Use Evo 2 to score DNA sequences, turn genomic regions into numerical representations for classification, and generate sequences from a supplied prefix. Compare reference and variant sequences or assess coding and regulatory regions across species.
中文简介正在准备中。当前显示英语简介。
Score, embed, and generate DNA sequences with Evo 2, a long-context genomic foundation model. Use this skill when: (1) Computing per-nucleotide or per-sequence likelihoods for variant effect scoring, (2) Embedding genomic windows for downstream classification, (3) Generating DNA conditioned on a prefix, (4) Scoring regulatory or coding regions across species.
安装前可以查看交给 AI 代理的指令内容。
| Requirement | Minimum | Recommended |
|---|---|---|
| Python | 3.11 | 3.12 (<3.13) |
| CUDA | 12.1+ | 12.4+ |
| GPU VRAM | 24 GB (7B bf16) | 80 GB (40B) |
| RAM | 32 GB | 128 GB |
pip install evo2
# Weights pulled from Hugging Face on first model load.
from evo2 import Evo2
model = Evo2("evo2_7b") # or "evo2_40b" — see model table
seqs = ["ATCG" * 50, "GGGCTTAA" * 25]
ll = model.score_sequences(seqs) # → list[float], mean per-token log-likelihood
print(ll)
out = model.generate(
prompt_seqs=["ATGAAAGCT"],
n_tokens=256,
temperature=0.7,
)
print(out.sequences[0])
| Name | Params | Context | VRAM (bf16) | Notes |
|---|---|---|---|---|
evo2_7b | 7 B | 1 M nt | ~22 GB | Default; fits on a single 24 GB+ GPU |
evo2_40b | 40 B | 1 M nt | ~78 GB | H100 80 GB or multi-GPU |
evo2_1b_base | 1 B | 8 K nt | ~6 GB | FP8 path requires sm_89+ (H100) |
score_sequences returns a list[float] (or np.ndarray) of mean log-likelihoods,
one per input sequence. More negative ⇒ less likely under the model. For variant
effect, compute Δll = ll_alt - ll_ref over a fixed window.
generate returns a GenerationOutput with .sequences (list[str]), .logits
(list[Tensor]), and .logprobs_mean (list[float]) — always populated, no flag required.
Need a DNA model?
│
├─ Per-base/per-sequence likelihood, generation → Evo 2 ✓
├─ Predict experimental tracks (expression, accessibility) → borzoi
└─ Protein, not DNA → fair-esm2 / esmfold2
7B/40B inference is GPU-bound (≥24 GB / 80 GB VRAM). Read
compute_details({provider, mode:'read'}) for an environment with evo2 +
flash-attn and a pre-cached HF weight mount, then submit:
c = host.compute.create(provider)
job = c.submitJob(
intent="Evo2-7B score 200bp variant window — 1×GPU, ~2 min",
inputs=[{"src": "score_evo2.py", "dstFilename": "score_evo2.py"}],
command="python3 score_evo2.py", # env selection is host-specific — see compute_details for your provider
outputs=["scores.json"],
timeoutSeconds=1800,
)
print(job.job_id) # cell ends here — kernel never blocks on compute
Retain the exact returned job_id. Query that saved ID with the non-blocking
c.attachJob(job_id).status() or .result() when its state or result is relevant; do not scan Job
history. A final .result() read reports whether its follow-up was suppressed or had already been
committed; otherwise the app starts the later analysis turn for an unread final result. See the
remote-compute-ssh skill for details.
Inside score_evo2.py, point HF_HOME at the provider's weight-cache mount
(path is in compute_details) and set HF_HUB_OFFLINE=1 so the loader
doesn't try to write refs/ into a read-only mount. Weight footprint:
~15 GB (7B), ~80 GB (40B).
| Task | 7B on H100 | Notes |
|---|---|---|
| Model load (cached) | ~5-7 min | First call hydrates weights |
score_sequences, 200×200bp | ~10-20 s | After load |
generate, 1×512 nt | ~15 s |
| Symptom | Cause | Fix |
|---|---|---|
Transformer Engine not installed | No FP8 — falls back to bf16 | Informational only on non-H100; ignore |
| OOM on load | 40B on <80 GB GPU | Use evo2_7b or shard with device_map |
HF tries to write refs/main | HF_HOME points at RO mount | Set HF_HUB_OFFLINE=1 |
dtype mismatch in score_sequences | Passing tensors not strings | Pass list[str]; the API tokenises for you |
Next: pair with borzoi to predict track-level effects of the same
variants.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Predict structures for single proteins or protein complexes with AlphaFold2 through ColabFold. Use confidence scores and comparisons with a designed structure to assess whether a designed sequence folds as intended.
中文简介正在准备中。当前显示英语简介。
Use Boltz-2 to predict structures of complexes containing proteins, DNA, RNA, and small molecules. Obtain structure files and confidence scores to assess designed binders, with optional predictions of binding strength between a protein and a small molecule.
中文简介正在准备中。当前显示英语简介。
Predict patterns of gene expression and other genomic measurements from DNA sequences with Borzoi. Compare reference and altered sequences to estimate regulatory effects, generate predicted coverage tracks, and prioritize variants in regions that do not encode proteins.
中文简介正在准备中。当前显示英语简介。
Predict three-dimensional structures of complexes containing proteins, RNA, DNA, and small molecules with Chai-1. Compare candidate structures using confidence scores and clash indicators, or run repeated predictions from Python to evaluate designed binders.
中文简介正在准备中。当前显示英语简介。
Prepare reproducible setup and removal instructions for software environments on Open-Science SSH hosts, including Slurm clusters. Check available tools, define named activation settings, and validate the environment where jobs run after the user installs it.
中文简介正在准备中。当前显示英语简介。
Create and manage focused Specialist agents and Skills through conversation. Draft agent roles, review available Skills and Connectors, configure access, and manage changes, with Skill creation and publishing handled by the application's internal skill-creator.
中文简介正在准备中。当前显示英语简介。