Skip to content
Other

ec-esm

enzyme classification from protein sequences

2 min read

This model predicts four-level Enzyme Commission numbers (EC — the international enzyme classification system, such as 1.1.1.131) directly from amino acid sequences. It builds on Meta's compact ESM-2 35M protein language model (facebook/esm2_t12_35M_UR50D) with a fine-tuned classifier covering 6,939 valid EC numbers. You feed it a Parquet table of sequences, and it outputs a JSONL file listing the predicted codes for each protein. It is built for bioinformaticians and researchers looking to automate functional enzyme annotation.

Key features

  • Built on the lightweight ESM-2 35M protein architecture with an added classification head (Linear + GELU + Dropout).
  • Covers a vocabulary of 6,939 valid four-level EC numbers, evaluated primarily on fourth-level accuracy (L4 accuracy).
  • Supports a configurable probability threshold via EC_THRESH (0.5 by default) to return either multiple candidates or only the top-scoring class.
  • Ships with a ready-to-use Dockerfile, inference script, and input validation.

What you can do

  • Drop the Docker container into a bioinformatics pipeline to batch-annotate protein sequences from Parquet files.
  • Set EC_THRESH=1.1 to strictly output only the single most probable EC number during initial protein triage.

Example

python /app/predict.py --input val_inputs.parquet --output predictions.jsonl
Read moreDetails
Architecture: ESM-2 35M (hidden_size 480) + two-layer classifier
Vocabulary: 6,939 EC classes (levels L1–L4)
Default runtime settings: batch size EC_BATCH=32, time limit EC_TIME_BUDGET=660 sec
Specific benchmark numbers are not reported in the model card
Discuss Telegram 0

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # sahithimacharapu2/ec-esm
    # ===== Tokens: paste yours between the quotes =====
    export HF_TOKEN=""        # huggingface.co/settings/tokens, for gated models
    # ==========================================
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download sahithimacharapu2/ec-esm --local-dir ~/models/ec-esm
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to ec-esm
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of ec-esm (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

sahithimacharapu2/ec-esm
Author
sahithimacharapu2
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments

Other

DAVID-Net

DAVID-Net