evo2

A framework for state-of-the-art pre-trained bio foundation models on genomics and transcriptomics modalities.

Decentralized genomic variant analysis on Google's Evo 2 + BNB Chain: Genome NFTs, DAO governance, and serverless GPU inference. BNB Hack Kerala winner.

AI-powered variant pathogenicity prediction using the Evo2 genomic language model on GPU, cross-validated against ClinVar.

Fine-tuning pipeline for NVIDIA BioNeMo Evo2 1B on genomics sequences, including preprocessing, training, inference, embeddings, and evaluation.

Native Windows app that runs the Evo 2 genomic language model on a GPU in your own Google Cloud project and returns per-base Shannon entropy tracks for IGV, Geneious, SnapGene and Benchling.

Native Apple Silicon support for Arc Institute's Evo2. Run StripedHyena models locally on macOS via Metal Performance Shaders.

An Analysis application that allows single nucleotide variations in genes to help predict "pathogenic" or not using the Evo2 LLM Model

Exploratory Evo 2 promoter-grammar analysis with frozen aggregate statistics and figures for human protein-coding genes.

?? Ultron Workflow - Where Science Meets Sarcasm ??? | Visual workflows, protein & DNA labs, drug discovery, 3-D mech/circuit, all powered by Ultron - your brilliantly sarcastic superintelligence. He roasts, then results. ????

Prompt-conditioned generation and LoRA finetuning of Evo 2 on rbcL, with a reproducible evaluation harness

a web app that can classify how likely specific mutations in DNA are to cause diseases (variant effect prediction). We will deploy and use the state-of-the-art Evo2 large language model, and use it to predict the pathogenicity of single nucleotide variants (SNVs)

A deep learning model (EVO2-500M) for predicting host specificity of eukaryote-infecting viruses CDNA sequence

PlatyGeno is a Python package for identifying genomic landmarks directly from raw sequence data. By leveraging the Evo 2 foundation model, it performs annotation-independent identification of biologically significant structures—such as promoters, coding sequences, and precise motifs—relying purely on model confidence and latent feature.

This project critically analyzes the data pipeline of EVO 2, a cutting-edge AI model for genomic research developed by the Arc Institute and collaborators. The focus lies in evaluating its risks, ethical challenges, and governance structures in the context of data privacy, genome editing, and AI-driven biomedical innovation.

Aligning multi-hybrid foundation models for genome generation with domain adaptation to human genetics

Evo 2 の内部表現をスパースオートエンコーダで読むハンズオン

Genome analysis toolkit powered by Evo foundation models.

A Full Stack Variant Effect Predictor leveraging the Evo2 LLM for Single Nucleotide Variant(SNV) pathogenicity analysis, powered by a Modal‑deployed FastAPI GPU backend and a Next.js + Shadcn UI frontend

Unofficial PyTorch and Transformers port of Arc Institute Evo 2 for DNA modeling and generation. Loads official checkpoints with no Vortex or custom kernels needed.

Predicts pathogenic vs benign DNA variants using Evo2 (7B-param genomic foundation model)