A Deep Learning Framework for Robust and Accurate Prediction of ncRNA-Protein Interactions Using Evolutionary Information
Problem
Identifying which proteins bind specific non-coding RNAs with wet-lab high-throughput techniques is expensive and time-consuming, so computational prediction is needed.
Why it matters
ncRNA-protein interactions underlie many biological processes, and faster prediction could accelerate biological research.
What the AI did
The RPI-SAN model uses a deep-learning stacked auto-encoder network to extract hidden high-level features from RNA and protein sequences, feeding them into a random forest classifier, with stacked assembling to improve accuracy.
Results
not stated in source
Impact
Provides a computational alternative to wet-lab identification, benchmarked against five established prediction tools on four datasets.
Record
| Venue | Molecular Therapy — Nucleic Acids |
|---|---|
| Published | 2018 |
| Work type | article |
| Authors | Hai-Cheng Yi, Zhu‐Hong You, De-Shuang Huang, Li Xiao, Tong-Hai Jiang, Liping Li |
| Citations | 134 (OpenAlex, at import time) |
| References | 81 |
| Model or method | RPI-SAN (stacked auto-encoder network + random forest) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1016/j.omtn.2018.03.001 |
| Open access | gold |
| Keywords | deep learning, ncRNA, protein interactions, stacked autoencoder, random forest, benchmark |
Importance rating
real but narrower advance
ncRNA-protein interactions matter in many biological processes and wet-lab identification is expensive and time-consuming, but the abstract frames this as a computational benchmark advance for biomedical research, not a quantified-impact problem.