A web app that can classify how likely specific mutations in DNA are to cause diseases (variant effect prediction). Using the state-of-the-art Evo2 large language model for prediction the pathogenicity of single nucleotide variants (SNVs). Deployed on FastAPI Python backend on an H100 serverless GPU with Modal for analysis. A web app where users can select a genome assembly, browse its chromosomes or search for specific genes like BRCA1, and view the gene's reference genome sequence. The user can input a mutation in the gene and predict its pathogenicity with AI, but the user can also pick from a list of existing known variations, and compare the Evo2 prediction (pathogenic/benign) against existing ClinVar classifications.
🧬 Evo2 Genome modeling and design across all domains of life
🔬 Explore gene and variants data (NCBI ClinVar/E-utilities)
🟢 Python backend deployed with Modal
👋 FastAPI endpoint
⚡ GPU-accelerated (H100) variant scoring via Modal
📱 Responsive Next.js 15, React 19 and is based off of the T3 Stack.
🎨 Modern UI with Tailwind CSS, and Shadcn UI
[🚧] This is on-going project to add for internal use for Anamnesai.com a Medical data analysis app that I'm currently building...
[•] Gigabite size CSV file analysis with context window more than 1M+ rows and 20+ columns EEG analysis for K. Anna from Epilepsy Reasearch Clinic Almaty.(Waiting for proper open source model with 10M+ context window to test 880MB cvs with EEG raw data)
[•] Collabrate with Atyrau biodoc ...
Check out the paper behind the model.
Follow these steps to install and set up the project.
Navigate to backend folder:
cd evo2-backendInstall dependencies:
pip install -r requirements.txtModal setup:
modal setupRun on Modal:
modal run main.pyDeploy backend:
modal deploy main.pyInstall dependencies:
cd evo2-frontend
npm installRun:
npm run dev