ideally, we want to find a way to record the most common contact points for all the 1000+ docking poses that Hex generates on each run. That way, we can overlap this with other systems to avoid predicting weird contact points (like in the center of the antigen, etc).

Let me know if you find something related to this, or if you come across anything else interesting / cool in the software.