Access the AlphaFold Server:
Go to https://alphafoldserver.com/.
Sign in with a Google Account:
You must log in using a Google account. If someone doesn't have one, they can work in small groups to complete the activity.
Configure the Prediction Job:
In the "Entity type" section, select protein.
Set the number of copies to 1.
In the "Input" box, paste the protein sequence using single-letter amino acid codes without any other characters.
Submit the Job:
Click "Continue and preview job".
Assign a name to your prediction.
Click "Confirm and preview job" again.
Processing Time:
The time required depends on the sequence length. For sequences of 35–50 amino acids, it typically takes about 1 minute. However, processing time may vary based on your internet connection and computer performance.
Viewing Predictions:
All submitted predictions will appear in a list at the bottom of the page. Once a prediction is complete, click on its name to view the results.
Access the Visualization Screen:
After the prediction is complete, it will be listed at the bottom of the page. Click on the prediction name to access its visualization.
Interpret the Protein Model:
In the visualization screen, you'll see:
3D Structure: The protein's three-dimensional structure is color-coded based on prediction confidence:
Blue: High confidence.
Yellow: Medium confidence.
Orange/Red: Low confidence, indicating less reliable regions.
plDDT (Predicted Local Distance Difference Test): This metric indicates the confidence level for each residue's position.
PAE (Predicted Aligned Error) Plot: This plot helps assess the confidence in the spatial relationship between residues:
Axes X and Y: Represent the protein's residues.
Dark Green (Low PAE values): High confidence in the relative positioning of residues.
Lighter Colors at Edges: Suggests increased uncertainty in some domain arrangements.
Entirely Dark Green Box: Indicates a highly confident prediction with well-defined domain orientations.
Interactive Features: You can rotate and zoom in on the 3D structure to examine specific details.
Download the Results:
Various download options are available:
PDB File: For analyzing the structure with other molecular modeling tools.
PAE Plot: Visual representation of prediction accuracy for each residue.
Ranking File: Displays the confidence ranking of the generated models. The program produces five different models, assigning them indices and ordering them from highest to lowest probability.
Advanced Options (If Necessary):
In-Depth Analysis: For more detailed examination, you can load the PDB file into programs like PyMOL, Chimera, or VMD
Re-running Predictions: To rerun a prediction with modifications, click "Clone and Reuse" at the top of the page.
Note on Usage Limits:
Please be aware that with this open server, each Google account has a daily limit on the number of predictions it can perform. You can check your remaining quota at the top right corner under "Remaining jobs".