Research map / Architectures and efficiency / RNNs and normalization
Deep Recurrent Neural Network-Based Identification of Precursor microRNAs
Seunghyun Park, Seonwoo Min, Hyun-Soo Choi, Sungroh Yoon
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use both sequence and secondary structure instead of sequence only (ablation) | LSTM+ATT pre-miRNA classifier · Human CV | F-score: 0.649 -> 0.839 | Improved |
| use both sequence and structure instead of structure only (ablation) | LSTM+ATT pre-miRNA classifier · Human CV | F-score: 0.795 -> 0.839 | Improved |
| use both modalities instead of sequence only (ablation) | LSTM+ATT pre-miRNA classifier · Human CV | sensitivity: 48% point increase | Improved |
| use both modalities instead of sequence only (ablation) | LSTM+ATT pre-miRNA classifier · Human CV | specificity: 0.987 -> 0.988 | No clear effect |
| add attention to 2-layer LSTM (ablation) | LSTM(2) · Human CV | F-score: 0.751 -> 0.839 | Improved |
| replace CNN with LSTM(2)+ATT (ablation) | 1D-CNN(2) · Human CV | F-score: 0.771 -> 0.839 | Improved |
| add LSTM layers on top of 1D-CNN (ablation) | 1D-CNN(2) · Human CV | F-score: 0.771 -> 0.738 | Hurt |
| add attention to CNN+LSTM (ablation) | 1D-CNN(2)+LSTM(2) · Human CV | F-score: 0.738 -> 0.739 | No clear effect |
| use two LSTM layers instead of one (ablation) | LSTM+ATT · Human CV | F-score: 0.824 -> 0.839 | Improved |
| make the LSTM bidirectional (ablation) | LSTM(2)+ATT · Human CV | F-score: 0.839 -> 0.834 | Hurt |
| use proposed method instead of microPred | pre-miRNA identification · Human test | F-score: 0.827 -> 0.868 | Improved |
| use proposed method instead of microPred | pre-miRNA identification · Cross-species test | F-score: 0.863 -> 0.906 | Improved |
Closest papers by meaning
- Nüwa.RNA: An RNA Foundation Model for Unified Representation with Deep Structure Infusion (NeurIPS 2026 · similarity 0.83)
- Understanding Transfer Learning of RNA Foundation Models on Downstream Tasks (ICML 2026 · similarity 0.83)
- PriFold: Biological Priors Improve RNA Secondary Structure Predictions (AAAI 2025 · similarity 0.82)
- BEACON: Benchmark for Comprehensive RNA Tasks and Language Models (NeurIPS 2024 · similarity 0.81)
- Deciphering RNA Secondary Structure Prediction: A Probabilistic K-Rook Matching Perspective (ICML 2024 · similarity 0.80)
- KINDLE: Knowledge-Guided Distillation for Prior-Free Gene Regulatory Network Inference (NeurIPS 2025 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
