Check TED/HMM matches

Check domain boundaries between HMM matches and TED consensus predictions.

The incoming data is a subset of all HMM/TED domains corresponding to UniProt accessions from human.

Currently the processing is done in a Jupyter notebook which:

• filters the incoming data based on minimum domain length
• creates bins according to overlap and evalue
• randomly selects 10 entries per bin

Visualizing the domain boundaries

python -m venv myenv
source myenv/bin/activate
pip install -r requirements.txt

For domain visualization and analysis, please refer to the Jupyter notebook: domain_visualization.ipynb

The notebook uses a custom function visualize_domains_and_plot(df) which provides an interactive visualization of domain boundaries and related metrics. This function does the following:

1. 3D Protein Structure Visualization:

   • Displays the 3D structure of a selected protein using PyMOL.
   • Colors different regions of the protein to indicate:
     • HMM-only domains (blue)
     • TED-only domains (red)
     • Overlapping HMM and TED domains (purple)
     • Regions without domain predictions (grey)

2. Interactive Scatter Plot:

   • Creates a scatter plot of HMM E-value vs. Overlap Percentage for all proteins in the dataset.
   • The x-axis (HMM E-value) uses a logarithmic scale.
   • Each point represents a protein, with the currently selected protein highlighted in red.

3. User Controls:

   • A slider and navigation buttons to move through the dataset.
   • A dropdown menu to sort the data by Index, HMM E-value, or Overlap Percentage.