-
Notifications
You must be signed in to change notification settings - Fork 0
Troubleshooting
Carolina Monzó edited this page Dec 15, 2025
·
1 revision
Error: gtfToGenePred: command not found
Solution:
- Run the installation script:
bash install_ucsc_tools.sh
- Or add the tools to your PATH:
export PATH="/path/to/tools:$PATH"
Problem: Hub URL doesn't work in UCSC Genome Browser
Solutions:
- Ensure files are publicly accessible (no authentication required)
- Check that
hub.txtURL is correct - Verify all referenced files exist at the specified URLs
- Use raw URLs for GitHub (
https://raw.githubusercontent.com/...)
Problem: Hub not updating after requesting changes
Solutions:
- Regenerate the hub (the issue was fixed in recent versions)
- Clear your UCSC browser cache
- Ensure track is in "full" or "pack" mode (not "dense")
- Insert 'udcTimeout=5&' just after the "hgTracks" portion of the URL so that it reads http://genome.ucsc.edu/cgi-bin/hgTracks?udcTimeout=5& (with the remainder of the URL following the ampersand)
Problem: Dropdown filters or range sliders not appearing
Solutions:
- Right-click on the track → Configure (not just the main filter page)
- Regenerate the hub with the latest version
- Check that the bigBed file was created successfully
- Insert 'udcTimeout=5&' just after the "hgTracks" portion of the URL so that it reads http://genome.ucsc.edu/cgi-bin/hgTracks?udcTimeout=5& (with the remainder of the URL following the ampersand)
Problem: Searching in UCSC returns "no results"
Solutions:
- Ensure
trix.ixandtrix.ixxfiles were generated - Check that
trackDb.txtincludes thesearchTrixdirective - Use the correct search syntax (e.g.,
structural_category_novel_in_catalog) - Refresh the hub in UCSC (remove and re-add)
- Insert 'udcTimeout=5&' just after the "hgTracks" portion of the URL so that it reads http://genome.ucsc.edu/cgi-bin/hgTracks?udcTimeout=5& (with the remainder of the URL following the ampersand)
Error: bedToBigBed failed or file not sorted
Solutions:
- Ensure your GTF file has valid chromosome names matching the genome
- Check for special characters in transcript IDs
- Use
--keep-tempto inspect intermediate files
Problem: Script runs out of memory with large files
Solutions:
- Ensure you have sufficient RAM (4GB+ recommended)
- Close other applications
- Try processing on a machine with more memory
Edit the script to change logging level:
logging.basicConfig(level=logging.DEBUG, ...)Use --keep-temp to preserve intermediate files:
python sqanti3_to_UCSC.py ... --keep-tempThis helps identify where the process is failing.
Use --validate-only to check inputs without processing:
python sqanti3_to_UCSC.py ... --validate-onlyUse --dry-run to test up to BED generation:
python sqanti3_to_UCSC.py ... --dry-run- Check the GitHub Issues for similar problems
- Run
python test_installation.pyto verify your setup - Open a new issue with:
- Error message
- Command used
- Input file sizes
- Python and tool versions
- Python: 3.7+
- pandas: 1.0+
- UCSC Tools: Latest versions recommended
Check your versions:
python --version
python -c "import pandas; print(pandas.__version__)"
gtfToGenePred 2>&1 | head -1