datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
eccdna_copy_number_level_prediction_threshold_4.5
eccDNA Copy-Number Level Prediction (Threshold = 4.5)
This dataset supports the eccDNA copy-number prediction task,where the model learns to infer relative copy-number levels directly from eccDNA sequences.
Each entry contains:
sequence: DNA string (A/T/C/G)
label: copy-number stratum (e.g., 0 = low, 1 = high)
📁 Folder Structure
eccdna_copy_number_prediction/
├── data/
│ └── eccdna_copy_number_prediction.csv
├── hf_data/
│ ├── dataset_info.json
│ └── data.arrow… See the full description on the dataset page: https://huggingface.co/datasets/eccDNAMamba/eccdna_copy_number_level_prediction_threshold_4.5.real_vs_pseudo_eccdna_gallus_gallus
Real vs. Pseudo-eccDNA Discrimination (Gallus gallus)
This dataset supports the Real vs. Pseudo-eccDNA Discrimination task for gallus gallus eccDNA.The goal is to train models that can distinguish true eccDNA sequences from pseudo-eccDNAsrandomly extracted from linear genomic regions with matched length distributions.
Each entry contains:
sequence: raw eccDNA sequence (A/T/C/G)
label:
1 → Real eccDNA
0 → Pseudo-eccDNA (negative control)
📁 Folder Structure… See the full description on the dataset page: https://huggingface.co/datasets/eccDNAMamba/real_vs_pseudo_eccdna_gallus_gallus.real_vs_pseudo_eccdna_homo_sapiens
Real vs. Pseudo-eccDNA Discrimination (Homo sapiens)
This dataset supports the Real vs. Pseudo-eccDNA Discrimination task for human eccDNA.The goal is to train models that can distinguish true eccDNA sequences from pseudo-eccDNAsrandomly extracted from linear genomic regions with matched length distributions.
Each entry contains:
sequence: raw eccDNA sequence (A/T/C/G)
label:
1 → Real eccDNA
0 → Pseudo-eccDNA (negative control)
📁 Folder Structure… See the full description on the dataset page: https://huggingface.co/datasets/eccDNAMamba/real_vs_pseudo_eccdna_homo_sapiens.real_vs_pseudo_eccdna_arabidopsis_thaliana
Real vs. Pseudo-eccDNA Discrimination (Arabidopsis thaliana)
This dataset supports the Real vs. Pseudo-eccDNA Discrimination task for arabidopsis thaliana eccDNA.The goal is to train models that can distinguish true eccDNA sequences from pseudo-eccDNAsrandomly extracted from linear genomic regions with matched length distributions.
Each entry contains:
sequence: raw eccDNA sequence (A/T/C/G)
label:
1 → Real eccDNA
0 → Pseudo-eccDNA (negative control)
📁 Folder… See the full description on the dataset page: https://huggingface.co/datasets/eccDNAMamba/real_vs_pseudo_eccdna_arabidopsis_thaliana.cancer_eccdna_prediction_short
Cancer-associated eccDNA prediction (short)
This dataset contains Cancer-associated eccDNA prediction (short) sampled from healthy and cancer tissues.
Each entry contains:
sequence: DNA string (A/T/C/G)
label: 0 = healthy, 1 = cancer
📁 Folder Structure
cancer_eccdna_prediction_short/
├── data/
│ └── cancer_eccdna_prediction_short.csv
├── hf_data/
│ ├── dataset_info.json
│ └── data.arrow
└── README.md
🚀 Quick Usage
from datasets import… See the full description on the dataset page: https://huggingface.co/datasets/eccDNAMamba/cancer_eccdna_prediction_short.eccdna_copy_number_level_prediction_threshold_10
eccDNA Copy-Number Level Prediction (Threshold = 10)
This dataset supports the eccDNA copy-number prediction task,where the model learns to infer relative copy-number levels directly from eccDNA sequences.
Each entry contains:
sequence: DNA string (A/T/C/G)
label: copy-number stratum (e.g., 0 = low, 1 = high)
📁 Folder Structure
eccdna_copy_number_prediction/
├── data/
│ └── eccdna_copy_number_prediction.csv
├── hf_data/
│ ├── dataset_info.json
│ └── data.arrow… See the full description on the dataset page: https://huggingface.co/datasets/eccDNAMamba/eccdna_copy_number_level_prediction_threshold_10.