Add notebook for running OpenFE with OF3 generated inputs

[hannahbaumann]

This is a first pass at a SepTop notebook that used OF3 generated protein-ligand structures.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[github-actions]

👈 Launch a Colab session on branch of3_septop_example

[IAlibay]

Here is my proposal for how we can structure this:

Title: OpenFE binding affinity predictions from OpenFold3 structures

0. Introduction

• Why and What are we presenting in this cookbook?
• Provide a brief overview

1. Cofolding a set of TYK2 ligands

• Technically we only need one, but several is ok too.

2. Processing the OpenFold3 outputs for use with OpenFE

• Aligning (make it clear this is not necessary for ABFEs)
• Protonating
• Extracting
• Adding partial charges

3. Loading structures into OpenFE Components
4. Creating an ABFE Transformation
5. Creating a SepTop RBFE Transformation
6. Creating a Hybrid Topology RBFE Transformation

[IAlibay]

TODO:

• Say we used the first model that was inferred. In practice you might want to use something to select amongst all the inferred models.

[atravitz]

Looks good, just some grammatical/organization suggestions!

[atravitz]

Here is my proposal for how we can structure this:

Title: OpenFE binding affinity predictions from OpenFold3 structures

0. Introduction


* Why and What are we presenting in this cookbook?

* Provide a brief overview


1. Cofolding a set of TYK2 ligands


* Technically we only need one, but several is ok too.


2. Processing the OpenFold3 outputs for use with OpenFE


* Aligning (make it clear this is not necessary for ABFEs)

* Protonating

* Extracting

* Adding partial charges


3. Loading structures into OpenFE Components

4. Creating an ABFE Transformation

5. Creating a SepTop RBFE Transformation

6. Creating a Hybrid Topology RBFE Transformation

would it make more sense to make the last 3 their own section, since they're independent of each other?

[IAlibay]

make the last 3 their own section

Sorry I'm not sure what you mean @atravitz - are they not already their own section? (I might be misunderstanding what we mean by section)

[hannahbaumann]

For step 2, maybe choose between either bullet points or a,b,c.

Also in step 2: ncessary to necessary

Reply via ReviewNB

[IAlibay]

Done thanks!

[hannahbaumann]

I don't know if it's important to add here, but for the protein sequence I did not use the full fasta sequence from uniprot, but the squence from the Xray structure we've been using in the benchmarking.

Reply via ReviewNB

[IAlibay]

Ah thanks! I'll add a bit of text to make that clearer.

[jthorton]

This is great thanks for finishing this up, LGTM!