Quickstart Guide

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

​

Sign In to LiteFold

After creating your account, sign in to the LiteFold workspace. The workspace serves as the central location for all projects, experiments, generated structures, docking studies, simulations, and design jobs.

​

Create Your First Project

All work in LiteFold is organized into projects. Select New Project from the sidebar and enter a project name. Projects provide a shared workspace for structures, molecules, simulations, and analysis results. After creation, the project becomes available throughout the platform and can be selected when launching new experiments.

​

Upload Files

Before starting an experiment, upload the files required for your study. Open the Files section from the sidebar and upload any supported research assets, including:

• FASTA sequences
• Protein structures (PDB, CIF)
• Ligands (SDF, MOL2)
• Simulation outputs
• Supporting datasets

Uploaded files are stored within your workspace and become available across LiteFold tools.

​

Choose a Research Workflow

From the Lab workspace, select the workflow that matches your research objective.

​

Structure Prediction

Predict protein, protein-complex, protein-ligand, protein-DNA, or protein-RNA structures from sequence data.

​

Molecular Docking

Identify binding modes and estimate binding affinity between ligands and biological targets.

​

DeNovo Design

Generate new molecules, peptides, proteins, or aptamers optimized for a target structure.

​

Dynamo

Run molecular dynamics simulations to evaluate stability, flexibility, and molecular interactions over time.

​

Rosalind AI

Use LiteFold’s AI co-scientist to search databases, interpret results, suggest experiments, and automate research tasks.

​

Run Your First Experiment

Every LiteFold workflow follows a similar pattern:

1. Select a project.
2. Choose input files.
3. Configure experiment parameters.
4. Review settings.
5. Submit the job.

The platform automatically provisions compute resources and tracks progress through the experiment dashboard.

​

Monitor Progress

Running experiments appear in their respective workflow pages. Depending on the workflow, LiteFold provides:

• Job status tracking
• Progress indicators
• Live logs
• Intermediate outputs
• Downloadable results

Completed jobs remain associated with the project for future analysis.

​

Analyze Results

Each LiteFold workflow includes dedicated analysis tools. Structure prediction results provide confidence metrics and interactive structure visualization. Docking experiments provide binding poses, interaction maps, and affinity scores. DeNovo design workflows provide generated candidates along with drug-likeness and docking evaluations. Molecular dynamics simulations provide trajectory analysis, RMSD, RMSF, hydrogen-bond occupancy, and energy profiles.

​

Work with Rosalind AI

Rosalind can assist throughout the research process. Researchers can ask Rosalind to:

• Search scientific literature
• Retrieve known ligands and targets
• Suggest experimental strategies
• Configure workflows
• Interpret computational results
• Recommend follow-up studies

Rosalind integrates directly with LiteFold projects and can access the context of your ongoing research. 

​

Recommended First Workflow

For new users, the following sequence provides a complete introduction to the platform:

1. Upload a protein structure.
2. Run Molecular Docking.
3. Analyze binding interactions.
4. Generate optimized candidates using DeNovo Design.
5. Validate promising molecules with Dynamo simulations.
6. Use Rosalind to summarize findings and suggest next experiments.

This workflow demonstrates how LiteFold connects prediction, design, simulation, and analysis within a single research environment.

​

Need Help?

​

Community Resources

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

• Blog: Read the latest research and tutorials at litefold.ai/blog
• HuggingFace: Access our models and datasets at huggingface.co/LiteFold
• Publications: Explore our research at litefold.ai/research