🧬 Gene Annotation Integration

This project was developed for the Biological Databases course (MSc Bioinformatics, University of Edinburgh) in 2024. It integrates gene annotation data from multiple biological databases into a local MySQL database using an R script.

Description

Given a list of 10 mouse genes (Ensembl IDs), this script:

• Retrieves annotation data from Ensembl, NCBI Entrez, UniProt, and STRING-db
• Integrates the results into a local MySQL database
• Produces a final SQL summary table combining all annotations

Requirements

• R 4.2+
• MySQL server (on localhost)
• R packages:
  biomaRt, httr, jsonlite, queryup, RMySQL

You can install all required R packages with:

install.packages(c("biomaRt", "httr", "jsonlite", "queryup", "RMySQL"))

Usage

Inputs

• The list of 10 Ensembl gene IDs is hardcoded in the script.
• All database queries are performed programmatically via API or library.

How to Run

1. Make sure your MySQL server is running.

2. Supply your own database and API credentials by creating a .Renviron file in the project root with the following variables:

   DB_USER=your_db_username
   DB_PASSWORD=your_db_password
   DB_NAME=your_db_name
   DB_HOST=localhost
   DB_PORT=3306
   NCBI_API_KEY=your_ncbi_api_key

   You can also start by copying Renviron.template:

   cp .Renviron.template .Renviron

3. Open the R script (create_integrated_db.R).

4. Run the script in full (recommended inside an RStudio session or terminal).

5. The script will:

   • Create 5 MySQL tables (Ensembl_table, NCBI_table, Uniprot_table, String_table, Summary_table)

   • Write the merged gene annotations into Summary_table.

Output

The final Summary_table contains an integrated view across all data sources. Each row corresponds to a gene, with fields from:

• Ensembl: gene location, gene type
• NCBI: gene summary, genetic source
• UniProt: protein functions, GO terms
• STRING: top 3 protein interaction partners

Project Structure

    .
    ├── create_integrated_db.R      # Main integration script
    ├── schema.sql                  # MySQL table schema
    ├── data.sql                    # Sample integrated data
    ├──.Renviron.template           # Template for user credentials
    ├── README.md                   # Project overview and instructions
    └──.gitignore                   # Prevents committing sensitive files

Notes

• This project demonstrates practical biological data integration using APIs and SQL.

• Designed as a fully automated script with error handling and compatibility for re-running.

• Code is structured to allow generalisation for other gene sets.

License

This project is licensed under the MIT License.