Microenvironment awareness pragmatic analysis for generative enzyme function reasoning
Abstract
Motivation: Accurate enzyme function annotation is critical for interpreting rapidly expanding sequence data, yet experimental methods are costly and low-throughput, while computational approaches face two key limitations. Most existing methods are discriminative models confined to known EC labels, making it difficult to generalize to emerging enzymatic activities under cold-start conditions. Moreover, they typically treat enzymes as isolated molecular entities and overlook microenvironment factors such as optimal pH and temperature, which provide important contextual clues for functional annotation.
Results: This study proposes Microenvironment awareness pragmatic analysis for generative enzyme function reasoning, a generative framework that integrates microenvironment as pragmatic context into enzyme function reasoning. By fine-tuning the large language model on a newly constructed Enzyme Microenvironment Database, the model learns to infer microenvironment from sequence and then reason function within that context, generating EC labels with explicit reasoning chains. Multi-task enzyme microenvironment prediction imputes missing pH and temperature values to complete the Enzyme Microenvironment Database. On public benchmark dataset, Microenvironment Awareness Pragmatic Analysis for Generative Enzyme Function Reasoning achieves an absolute F1 improvement of 0.1068 over the state-of-the-art GraphEC model. Under cold-start conditions, the model generates reference-worthy function labels, demonstrating its practical utility for emerging enzymatic activities.
Availability and implementation: The source code and data of MPA-EC can be accessed at https://bliulab.net/MPA-EC, https://zenodo.org/records/22152329 and https://github.com/bliulab/MPA-EC.
Related
- BioSecBench-Function: A Verifiable Benchmark for Reasoning about Biological Function from Experimental Data
- Bolek: A Multimodal Language Model for Molecular Reasoning
- Biochemical Characteristics of the Vascular Microenvironment in Pathological Conditions and Precision Therapeutic Strategies
- Mapping cell-type-specific spatial gene programs uncovers tissue architecture and microenvironment organization
- Quantum-sensing-based super-resolution live cell microenvironment imaging
BibTeX
@article{hao2026microenvironment,
title = {{Microenvironment awareness pragmatic analysis for generative enzyme function reasoning}},
author = {Jing Hao and Ke Yan and Bin Liu},
journal = {Bioinformatics},
year = {2026},
month = oct,
doi = {10.1093/bioinformatics/btag769},
url = {https://doi.org/10.1093/bioinformatics/btag769},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Microenvironment awareness pragmatic analysis for generative enzyme function reasoning
AU - Jing Hao
AU - Ke Yan
AU - Bin Liu
PY - 2026
DA - 2026/10/13
JO - Bioinformatics
DO - 10.1093/bioinformatics/btag769
UR - https://doi.org/10.1093/bioinformatics/btag769
AB - Motivation: Accurate enzyme function annotation is critical for interpreting rapidly expanding sequence data, yet experimental methods are costly and low-throughput, while computational approaches face two key limitations. Most existing methods are discriminative models confined to known EC labels, making it difficult to generalize to emerging enzymatic activities under cold-start conditions. Moreover, they typically treat enzymes as isolated molecular entities and overlook microenvironment factors such as optimal pH and temperature, which provide important contextual clues for functional annotation. Results: This study proposes Microenvironment awareness pragmatic analysis for generative enzyme function reasoning, a generative framework that integrates microenvironment as pragmatic context into enzyme function reasoning. By fine-tuning the large language model on a newly constructed Enzyme Microenvironment Database, the model learns to infer microenvironment from sequence and then reason function within that context, generating EC labels with explicit reasoning chains. Multi-task enzyme microenvironment prediction imputes missing pH and temperature values to complete the Enzyme Microenvironment Database. On public benchmark dataset, Microenvironment Awareness Pragmatic Analysis for Generative Enzyme Function Reasoning achieves an absolute F1 improvement of 0.1068 over the state-of-the-art GraphEC model. Under cold-start conditions, the model generates reference-worthy function labels, demonstrating its practical utility for emerging enzymatic activities. Availability and implementation: The source code and data of MPA-EC can be accessed at https://bliulab.net/MPA-EC, https://zenodo.org/records/22152329 and https://github.com/bliulab/MPA-EC.
SN - 1367-4803
SN - 1367-4811
ER -