SNNRice6mA: A Deep Learning Method for Predicting DNA N6-Methyladenine Sites in Rice Genome
Problem
Identifying DNA N6-methyladenine (6mA) sites in the rice genome experimentally is time-consuming and laborious, and existing machine learning predictors had room for improvement.
Why it matters
6mA is an important epigenetic modification involved in many biological regulatory processes, and accurate identification aids understanding of gene regulation in a staple crop.
What the AI did
A simple, lightweight deep learning model (SNNRice6mA) predicts 6mA sites from rice genome sequence data without requiring prior 6mA knowledge or manually crafted sequence features.
Results
Average prediction accuracy of ~93% and ~92% on two rice 6mA benchmark datasets
Impact
The model achieved high benchmark accuracy and was compared against existing 6mA predictors (comparison truncated in abstract).
Record
| Venue | Frontiers in Genetics |
|---|---|
| Published | 2019 |
| Work type | article |
| Authors | Haitao Yu, Zhiming Dai |
| Citations | 94 (OpenAlex, at import time) |
| References | 24 |
| Model or method | SNNRice6mA (deep learning model) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3389/fgene.2019.01071 |
| Open access | gold |
| Keywords | 6mA, epigenetics, deep learning, rice genome, methylation, prediction |
Importance rating
incremental / niche
Niche advance: faster computational identification of 6mA sites in rice genome, aiding epigenetic regulatory studies in one staple crop; no broader scale claimed.