Missing full feature CSVs / trained models for TLPath inference and reproduction

[mizukihonda6628]

Dear TLPath authors,

Thank you very much for making TLPath publicly available.

My name is Mizuki Honda, MD, from Kyoto Prefectural University of Medicine. I am trying to reproduce the TLPath pipeline using the public GitHub repository.

I was able to set up the environment successfully. However, I could not find the full tissue-level feature CSV files or the trained tissue-specific model files required for inference.

At the moment, I could only find the three example CSV files:

• Lung.csv
• Pancreas.csv
• Prostate.csv

In the README, the section on reproducibility mentions that the whole-slide level features are provided, but the Zenodo link appears as a placeholder. Because of this, I am not sure where to obtain:

1. the full feature CSV directory used for training/reproduction
2. the trained tissue-specific model files (for example, .joblib files for inference)

Could you kindly let me know whether these files are publicly available, and if so, where they can be accessed?

I would greatly appreciate your guidance.

Sincerely,
Mizuki Honda, MD
Kyoto Prefectural University of Medicine
mhonda@koto.kpu-m.ac.jp