Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -269,7 +269,7 @@ dataset_info:
|
|
| 269 |
dataset_size: 334784354
|
| 270 |
---
|
| 271 |
|
| 272 |
-
#
|
| 273 |
|
| 274 |
A benchmark of **sequence-level perturbation tasks** for evaluating DNA foundation models.
|
| 275 |
Each task presents pairs of genomic sequences — one real (unperturbed) and one structurally
|
|
@@ -340,17 +340,17 @@ from datasets import load_dataset
|
|
| 340 |
|
| 341 |
# Synonymous codon substitution — human
|
| 342 |
syn_human = load_dataset(
|
| 343 |
-
"HuggingFaceBio/
|
| 344 |
)
|
| 345 |
|
| 346 |
# Synonymous codon substitution — mouse
|
| 347 |
syn_mouse = load_dataset(
|
| 348 |
-
"HuggingFaceBio/
|
| 349 |
)
|
| 350 |
|
| 351 |
# CAG motif insertion
|
| 352 |
motif = load_dataset(
|
| 353 |
-
"HuggingFaceBio/
|
| 354 |
)
|
| 355 |
```
|
| 356 |
|
|
|
|
| 269 |
dataset_size: 334784354
|
| 270 |
---
|
| 271 |
|
| 272 |
+
# Perturbation Bench
|
| 273 |
|
| 274 |
A benchmark of **sequence-level perturbation tasks** for evaluating DNA foundation models.
|
| 275 |
Each task presents pairs of genomic sequences — one real (unperturbed) and one structurally
|
|
|
|
| 340 |
|
| 341 |
# Synonymous codon substitution — human
|
| 342 |
syn_human = load_dataset(
|
| 343 |
+
"HuggingFaceBio/perturbation-bench", "syn_human", split="test"
|
| 344 |
)
|
| 345 |
|
| 346 |
# Synonymous codon substitution — mouse
|
| 347 |
syn_mouse = load_dataset(
|
| 348 |
+
"HuggingFaceBio/perturbation-bench", "syn_mouse", split="test"
|
| 349 |
)
|
| 350 |
|
| 351 |
# CAG motif insertion
|
| 352 |
motif = load_dataset(
|
| 353 |
+
"HuggingFaceBio/perturbation-bench", "motif_human", split="test"
|
| 354 |
)
|
| 355 |
```
|
| 356 |
|