Update genome annotation file with multiple samples?

[jasonleongbio]

Hi, sorry for having posted so many issues recently.

I was simply wondering whether it's possible to update the old genome annotation file with BAM files from multiple samples. I have multiple samples from the same species (which is a non-model organism), so what I tried was to update the genome annotation file with the BAM file from one single-cell sample at a time. However, when I tried to update the already-updated gff file with the BAM file from a different single-cell sample, I got the following error message (i.e., original gtf/gff → updated version 1 → updated version 2; it failed at the conversion step from updated version 1 to updated version 2) :

INFO     Iterating over reads to determine SPAT pileups:   0%|           | [20:53<?]
2023-02-01 22:28:29,771 - INFO - Merging SPAT outputs.
2023-02-01 22:28:30,058 - INFO - Creating gff db.
2023-02-01 22:28:30,060 - INFO - Calling peaks for forward strand with MACS2.
2023-02-01 22:28:30,069 - INFO - Calling peaks for reverse strand with MACS2.
2023-02-01 22:28:30,069 - INFO - Populating features
2023-02-01 22:28:30,069 - INFO - Populating features
2023-02-01 22:28:30,227 - INFO - Clearing cache.ons: 2000 features
Traceback (most recent call last):
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/bin/peaks2utr", line 8, in <module>
    sys.exit(main())
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/peaks2utr/__init__.py", line 49, in main
    asyncio.run(_main())
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/asyncio/runners.py", line 44, in run
    return loop.run_until_complete(main)
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/asyncio/base_events.py", line 647, in run_until_complete
    return future.result()
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/peaks2utr/__init__.py", line 102, in _main
    db, _, _ = await asyncio.gather(
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/peaks2utr/preprocess.py", line 120, in create_db
    await sync_to_async(gffutils.create_db)(gff_in, gff_db, force=True, verbose=True, disable_infer_genes=True, disable_infer_transcripts=True)
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/asgiref/sync.py", line 448, in __call__
    ret = await asyncio.wait_for(future, timeout=None)
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/asyncio/tasks.py", line 442, in wait_for
    return await fut
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/concurrent/futures/thread.py", line 58, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/asgiref/sync.py", line 490, in thread_handler
    return func(*args, **kwargs)
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/gffutils/create.py", line 1292, in create_db
    c.create()
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/gffutils/create.py", line 507, in create
    self._populate_from_lines(self.iterator)
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/gffutils/create.py", line 589, in _populate_from_lines
    self._insert(f, c)
  File "/~~~home_dir~~~/tools/miniconda3/envs/peaks2utr/lib/python3.9/site-packages/gffutils/create.py", line 530, in _insert
    cursor.execute(constants._INSERT, feature.astuple())
sqlite3.InterfaceError: Error binding parameter 11 - probably unsupported type.

The code I ran was:

peaks2utr --extend-utr -p 25 -o <GENOME_ANNOTATION>.update2.gff3 /path/to/<GENOME_ANNOTATION>.update1.gff3 /path/to/the/second/replicate/sample/outs/possorted_genome_bam.bam

Thank you so much

[haessar]

Interesting, I've not encountered this before. Allowing multiple BAM files as input is on the feature wish-list so watch this space. I wonder, can you verify that /path/to/<GENOME_ANNOTATION>.update1.gff3 is indeed in the correct format (i.e. not an empty file)? This is one of the reasons we use genometools as a post-processing step (see issue #3 ) - it performs validation to ensure everything is correctly formatted. If you have access to genometools, could you try running "gt gff3 -sort -retainids -tidy /path/to/<GENOME_ANNOTATION>.update1.gff3" to ensure no error is thrown.