Welcome to OpenProtein’s documentation!

Welcome to OpenProtein.AI!

OpenProtein.AI is a powerful platform that seamlessly integrates state-of-the-art machine learning and generative models into protein engineering workflows.

We help you to design better proteins faster by giving you access to cutting-edge protein language models, prediction and design algorithms, as well as data management and model training tools. Easily build and deploy high-performance Bayesian protein function predictors, or apply generative protein language models to design sequence libraries, all via our integrated platform. OpenProtein.ai can be accessed via web App, API and this python client, making it great for both biologists and protein engineers, as well as data scientists and software engineers.

The documentation is divided into workflows below. For each workflow you can read the docs and see a demo of usage.

Table of Contents

Workflow	Description
Installation	Install guide for pip and conda.
Session management	An overview of the OpenProtein Python Client & the asynchronous jobs system.
Asssay-based Sequence Learning	Covers core tasks such as data upload, model training & prediction, and sequence design
De Novo prediction & generative models (PoET)	Covers PoET, a protein LLM for de novo scoring, as well as sequence generation.
Protein Language Models & Embeddings	Covers methods for creating sequence embeddings with proprietary & open-source models

Tutorials

We have a range of tutorials to get you started:

Quick-start Guides

• Sequence-based learning

• Sequence embeddings

• Getting started with Poet

Case-studies

• Designing new Chorismate mutase enzymes with PoET notebook

• Visualizing antibody embeddings

• PoET: Aliphatic amidase sequence analysis (substitutions and deletions)

Index

• Installation
  • pip
  • conda
  • Github
• Session management
  • OpenProtein login
  • OpenProtein Job System
  • OpenProtein API session
• Asssay-based Sequence Learning
• De Novo prediction & generative models (PoET)
• Protein Language Models & Embeddings