ALSEBO

Getting Started

  • Installation
    • Requirements
    • Standard install
    • Virtual environment (recommended)
    • Dependencies
    • Verify the installation
    • Building the documentation locally
  • Quickstart
    • Assumed directory layout
    • Step 1 — Featurize the generated sequence space
    • Step 2 — Sample a diverse initial training set
    • Step 3 — Record initial experimental results
    • Step 4 — Run Bayesian Optimisation to select the next batch
    • Step 5 — Append new results and repeat
    • Multi-objective example

User Guide

  • User Guide
    • Overview
      • What is ALSEBO?
      • The ALSEBO pipeline
      • Modules at a glance
      • Choosing a featurization method
      • References
    • VAE Latent Landscape
      • Conceptual background
      • Architecture
      • Training
      • Generating the sequence space
      • Latent landscape as a functional map
      • References
    • Sequence Space
      • Featurization methods
        • DCA — Direct Coupling Analysis
        • ESM — Protein Language Model Embeddings
        • Latent coordinates
      • Output format
    • Training Space
      • Why diversity matters
      • Sampling strategy
      • Recording experimental results
      • The obj_config dictionary
    • Bayesian Optimisation
      • The surrogate model — Gaussian Process Regression
      • The acquisition function — UCB
      • Tuning beta
      • Appending results and closing the loop
      • Round-by-round improvement

Tutorials

  • Tutorials
    • avGFP Fluorescence Optimisation with ALSEBO
      • Biological context
      • Files in this tutorial
      • Pipeline
      • Setup
      • Step 1 — Featurise the Sequence Space
      • Step 2 — Explore the Fitness Landscape
      • Step 3 — Sample a Diverse Initial Training Set
      • Step 4 — Simulate Initial Experiments
      • Step 5 — Bayesian Optimisation Loop
      • Step 6 — Results
        • Convergence
        • Optimisation Trajectory
      • Summary
        • Things to try

Project

  • Overview
    • Pipeline overview
    • Key features
    • Installation
  • Contributions & Help
    • Issue Reports
    • Documentation Improvements
    • Code Contributions
      • Submit an issue
      • Create an environment
      • Clone the repository
      • Implement your changes
      • Submit your contribution
      • Troubleshooting
    • Maintainer tasks
      • Releases
  • License
  • Authors
  • Changelog
    • Version 0.1

API Reference

  • Module Reference
    • alsebo package
      • Subpackages
        • alsebo.VAE namespace
      • Submodules
      • alsebo.optimizer module
        • acquisition_function()
        • get_next_seq_bo()
        • gpr()
        • read_seq_files()
        • save_next_batch_results()
        • seq_space_prediction()
      • alsebo.seq_space module
        • compute_dca_features()
        • generate_dca_features()
        • generate_esm_features()
        • generate_latent_features()
        • generate_seq_space()
      • alsebo.skeleton module
        • fib()
        • main()
        • parse_args()
        • run()
        • setup_logging()
      • alsebo.training_space module
        • generate_sequence_training_file()
        • sample_initial_training_sequnces()
      • Module contents
ALSEBO
  • Python Module Index

Python Module Index

a
 
a
- alsebo
    alsebo.optimizer
    alsebo.seq_space
    alsebo.skeleton
    alsebo.training_space
    alsebo.VAE
    alsebo.VAE.model
    alsebo.VAE.model.generator
    alsebo.VAE.model.layers

© Copyright 2025, dulithaprasanna.

Built with Sphinx using a theme provided by Read the Docs.