Introduction to OpenProtein

OpenProtein.AI is a powerful platform that leverages state-of-the-art machine learning techniques to facilitate protein engineering and analysis.

We offer multiple toolsets to get a project started, including:

Each of these workflows offers unique functionalities to help you engineer better proteins!

We offer our tools through a web application or a suite of accessible APIs. Both the app and API suite contain the same functionality, so that it can suit any workflows.

Starting your first project

When embarking on the first project, it is essential to choose the appropriate module based on the dataset.
To create a design, you will need:
  • Dataset with mesurements
  • Target design objectives
To use PoET, you will need:
  • Seed sequence
  • Query (optional)

For users with functional measurements, who want to extrapolate sequence - function relationships using OpenProtein’s suite of machine learning tools, this is the place to start. Projects enable visualizing data, use existing data to predict novel sequence properties, and design new sequences with user-specified constraints. To learn how to initiate a project, please refer to our detailed step-by-step tutorial.

For users who want to start de novo, without wet lab measurements, use our generative PoET workflow to score and generate sequences. PoET infers sequence fitness based on evolutionary fitness from a user-defined prompt, most commonly a multiple sequence alignment. For more details on PoET, please refer to our detailed step-by-step tutorial.

To learn how to interact with the platform’s functionality via API, view more details here. The suite of API allows integrating OpenProtein.AI into specific applications or workflows which also allows access to features such as project management, PoET capabilities, and data analysis.

What’s next?

For the next steps, initiate a project or run a job via PoET. If you are looking for more in depth information, take a look at our walkthrough section and learn how to do more specific things like exploring a dataset, setting design criteria, and defining prompts.