AI-powered Biomedical Literature Research Assistant built using RAG (Retrieval-Augmented Generation), FAISS, PubMed, and Google Gemini.
👉 Try the application here (https://bioresearch-ai.streamlit.app/)
BioResearch AI is an AI-powered biomedical literature assistant that automatically retrieves scientific publications from PubMed, performs semantic search using FAISS embeddings, and generates structured literature reviews using Google's Gemini model.
Instead of manually reading dozens of research papers, researchers can ask a biological question and receive an evidence-based literature review in seconds.
✅ PubMed Literature Retrieval
✅ Semantic Search using Sentence Transformers
✅ FAISS Vector Database
✅ Retrieval-Augmented Generation (RAG)
✅ AI Literature Review Generation
✅ Biological Research Assistant
User Query
│
▼
PubMed Search
│
▼
Paper Collection
│
▼
Sentence Embeddings
│
▼
FAISS Vector Search
│
▼
Relevant Papers
│
▼
Gemini AI
│
▼
Structured Literature Review
- Python
- Streamlit
- PubMed API
- Sentence Transformers
- FAISS
- Google Gemini
- NumPy
BioResearch-AI
│
├── app
│ ├── agents
│ ├── services
│ ├── utils
│ ├── streamlit_app.py
│ └── main.py
│
├── data
│ ├── raw
│ └── processed
│
├── README.md
├── requirements.txt
└── LICENSE
git clone https://github.com/sarojjriyaaa/BioResearch-AI.git
cd BioResearch-AI
pip install -r requirements.txt
streamlit run app/streamlit_app.pyDNMT1 inhibitors in colorectal cancer
HIV-1 protease drug discovery
The application generates:
- Overview
- Major Findings
- Biological Mechanisms
- Important Genes
- Important Drugs
- Contradictory Evidence
- Research Gaps
- Future Directions
- References (PMIDs)
- Multi-Agent Research System
- PDF Literature Review Export
- Chat History
- ChromaDB Support
- Citation Network Visualization
- Paper Recommendation Engine
- Full-text Paper Retrieval (PMC)
Riya Saroj
B.Tech Biotechnology
Interested in Computational Biology • Bioinformatics • AI for Healthcare
MIT License