Experiences developing structural biology tools with AI coding assistance
- 1Molecular Biophysics and Integrated Bioimaging Division Lawrence Berkeley National Laboratory Berkeley California USA
- 2Department of Bioengineering University of California Berkeley Berkeley California USA
Abstract
Large language models (LLMs) offer significant potential to accelerate scientific software development, yet their application to structural biology algorithms requires rigorous evaluation. This challenge is examined through the development of xcif, a new native parser for the crystallographic information file (CIF) format, designed to replace a stable but slow parser (ucif) in the Computational Crystallography Toolbox (cctbx). Using an expert-led, artificial intelligence (AI)-assisted workflow, implementation time was reduced from an estimated several months to approximately one week of active human effort. However, this efficiency relied entirely on domain knowledge to specify constraints, intercept plausible but flawed machine-generated code and design adversarial tests. The machine assistance did not replace programming expertise but rather shifted the human engineering burden from writing code to independent validation. Applying the new xcif parser across a corpus of approximately 550 000 archive files confirmed that it produces output identical to both the legacy implementation and GEMMI, a modern high-performance parser. xcif parses roughly an order of magnitude faster than the legacy cctbx parser and 1.3–1.5 times faster than GEMMI, while reducing peak memory use for larger files. This case study demonstrates that AI assistance provides a highly practical, cost-effective pathway for the renewal of mature structural biology infrastructure, provided that the development is strictly governed by expert oversight and objective verification.
Related
- Theoretical Framework for Predicting Dynamic Covalent Exchange Based on Proton Affinities and Proton‐Transfer Assistance
- Developing Rapid Urine-Penetrating, Self-Drying, and Breathable Materials for Disposal Diapers
- Transmission-type coding metasurface for spatially tailored Rydberg-atom microwave sensing
- The crystalline style of molluscs: the nonlinear dynamics of a macroscale rotating structure in biology.
- Developing An Entirely Mechanochemical Synthesis of Cram’s Cavitands and Carcerands
BibTeX
@article{sobolev2026experiences,
title = {{Experiences developing structural biology tools with AI coding assistance}},
author = {Oleg V. Sobolev and Dorothee Liebschner and Paul D. Adams},
journal = {Acta Crystallographica Section D: Structural Biology},
year = {2026},
month = oct,
doi = {10.1107/s2059798326010211},
url = {https://doi.org/10.1107/s2059798326010211},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Experiences developing structural biology tools with AI coding assistance
AU - Oleg V. Sobolev
AU - Dorothee Liebschner
AU - Paul D. Adams
PY - 2026
DA - 2026/10/19
JO - Acta Crystallographica Section D: Structural Biology
DO - 10.1107/s2059798326010211
UR - https://doi.org/10.1107/s2059798326010211
AB - Large language models (LLMs) offer significant potential to accelerate scientific software development, yet their application to structural biology algorithms requires rigorous evaluation. This challenge is examined through the development of xcif, a new native parser for the crystallographic information file (CIF) format, designed to replace a stable but slow parser (ucif) in the Computational Crystallography Toolbox (cctbx). Using an expert-led, artificial intelligence (AI)-assisted workflow, implementation time was reduced from an estimated several months to approximately one week of active human effort. However, this efficiency relied entirely on domain knowledge to specify constraints, intercept plausible but flawed machine-generated code and design adversarial tests. The machine assistance did not replace programming expertise but rather shifted the human engineering burden from writing code to independent validation. Applying the new xcif parser across a corpus of approximately 550 000 archive files confirmed that it produces output identical to both the legacy implementation and GEMMI, a modern high-performance parser. xcif parses roughly an order of magnitude faster than the legacy cctbx parser and 1.3–1.5 times faster than GEMMI, while reducing peak memory use for larger files. This case study demonstrates that AI assistance provides a highly practical, cost-effective pathway for the renewal of mature structural biology infrastructure, provided that the development is strictly governed by expert oversight and objective verification.
SN - 2059-7983
ER -