BioNeMo

A transformers-based tokenizer for DNA sequences

Home
BioNeMo Recipes
BioNeMo Recipes
- About
  About
  - Overview
- Get Started
  Get Started
- Recipes
  Recipes
  - Models
    
    Models
    
    Models Directory
    
    Amplify
    
    Amplify
    
    Codonfm
    
    Codonfm
    
    Tests
    Tests
    
    Common
    
    Common
    
    Esm2
    
    Esm2
    
    Tests
    Tests
    
    Common
    
    Common
    
    Geneformer
    
    Geneformer
    
    Model readme
    
    Llama3
    
    Llama3
    
    Tests
    Tests
    
    Common
    
    Common
    
    Mixtral
    
    Mixtral
    
    Tests
    Tests
    
    Common
    
    Common
    
    Qwen
    
    Qwen
    
    Tests
    Tests
    
    Common
    
    Common
  - Recipes
    
    Recipes
    
    Recipes Directory
    
    Context Parallelism
    
    Codonfm native te
    
    Codonfm native te
    
    Hydra config
    
    Hydra config
    
    Codonfm ptl te
    
    Codonfm ptl te
    
    Data scripts
    Data scripts
    
    Data curation
    
    Data curation
    
    Notebooks
    
    Notebooks
    
    0-Zero-Shot-Mutation-Cancer-Hotspot
    
    Mutation Datasets Preprocessing
    
    000 Annotation File Processing
    
    1-Zero-Shot-Mutation-Variant-DDD-ASD
    
    2-Zero-Shot-Mutation-Variant-Clinvar-Alphamissense
    
    3-Zero-Shot-Mutation-Variant-Clinvar-Synonymous
    
    Tutorial 4: RiboNN Translation Efficiency Prediction with Encodon
    
    Tutorial 5: mRFP Expression Prediction with Encodon
    
    Tutorial 6: mRNA Stability Prediction with Encodon
    
    Eden megatron
    
    Eden megatron
    
    Src
    Src
    
    Bionemo
    Bionemo
    
    Eden
    Eden
    
    Data
    
    Data
    
    Utils
    Utils
    
    Checkpoint
    
    Checkpoint
    
    Tokenizers
    Tokenizers
    
    Nucleotide fast tokenizer 256
    
    Nucleotide fast tokenizer 256
    
    Nucleotide fast tokenizer 512
    
    Nucleotide fast tokenizer 512
    
    Esm2 accelerate te
    
    Esm2 accelerate te
    
    Example 8m checkpoint
    
    Example 8m checkpoint
    
    Hydra config
    
    Hydra config
    
    Esm2 native te
    
    Esm2 native te
    
    Hydra config
    
    Hydra config
    
    Esm2 peft te
    
    Esm2 peft te
    
    Example 8m checkpoint
    
    Example 8m checkpoint
    
    Hydra config
    
    Hydra config
    
    Evo2 megatron
    
    Evo2 megatron
    
    Examples
    
    Examples
    
    Evaluating Evo2 with GFMBench-API (BioNeMo Recipe)
    
    Fine tuning tutorial
    
    Lora fine tuning tutorial
    
    Zero-shot prediction of BRCA1 variant effects with Evo 2
    
    Configs
    
    Configs
    
    Src
    Src
    
    Bionemo
    Bionemo
    
    Evo2
    Evo2
    
    Data
    
    Data
    
    Utils
    Utils
    
    Checkpoint
    
    Checkpoint
    
    Tokenizers
    Tokenizers
    
    Nucleotide fast tokenizer 256
    
    Nucleotide fast tokenizer 256
    
    Nucleotide fast tokenizer 512
    
    Nucleotide fast tokenizer 512
    
    Fp8 analysis
    
    Fp8 analysis
    
    Geneformer native te mfsdp fp8
    
    Geneformer native te mfsdp fp8
    
    Hydra config
    
    Hydra config
    
    Model
    
    Model
    
    Llama3 native te
    
    Llama3 native te
    
    Hydra config
    
    Hydra config
    
    Tokenizers
    Tokenizers
    
    Nucleotide fast tokenizer
    
    Nucleotide fast tokenizer
    
    Opengenome2 llama native te
    
    Opengenome2 llama native te
    
    OpenGenome2 Metagenome Dataset: Sharding, Shuffling, and Tokenization in the LlaMA 3 Recipe
    
    Hydra config
    
    Hydra config
    
    Tokenizers
    Tokenizers
    
    Nucleotide fast tokenizer
    
    Nucleotide fast tokenizer
    
    Vit
    
    Vit
    
    Config
    
    Config
    
    Vllm inference
    Vllm inference
    
    Esm2
    
    Esm2
- Tutorials
  Tutorials
- Data Sets
  Data Sets
  - CELLxGENE
  - UniProt Dataset
- Contributing
  Contributing
  - Code Review
  - Contributing
  - Writing Documentation
    
    Writing Documentation
    
    Jupyter Notebook Support
    
    MkDocs
- Frequently Asked Questions
Models
Models
- Amplify
- Evo2
- Geneformer
- ESM 2
  ESM 2
  - Pre-trained Checkpoints

A transformers-based tokenizer for DNA sequences

This tokenizer is similar to the one used in the Evo-2 DNA model, and tokenizes DNA sequences using the ord() function.