File size: 1,724 Bytes
b42e583 b48bc49 4feb434 b48bc49 4feb434 b48bc49 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 |
---
title: Protein Similarity API
emoji: 🧬
colorFrom: blue
colorTo: indigo
sdk: docker
app_port: 7860
python_version: 3.10
short_description: FastAPI for comparing protein sequences.
tags:
- bioinformatics
- protein
- api
- fastapi
- transformer
- huggingface
pinned: true
---
# protein-similarity-api
A lightweight, production-ready FastAPI service for comparing protein sequences using state-of-the-art transformer embeddings (ESM-2).
It provides a simple REST API to measure semantic similarity between protein sequences, returning a similarity score based on cosine distance in embedding space.
---
## Why this matters?
Protein similarity is at the heart of functional annotation, homology detection, and structure prediction. This API lets anyone (researcher, student, dev) compare protein sequences using powerful models.
---
## Features
- Compare two protein sequences using [ESM-2](https://huggingface.co/facebook/esm2_t33_650M_UR50D)
- Returns cosine similarity + classification (homologous vs. non-homologous)
- Lightweight, deployable on free-tier platforms like Render or Railway
- Clean FastAPI structure, easy to extend (e.g. embeddings endpoint, caching)
---
## API Endpoints
### `POST /compare`
Compare two sequences and get similarity:
#### Request:
```json
{
"sequence_1": "MSSKVIFF...",
"sequence_2": "MTTRLIFF...",
"model": "esm_2_650m"
}
```
#### Response:
```json
{
"cosine_similarity": 0.57,
"classification": "moderate similarity (possible remote homolog)",
"model": "ESM-2 650M"
}
```
## How to run it?
```shell
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn main:app --reload
```
## How to deploy it?
```
```
|