Skip to main content
This guide will walk you through setting up your LiteFold account and running your first protein structure prediction and molecular docking experiment. You’ll go from zero to analyzing results in under 30 minutes.

​
Prerequisites

Before you begin, ensure you have:
  • A LiteFold account (sign up at litefold.ai)
  • A modern web browser (Chrome, Firefox, Safari, or Edge)
  • A protein sequence or PDB file for structure prediction
  • (Optional) A ligand molecule in SMILES or SDF format for docking

​
Create Your Account

1

Sign Up

Navigate to litefold.ai and click “Get Started” or “Sign Up”.Provide your email address and create a password. You’ll receive a verification email to activate your account.
2

Complete Your Profile

After verifying your email, log in and complete your researcher profile:
  • Organization/Institution
  • Research area
  • Primary use case (drug discovery, protein research, education, etc.)
3

Access the Platform

Once your profile is complete, you’ll be directed to the LiteFold dashboard where you can start creating projects.

​
Your First Experiment: Protein Structure Prediction

Let’s predict the structure of a protein using LiteFold’s AI-powered structure prediction engine.
1

Create a New Project

From the dashboard, click “New Project” and give it a name (e.g., “My First Structure Prediction”).Select “Structure Prediction” as the project type.
2

Upload or Paste Your Protein Sequence

You can either:
  • Paste a protein sequence in FASTA format
  • Upload a FASTA file from your computer
  • Search for a protein by UniProt ID or name
Example protein sequence (human insulin):
>Insulin_A_chain
GIVEQCCTSICSLYQLENYCN
3

Configure Prediction Settings

Choose your prediction parameters:
  • Model: Select from available AI models (AlphaFold2, ESMFold, etc.)
  • Confidence: Standard or high-confidence mode
  • Relaxation: Enable structure refinement (recommended)
For your first run, the default settings work well.
4

Run the Prediction

Click “Run Prediction” to start the structure prediction job.The platform will:
  • Process your sequence
  • Generate the 3D structure
  • Calculate confidence scores
  • Provide interactive visualization
This typically takes 2-10 minutes depending on protein size.
5

Visualize and Analyze Results

Once complete, you’ll see:
  • 3D Structure Viewer: Rotate and zoom the predicted structure
  • Confidence Scores: pLDDT scores colored on the structure
  • Secondary Structure: Helices, sheets, and loops identified
  • Download Options: PDB file, images, and reports

​
Next Step: Molecular Docking

Now that you have a protein structure, let’s dock a small molecule to it.
1

Prepare Your Ligand

From your structure prediction results, click “Add Docking Experiment”.Upload or draw a ligand molecule:
  • Paste a SMILES string (e.g., CC(=O)OC1=CC=CC=C1C(=O)O for aspirin)
  • Upload an SDF or MOL2 file
  • Use the built-in molecule sketcher
2

Define the Binding Site

You can either:
  • Auto-detect: Let Rosalind AI identify potential binding pockets
  • Manual: Specify residues or coordinates for the binding site
  • Import: Use a known binding site from the literature
For the first run, try auto-detection.
3

Run Docking

Click “Start Docking” to begin the molecular docking simulation.LiteFold will:
  • Generate ligand conformations
  • Perform rigid and flexible docking
  • Score binding poses
  • Rank by predicted affinity
4

Review Docking Results

Analyze the top-ranked poses:
  • Binding Affinity: Predicted ΔG values
  • Interaction Map: Visualize hydrogen bonds and hydrophobic contacts
  • RMSD: Clustering of similar poses
  • Export: Download poses for further analysis

​
Working with Rosalind AI

Pro Tip: Ask Rosalind for help at any stage!Click the “Ask Rosalind” button and try queries like:
  • “Show me similar proteins in the PDB database”
  • “What are known inhibitors for this target?”
  • “Set up a virtual screening experiment with 1000 compounds”
Rosalind will automatically configure experiments and fetch relevant data for you.

​
What’s Next?

Platform Overview

Explore all of LiteFold’s capabilities beyond structure prediction and docking.

Drug Discovery Workflows

Learn how to run complete drug discovery campaigns from target to candidate.

Molecular Dynamics

Validate binding predictions with MD simulations and calculate binding free energies.

De Novo Design

Generate novel molecules tailored to your protein target using generative AI.

​
Need Help?

Get Support

Our team is here to help! Reach out with questions, feedback, or if you encounter any issues.Email: support@litefold.ai

​
Community Resources

Join the LiteFold community to learn from other researchers and share your experiences: