ASE-GA

Official Resources

• Homepage: https://wiki.fysik.dtu.dk/ase/ase/ga/ga.html
• Documentation: https://wiki.fysik.dtu.dk/ase/ase/ga/ga.html
• Source Repository: https://gitlab.com/ase/ase
• License: GNU Lesser General Public License v2.1

Overview

ASE-GA is the genetic algorithm module within the Atomic Simulation Environment (ASE). It provides a flexible framework for performing global optimization of atomic structures, including clusters, crystals, and surfaces. Being part of ASE, it allows users to combine GA search strategies with any calculator (DFT or classical) supported by ASE.

Scientific domain: Genetic algorithms, structure prediction, global optimization
Target user community: Materials scientists, ASE users, method developers

Theoretical Methods

• Genetic Algorithm (GA)
• Evolutionary operators (cut-and-splice, strain, permutation)
• Population management
• Niching and diversity maintenance
• Local minimization

Capabilities (CRITICAL)

• Crystal structure prediction
• Cluster optimization
• Surface reconstruction search
• Variable composition search
• Modular design: Swap optimization methods and calculators easily
• Python-based customization

Sources: ASE documentation, J. Chem. Phys. 141, 044711 (2014)

Inputs & Outputs

• Input formats: Python script setup, initial population
• Output data types: SQLite database of structures, energy trajectory

Interfaces & Ecosystem

• ASE: Native integration
• Calculators: VASP, GPAW, LAMMPS, EMT, etc.
• Database: Uses ASE database for storing population

Workflow and Usage

1. Define reference structure (stoichiometry).
2. Initialize starting population (random).
3. Define GA operations (mating, mutation).
4. Run GA loop: Select parents -> Procreate -> Relax -> Add to DB.
5. Analyze database for global minimum.

Performance Characteristics

• Dependent on calculator speed
• Parallelization via independent relaxations
• Highly flexible

Application Areas

• Nanoclusters
• Surface alloys
• 2D materials
• Crystal prediction

Community and Support

• ASE Community
• Active mailing list
• Developed at DTU and collaborators

Verification & Sources

Primary sources:

1. ASE GA docs: https://wiki.fysik.dtu.dk/ase/ase/ga/ga.html
2. Publication: L.B. Vilhelmsen et al., J. Chem. Phys. 141, 044711 (2014)

Confidence: VERIFIED

Verification status: ✅ VERIFIED

• Website: ACTIVE
• Documentation: COMPREHENSIVE
• Source: OPEN (GitLab)
• Development: ACTIVE (ASE Community)
• Applications: GA structure prediction, ASE integration