# ScorpioLea/AiCE

Predicting high-fitness mutations based on protein inverse folding models

Repository: https://github.com/ScorpioLea/AiCE
Canonical: https://ross.abutalabs.com/products/aice
Language: Jupyter Notebook
License: NOASSERTION
License Family: other
Last push: 2025-09-30T03:16:29+00:00

## Health v2 (maintenance only)
Score: 40/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 44, release rhythm 35, longevity 41
- inputs: {"age_days": 580, "days_push": 337, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1144, forks 186 (observed 2026-08-28T04:03:45.139018+00:00)

## What it is
AiCE is a Python tool that predicts high-fitness protein mutations by sampling sequences from protein inverse folding models such as ProteinMPNN, LigandMPNN, ESM-IF1, and SaProt. It integrates structural and evolutionary constraints to nominate mutations that optimize protein function.

## Use cases
- predict high-fitness mutations in a protein
- nominate mutations for directed evolution experiments
- optimize protein function using inverse folding models
- build LD and SCA coupling matrices from sampled sequences
- predict multi-mutation combinations for a monomeric protein structure

## When to choose
- you have a protein structure (PDB) and want AI-nominated beneficial mutations
- you already use ProteinMPNN, LigandMPNN, ESM-IF1, or SaProt and want mutation prioritization
- you need structural and evolutionary constraints combined for protein engineering

## When to avoid
- you need mutation prediction for protein complexes without modifying the provided scripts
- you lack a protein 3D structure for your target
- you need a general-purpose protein language model rather than mutation nomination

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, data-science
- domain: bioinformatics, machine-learning, artificial-intelligence
- platform: python
- tags: protein-engineering, inverse-folding, mutation-prediction, computational-biology, proteinmpnn, directed-evolution, jupyter-notebook, linux, macos

## Member repositories
- ScorpioLea/AiCE (main) score 40

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:45.139018+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:34:43.274076+00:00, confidence not recorded.
  - readme: https://github.com/ScorpioLea/AiCE (fetched 2026-08-28T04:03:45.139018+00:00, sha 99ba14b5893a)
- Data as of 2026-08-30T08:39:29.467469+00:00.