Add intermediate structure metrics module (TM-score, lDDT)#36
Open
sherrylicodes wants to merge 1 commit into
Open
Add intermediate structure metrics module (TM-score, lDDT)#36sherrylicodes wants to merge 1 commit into
sherrylicodes wants to merge 1 commit into
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This pull request contributes to issue #9 by adding evaluation and visualization components for intermediate structure prediction in OpenFold. Specifically, I implemented TM-score and lDDT (mean + per-residue) metrics that compare generated intermediate structures against reference structures extracted during inferences. These metrics will provide feedback on the quality of intermediate predictions and are designed to integrate well with future visualization tools.
The new module can be used as part of a reproducible pipeline that runs:
Evoformer extraction → structure generation → structure evaluation.
It allows for quick experimentation and benchmarking across different intermediate layers.
I am working with Alexis Foster and Archible Sherman. Alexis create test_intermediate_structure.py, which runs a three-recycle forward pass through the Evoformer, extracts intermediate MSA and pair representations. Archible worked on the web interface that will visualize these intermediate structures and the evaluation metrics.
In intermediate_metrics.py, I added the StructureMetrics dataclass, TM-score + lDDT computation functions, and PDB-based comparison utilities.