🧬 TriRNASP

TriRNASP — by Tan-group, Wuhan University
A knowledge-based three-body statistical potential for accurate RNA 3D structure evaluation.

TriRNASP reads precomputed energy tables from the Energy/ folder, processes .pdb structures, and outputs the lowest-energy candidates.
The lower the energy (in kBT), the closer the structure is to the native state.

---

🧩 1. Quick Start

📥 Clone the Repository

git clone https://github.com/Tan-group/TriRNASP
cd TriRNASP

---

🔧 2. Build

✅ Recommended: Using Make

make

This will:

• Compile the executable TriRNASP with full OpenMP optimization.
• Automatically extract bundled datasets (Energy.zip, Training_set.zip, exmaple.zip).

After building, you should see the following structure:

TriRNASP/
 ├── TriRNASP           ← Executable file
 ├── Energy/            ← Energy tables
 ├── Training_set/      ← Training data
 └── exmaple/           ← Example dataset

🛠️ Manual Build (Alternative)

gcc -O3 -march=native -ffast-math -fno-math-errno -fopenmp \
    -Wall -Wextra -Wa,--noexecstack -Wl,-z,noexecstack \
    TriRNASP.c -lm -o TriRNASP

Requirements:

• GCC with OpenMP support
• Linux/Unix system

---

🚀 3. Run Example

Run TriRNASP on the provided example dataset:

./TriRNASP ./exmaple

Example Output:

Scanning  directory: ./exmaple/
Found PDBs: 182
R1205.pdb   -447.119919744277
R1205TS481_2.pdb   -391.548708213982
R1205TS481_1.pdb   -425.672444518381
R1205TS481_3.pdb   -292.657212955055
R1205TS481_4.pdb   -364.965577550650
Wall-clock time: 0.804421 seconds

✅ Notes:

• The output lists the Top 5 lowest-energy structures (in kBT).
• Lower energy indicates a structure closer to the native state.

---

⚙️ 4. Configuration

In TriRNASP.c:

#define num    1000  // Max PDBs per directory
#define path_l 300   // Max path length

If a stack-related core crash occurs, adjust num to fit the dataset size.

---

🧼 5. Clean Build Files

make clean

---

🧠 Tips

• If you see a permission error:

  chmod +x TriRNASP

• If you get command not found, make sure you are in the TriRNASP directory.

---

📧 Contact

Prof. Zhi-jie Tan
Wuhan University
✉️ zjtan@whu.edu.cn

---

🔖 Citation

Tongwei Yuan, En Lou, Zouchenyu Zhou, Ya-Lan Tan, Zhi-jie Tan.
TriRNASP: An efficient knowledge-based potential with three-body effects
for accurate RNA 3D structure evaluation. (2025)