Skip to main content
Now that you have signed in and understand the interface, you’re ready to start your first research session. This guide walks you through your first research session, from creating a project to running your first analysis.

​
Creating Your First Project

When you first sign in to OpenBio, you’ll be prompted to create your first project. Projects help you organize your research into separate workspaces.
1

Create a project

Create a project to organize your research:
  • Click the “New Project” button in the project selector
  • Type your project name and description (optional)
  • Choose a color for your project
  • Click “Create Project”
What is a Project?A project in OpenBio is a workspace that contains:
  • Chat conversations
  • Uploaded files
  • Notes and todos
  • Prediction jobs
All your research data is organized within projects.
You can create multiple projects to organize your research into different areas. And you can switch between projects by clicking the project selector in the sidebar.

​
Starting Your First Chat

The heart of OpenBio is the AI-powered chat interface. Let’s start your first conversation. After creating a project, you will be redirected to the project dashboard. On the right side of the screen, you will see the chat panel. Here you can start your first conversation with OpenBio.
1

Type Your First Question

In the chat input box in the right side of the screen, type a question or request. Here are some examples to get you started:
2

Send Your Message

Press Enter or click the Send button (arrow icon) to send your message.
Press Shift + Enter to add a new line in your message without sending it. This is useful for formatting longer queries.
3

Review the Response

OpenBio will process your request and you’ll see a few things:
  • Reasoning process
  • Tool calls (you can click on them to see the details)
  • Intermediate explanations of what OpenBio is doing
  • Interactive visualizations
    When applicable, visualizations will appear in the research panel in the center and you can select the tab both from the tool call or the right sidebar
  • Final response
The response appears in real-time as it’s generated, so you can see the AI “thinking” through your request.
You stop OpenBio anytime you want by clicking on the stop button in the chat input box and guide OpenBio better

​
Your First Research Task

Let’s complete a simple research task together.
1

Search for a Protein

In the chat, type:
Show me information about the protein insulin
OpenBio will search UniProt and return detailed information about insulin.
2

View the Structure

After getting the results, ask:
Show me the 3D structure of human insulin
OpenBio will query the PDB database and display an interactive 3D visualization of insulin’s structure in the research panel on the right.
3

Take Notes

OpenBio automatically takes notes for you as you go. You can also ask OpenBio to take notes for you by typing telling it to take notes for you.
You should be specific about what notes you want to take.
You can also add your own notes by clicking on the “Notes” tab in the research panel.
4

Upload a File (Optional)

If you have a protein sequence file (FASTA format):
  1. Click the “Files” section in the sidebar
  2. Click the “Upload” button (cloud with up arrow icon)
  3. Select your file
  4. Once uploaded, you can reference it in chat by typing @ followed by the filename (the suggestions show up as you type filenames)

​
Selecting AI Models

1

Choose an AI model

OpenBio supports multiple AI models for different use cases:
  1. Look at the bottom-left of the chat input box
  2. Click on the model name (default is “Gemini Flash 2.5”)
  3. A dropdown menu will appear with available models:
    • Claude Sonnet 4: Best for tasks that a smaller model like Gemini Flash cannot handle. We recommend using Sonnet 4 for workflows that require you to run predictions and complex reasoning.
    • Gemini Flash 2.5: Best for general research tasks, balanced performance, very fast. We recommend using Gemini Flash 2.5 for most research tasks.
    • We also support other models like GPT-4.1, Kimi K2, and GLM-4.6.
Different models have different capabilities and response speeds. You should try out various models to find out what works best for your prompting style and workflows

​
Tips for Success

The more specific you are, the better results you’ll get. Instead of “show me proteins,” try “show me human kinase proteins involved in cell signaling.”
OpenBio remembers the context of your conversation. You can ask follow-up questions like “show me more details” or “what about similar proteins?”
Upload your data files and reference them in conversations using @filename. OpenBio will read and analyze your files.
Use the Notes feature to save important findings. Add tags to organize notes by topic.
Once you find a useful sequence of steps, save it as a workflow to reuse later. Check out workflows for more information.

​
What’s Next?

Now that you’ve completed your first research session, explore these guides:

Interface Guide

Deep dive into all interface elements

File Management

Learn to organize and use files

Predictions

Learn about predictions

Workflows

Create reusable workflows

Notebooks

Organize your research notes

Tutorials

Step-by-step task guides
Remember: OpenBio automatically saves your work as you go. You don’t need to worry about manually saving conversations, notes, or files.