directed-evolution

A graph-based python framework for fitness landscape analysis

Directed evolution of proteins in sequence space with gradients

🧬 Fusion of protein sequence and structural information, using denoising pre-training network for zero-shot protein engineering (eLife 2025).

Protein Design by Machine Learning guided Directed Evolution

LatentDE: Latent-based Directed Evolution for Protein Sequence Design

A general harness enhances zero-shot fitness predition for any protein foundation model.

Computational model of laboratory directed evolution + experiments.

TeBaAb: Text-Based Antigen-Conditioned Antibody Redesign via Directed Evolution

Learning to Solve Multiresolution Matrix Factorization by Manifold Optimization and Evolutionary Metaheuristics

Generates randomly generated fastQ files from a template and upstream sequence.

Directed Evolution of Proteins via Bayesian Optimization in Embedding Space

Directed Evolution in Silico

GenLoop — a closed generative drug-design loop on dtSFM; directed evolution of small molecules (Reddy 2026). Generate → encoder-rerank → AlphaFold-3 verify → LoRA-refine.

Unified Python library and CLI for protein structure prediction and inverse folding.

Sequence Space Coverage Calculator

Calculates the probability of finding the top variant in a library of sequences.

Prior- and constraint-driven directed-evolution agent on the GB1 four-site landscape: seven-arm controlled comparison, auditable constraints, and an honest null result for the LLM

Active learning over real protein fitness landscapes: GP-UCB vs random under a fixed experiment budget, replicated multi-seed on GB1 and AAV2. Snakemake DAG + provenance.

A trial of whether life could be built on silicon rather than carbon — and an honest verdict of suspension.

Plan the next round of protein variant experiments from the measurements you already have: assay-specific evaluation, a stated selection rule, and reproducible exports.