tpbla-thermal 0.0.4

TPBLA_ThermAL: ML for amyloid thermodynamic landscape prediction.

ThermAL (Thermodynamics of Amyloid Landscapes)

---

🔧 Installation

Install from PyPI:

pip install tpbla-thermal

Then run:

tpbla-thermal

---

📦 Requirements

• Python 3.8+

Dependencies (exact pins where required):

pip install pandas numpy scipy==1.9.3 scikit-learn==1.2.2 seaborn matplotlib pillow joblib openpyxl

---

🧪 What is ThermAL?

ThermAL is a tool for predicting regions that stabilise amyloid fibrils.

Note: The feature extraction step can be time-limiting for larger sequences. These features are reusable for other ML tasks.

ThermAL takes one or more FASTA sequences as input, generates all single–residue variants, computes physicochemical features (AAC, DPC, sliding-window AUC), feeds them into a pre-trained Random Forest model, and produces the following key outputs:

• Predicted_fitness_with_1_letter_mutations.xlsx
• heatmap_simple.xlsx
• heatmap.png
• sliding_window.xlsx
• sliding_window_with_foldx.png

All outputs are written into per-job directories named after each input sequence.

---

📁 Project Structure

/ThermAL
│
├── required_docs/                  ← precomputed resources
│   ├── 3_B_Atlas.xlsx
│   ├── 3_BT_Atlas.xlsx
│   ├── 3_cDR_Atlas.xlsx
│   ├── 3_CF_Atlas.xlsx
│   ├── 3_Kd_Atlas.xlsx
│   ├── 3_P_Atlas.xlsx
│   ├── 3_DR_Atlas.xlsx
│   └── ThermAL.joblib
│
├── ThermAL.png                     ← logo displayed in GUI
├── src/tpbla_thermal/              ← package source
│   ├── __init__.py
│   ├── cli.py
│   └── gui.py
└── README.md

---

🚀 Usage (GUI)

1. Launch the GUI:

tpbla-thermal-gui

2. In the GUI:
   • Click Select FASTA File and choose your .fasta or .fa file.
   • Click Run Analysis.
   • Progress bars will update during AAC/DPC and feature processing.

When complete, you’ll find a subfolder per sequence in the working directory containing the outputs above.

---

📬 Contact

Any problems, feel free to reach out:
📧 [e-mail]
🔗 LinkedIn