ntv3
Datasets
All datasets matching “ntv3”NTv3_benchmark_dataset
NTv3 Benchmark Dataset
This dataset contains the functional tracks and genome annotations used in the NTv3 benchmark evaluation.
Dataset Organization
The dataset is organized as follows:
InstaDeepAI/NTv3_benchmark_dataset/
├── <species>/
│ ├── functional_tracks/
│ │ ├── <track_name>.bigwig
│ │ └── ...
│ └── genome_annotation/
│ ├── <track_name>.bed
│ └── ...
Available Species
arabidopsis
chicken
cattle
human
maize
rice… See the full description on the dataset page: https://huggingface.co/datasets/InstaDeepAI/NTv3_benchmark_dataset.omegagenome-ntv3-teacher-cacheNTv3_enhancer_generation
Enhancer generation dataset for NTv3-generative model
This dataset contains the processed STARR-seq data from the DeepSTARR study. Here we processed it with promoter context for conditional sequence generation training for NTv3-generative. Each enhancer is paired and inserted into two promoter contexts (RpS12 and DSCP), allowing the study of promoter-specific enhancer activity.
Source Data
Original Study: de Almeida et al., "DeepSTARR predicts enhancer activity… See the full description on the dataset page: https://huggingface.co/datasets/InstaDeepAI/NTv3_enhancer_generation.153-angiosperms-meaned-activations-NTV3-LargeNV-JEPA-DNA-NTv3NTV3-alllayer-152-genomes-32k-activations