Overview

Researchers spend countless hours manually searching papers and still miss valuable prior evidence buried in figures, supplementary materials, and experiments outside the paper's primary focus.

HiveBlot indexes Western Blots making previously hidden findings searchable. Our platform enables researchers to discover relevant experimental results across the literature in seconds.

The Problem

Western blot evidence is often buried in figures, captions, methods, and supplementary materials rather than the main text. Researchers must manually examine these sections across hundreds of papers to determine whether a protein has already been studied, often missing relevant findings and repeating experiments.

The Solution

HiveBlot automatically extracts and indexes experimental evidence from scientific papers, allowing researchers to search across:

1. What protein was measured

2. What experiment was performed

3. What biological result was observed

4. Under what conditions

5. With direct links to the original figure

Features

1. AI-powered scientific paper parsing
2. Evidence-level search
3. Experimental result indexing
4. Natural language search interface
5. Fast retrieval of relevant supporting evidence

Tech Stack

Frontend: Next.js, React Backend: Python, FastAPI AI: Qwen 3-VL (hosted on RunPod), GPT 4.1

Run at: https://qbi-prod.vercel.app