Contig_embeddings / README.md
spacemorrissey's picture
Dataset card; drop internal fasta_path column from manifests
1a97f33 verified
|
Raw History Blame Contribute Delete
4.55 kB
---
license: cc-by-4.0
pretty_name: IMAGINE-AI contig-level Evo 2 embeddings
tags: [biology, genomics, microbiology, embeddings, dna-language-model, evo2]
size_categories: [10K<n<100K]
viewer: false
---
# IMAGINE-AI contig-level Evo 2 embeddings
Per-contig **Evo 2** (7B base) embeddings, all 32 layers, for **11,862 bacterial and archaeal genomes**
(public NCBI assemblies), computed on ALCF Polaris (A100) in October 2026. Built for genotype→phenotype
modelling in IMAGINE-AI (DOE DE-FOA-0003612, Topic 2-B; PI Ember Morrissey, West Virginia University).
The genome-level vectors (Evo 2 and MicroGlot, one small file per layer) are in the companion dataset
[`ImagineG2P/genome-embeddings`](https://huggingface.co/datasets/ImagineG2P/genome-embeddings). Use this
dataset when you need **one vector per contig**, e.g. to re-pool genomes by contig length, drop plasmids or
short contigs, or study contig-level signal.
## What is in it
| Path | Contents |
|---|---|
| `contigs/evo2/<ddd>/<accession>.npz` | one file per genome (11,862 files, ~465 GB in total; 1 MB to 2.6 GB, median 15 MB) |
| `manifests/genomes.csv` | one row per listed genome (11,868; 6 have `status = rejected` and no `.npz`) |
| `manifests/pilot_genomes.csv` | selection checks for the 12 pilot genomes |
| `manifests/labels.csv` | phenotype labels for the pilot genomes only (Madin, BacDive, Bergey's) |
`<ddd>` is digits 4–6 of the accession number: `GCF_000007365.1` → `contigs/evo2/007/`
(in Python, `acc[7:10]`). This keeps every folder well under 10,000 files.
### Each `.npz`
| Key | dtype, shape | Meaning |
|---|---|---|
| `embedding` | float32 [32, 4096] | genome vector, all layers (identical to the row in `genome-embeddings/evo2/`) |
| `contig_means` | float32 [n_contigs, 32, 4096] | one vector per contig, all layers |
| `contig_ids` | str [n_contigs] | FASTA record IDs, same order as `contig_means` |
| `meta` | JSON string | per-genome provenance: model revision, layers, pooling policy, `fasta_sha256`, `n_contigs`, `n_windows`, `total_bp`, IUPAC cleaning per contig, runtime and software environment |
Layer *i* is `blocks.i`: the residual stream after Evo 2 block *i*, before the final norm.
```python
from huggingface_hub import hf_hub_download
import numpy as np, json
acc = "GCF_000007365.1"
p = hf_hub_download("ImagineG2P/Contig_embeddings", f"contigs/evo2/{acc[7:10]}/{acc}.npz",
repo_type="dataset")
z = np.load(p)
contigs = z["contig_means"][:, 24] # [n_contigs, 4096], layer blocks.24
ids = z["contig_ids"]
meta = json.loads(str(z["meta"]))
```
To fetch a subset without downloading everything, use `snapshot_download(..., allow_patterns=[...])` or
`hf download ImagineG2P/Contig_embeddings --repo-type dataset --include "contigs/evo2/007/*"`.
## Model and pooling
| | |
|---|---|
| Model | [`arcinstitute/evo2_7b_base`](https://huggingface.co/arcinstitute/evo2_7b_base) @ `074097e9` |
| Tokens | one byte per base (CharLevelTokenizer) |
| Window | 8,192 bp, forward strand only |
| Precision | bf16 inference, float32 pooling |
**Pooling.** Mean over tokens within each 8,192-bp window, then an equal-weight mean over the windows of a
contig (the short last window counts as much as a full one) → `contig_means`. The genome vector
`embedding` is an **equal-weight mean over contigs**, so a 1-kb contig counts as much as a 5-Mb chromosome.
For a length-weighted genome vector, weight `contig_means` by contig length (from the assembly FASTA; the
total is `total_bp` in `meta`). Non-ACGT IUPAC codes were set to N before embedding (counts in `meta`).
## Genomes
All public NCBI assemblies (GCF_/GCA_), each with a `priority` from the IMAGINE-AI master list: 0 pilot
genomes (11), 1 reference genomes for the WVU isolates' species (80), 2 QC-passed genomes from the
growth-rate master table (283), 3 other genomes in that table (554), 4
[`ImagineG2P/fasta`](https://huggingface.co/datasets/ImagineG2P/fasta) (10,940). `fna_sha256` in
`manifests/genomes.csv` is the hash of the exact FASTA embedded.
## Notes and limits
- Frozen-model features, not fine-tuned. MicroGlot contig vectors are not included.
- Draft assemblies vary widely in contig count (`n_contigs` in `manifests/genomes.csv`); see the pooling note.
- Code: `Hackathon/pilot/` in the IMAGINE-AI repository (`embed_evo2.py`, `export_genome_embeddings.py`).
## License and citation
CC-BY-4.0. Please cite IMAGINE-AI (DOE DE-FOA-0003612), Evo 2 (Brixi et al., *Nature* 2026; weights
Apache-2.0, Arc Institute) and NCBI for the assemblies.