Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2025
|
| Online Access: | https://doi.org/10.5281/zenodo.17538514 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866901153338359808 |
|---|---|
| author | BingShiun |
| author_facet | BingShiun |
| contents | <p>DNA-DETR, an adaptation of the DETR architecture for object detection in genomic sequences, using Non-B DNA elements as benchmarks. Our study reveals that conventional one-hot encoding inadequately captures complex sequence features. By comparing one-hot, dot matrix, and both representations, we show that representation choice significantly impacts detection accuracy and generalization. Combined representations consistently improve both localization and classification of functional genomic elements, highlighting the importance of representation-aware modeling in sequence analysis.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_17538514 |
| institution | Zenodo |
| language | |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | bio-it-station/DNA-DETR BingShiun <p>DNA-DETR, an adaptation of the DETR architecture for object detection in genomic sequences, using Non-B DNA elements as benchmarks. Our study reveals that conventional one-hot encoding inadequately captures complex sequence features. By comparing one-hot, dot matrix, and both representations, we show that representation choice significantly impacts detection accuracy and generalization. Combined representations consistently improve both localization and classification of functional genomic elements, highlighting the importance of representation-aware modeling in sequence analysis.</p> |
| title | bio-it-station/DNA-DETR |
| url | https://doi.org/10.5281/zenodo.17538514 |