add_LLM_add_GUI_add_dmg

.gitignore

@@ -147,3 +147,5 @@ testSQL.py
 /static/
 /templates/
 .idea/
+/sboannotator.dmg
+/src/ml_sbo/models/stage1_80_stage2_10/pytorch_model.bin

README.md

@@ -11,75 +11,64 @@
 [![PyPI downloads](https://img.shields.io/pypi/dm/SBOannotator.svg)](https://pypistats.org/packages/SBOannotator)
 [![DOI](https://img.shields.io/badge/DOI-10.1093%2Fbioinformatics%2Fbtad437-blue.svg?style=plastic)](https://doi.org/10.1093/bioinformatics/btad437)
 
-Developers : [Nantia Leonidou](https://github.com/NantiaL) & Elisabeth Fritze
+Developers : [Nantia Leonidou](https://github.com/NantiaL) & Elisabeth Fritze& Jiahui Hu
 ___________________________________________________________________________________________________________
 
-### How to cite the SBOannotator?
-
-The SBOannotator is described in this article: https://doi.org/10.1093/bioinformatics/btad437
 
 ### Overview
-SBOannotator is the first standalone tool that automatically assigns SBO terms to multiple entities of a given SBML model. 
-The main focus lies on the reactions, as the correct assignment of precise SBO annotations requires their extensive classification. 
-Our implementation does not consider only top-level terms but examines the functionality of the underlying enzymes to 
-allocate precise and highly specific ontology terms to biochemical reactions. 
-Transport reactions are examined separately and are classified based on the mechanism of molecule transport. 
-Pseudo-reactions that serve modeling purposes are given reasonable terms to distinguish between biomass production and the 
-import or export of metabolites. Finally, other model entities, such as metabolites and genes, are annotated with appropriate terms. 
-Including SBO annotations in the models will enhance the reproducibility, usability, and analysis of biochemical networks.
+This project transforms SBOannotator from a static, hard-coded tool into a dynamic, intelligent system for annotating Systems Biology Ontology (SBO) terms in SBML models. The enhanced system integrates real-time SBO term retrieval, multiple enzymology data sources, and LLM-assisted annotation—significantly improving accuracy and usability while preserving the core rule-based strengths. A standalone desktop GUI with interactive visualization makes powerful annotation capabilities accessible to a broader community of systems biology researchers.
+The system delivers three key innovations:
+- Automated GitHub integration for real-time SBO file updates;
+- A three-layer rule-based annotation workflow based on 4 database(bigg, kegg, reactome, seed);
+- A finetuned LLM achieving 94% accuracy to predict SBO terms within 42 candidates for EC number .
 
-### Web Application
-Web application hosted at [TueVis](https://tuevis.cs.uni-tuebingen.de/sboannotator/) is accessible and ready to use at [sbo-annotator-tuevis.cs.uni-tuebingen.de/](https://sbo-annotator-tuevis.cs.uni-tuebingen.de/)
+### Input data
++ an SBML document
++ a sbo terms (.json or obo)(optional)
 
-### Installation
-```
-pip install SBOannotator
-```
 
-### Prerequisites
+### Outputs
++ Rule Based Enhanced Annotated libSBML model
++ LLM Annototed libSBML model
++ Uptodate SBO terms (.json) 
 
+### Usage with terminal
 This tool has the following dependencies:
 
-python >=3.8.5
+**Python** >= 3.9.6
 
-Packages:
-* sqlite3
-* libsbml
-* collections
-* requests
-* json
-* time
+**Packages:**
+- sqlite3
+- libsbml
+- collections
+- requests
+- json
+- time
 
-### Input data
-+ `doc`: an SBML document
-+ `model_libsbml`: SBML model of interest
-+ `modelType`: type of modelling framework (see below)
-+ `database_name`: name of imported database, without extension
-+ `new_filename`: file name for output model
-
-Types of modelling framework accepted:
-- constraint-based
-- logical
-- continuous
-- discrete
-- hybrid
-- logical
+**Trained Model (423MB):**
+- Download pretrained model: https://drive.google.com/file/d/1Kypb5YmLKUbFY9tZuzk0p1Mn_FWaD0g_/view?usp=drive_link
+- Place the downloaded file at: `src/ml_sbo/models/stage1_80_stage2_10/pytorch_model.bin`
 
-### Outputs
-+ `model_libsbml`: Annotated libSBML model
+**Run**
+- `pip install -r requirements.txt`
+- `python main.py` in the command line within the project folder. If ERROR occurs, check the current version of Python: 
+
+- `python --version`
+- `conda install python>=3.9.6`
 
-### Usage
-To run SBOannotator use the `__main__.py` script and modify the parameters in the `readSBML` and `sbo_annotator` 
-functions as wished.
 
-Alternatively run `python __main__.py` in the command line within the project folder.
+### Desktop App
 
-If ERROR occurs, check the current version of Python: 
+1. macOS: Download the DMG  
+  https://drive.google.com/file/d/1Ltm2nsFXpuVh7wgmQAY9Jg67EPJEooHW/view?usp=drive_link
 
-- `python --version'`
-- `conda install python>=3.8.5`
+2. Open the `.dmg`, then drag the app into **Applications**.
+3. First launch: if Gatekeeper blocks the app, remove the quarantine attribute in Terminal:
+   ```bash
+   sudo xattr -r -d com.apple.quarantine "/Applications/main.app"
 
 ### Exemplary models and results
-The folder `models/BiGG_Models` contains all the tested models as they were downloaded from
+The folder `models/Customer_Models` contains a  models as it were downloaded from
 the BiGG database. 
-The annotated models after using the SBOannotator are listed in the folder named `models/Annotated_Models`.
+The annotated models after using the Rule based SBOannotator and target reaction filtered for llm are listed in the folder named `models/Enhanced_Annotated_Models`.
+The annotated models after using the LLM recommendationi are listed in the models/LLM_Annotated_Models