Dear PIP-eco Developers,
First of all, thank you for developing this tool. It is clear that a lot of work has gone into creating a streamlined and practical pipeline for pathogenomic analysis. I especially appreciate the modularity and breadth of functionality.
However, while attempting to install the tool, I encountered several challenges that I believe may hinder other users as well. Below are some issues and observations I hope you can consider:
- Environment & Compatibility Issues
Python version: The current codebase appears to require Python ≥3.9 (e.g., for dict[str, float] type annotations) but the documentation or code suggests Python 3.7 compatibility.
Dependency mismatch: Some modules/libraries expect newer versions (e.g., ipywidgets, IProgress) not automatically installed in the environment.
- Software Dependency
The usearch binary is expected at include/software/usearch, but this folder is empty by default after configuring conda env.
Since usearch is not free for all use cases, I tried substituting it with vsearch, but this required editing the call from -ublast to --usearch_global and adjusting other flags accordingly.
Would you consider either bundling vsearch compatibility or recommending it as the default option in the documentation?
- Input Path Confusion
Clarifying expected folder structure and naming in the README would be very helpful, since there is no specification of where to put input genomes.
Again, thank you so much for this innovative tool. Any help would be much appreciated.
Best regards,
Medelin
Dear PIP-eco Developers,
First of all, thank you for developing this tool. It is clear that a lot of work has gone into creating a streamlined and practical pipeline for pathogenomic analysis. I especially appreciate the modularity and breadth of functionality.
However, while attempting to install the tool, I encountered several challenges that I believe may hinder other users as well. Below are some issues and observations I hope you can consider:
Python version: The current codebase appears to require Python ≥3.9 (e.g., for dict[str, float] type annotations) but the documentation or code suggests Python 3.7 compatibility.
Dependency mismatch: Some modules/libraries expect newer versions (e.g., ipywidgets, IProgress) not automatically installed in the environment.
The usearch binary is expected at include/software/usearch, but this folder is empty by default after configuring conda env.
Since usearch is not free for all use cases, I tried substituting it with vsearch, but this required editing the call from -ublast to --usearch_global and adjusting other flags accordingly.
Would you consider either bundling vsearch compatibility or recommending it as the default option in the documentation?
Clarifying expected folder structure and naming in the README would be very helpful, since there is no specification of where to put input genomes.
Again, thank you so much for this innovative tool. Any help would be much appreciated.
Best regards,
Medelin