replace talon.db about swan_vis HOT 10 CLOSED

from swan_vis.

Please look at this section of the documentation: https://freese.gitbook.io/swan/tutorials/getting_started#adding-transcript-models-from-a-gtf

from swan_vis.

Thank you for the tutorial. There is still an issue regarding the "merge on int64" as below

Traceback (most recent call last):
  File "/Swan/Swan.py", line 24, in
    sg.add_abundance(ab_file)
  File "/conda_envs/swan/lib/python3.9/site-packages/swan_vis/swangraph.py", line 276, in add_abundance
    df = df.merge(self.t_df['tid'].to_frame(), how='right', left_on='tid', right_index=True)
  File "/conda_envs/swan/lib/python3.9/site-packages/pandas/core/frame.py", line 9191, in merge
    return merge(
  File "/conda_envs/swan/lib/python3.9/site-packages/pandas/core/reshape/merge.py", line 105, in merge
    op = _MergeOperation(
  File "/conda_envs/swan/lib/python3.9/site-packages/pandas/core/reshape/merge.py", line 702, in init
    self._maybe_coerce_merge_keys()
  File "/conda_envs/swan/lib/python3.9/site-packages/pandas/core/reshape/merge.py", line 1255, in _maybe_coerce_merge_keys
    raise ValueError(msg)
ValueError: You are trying to merge on int64 and object columns. If you wish to proceed you should use pd.concat

from swan_vis.

Can you send me a screenshot of your abundance file please?

from swan_vis.

Yes. Thank you

from swan_vis.

Can you also please send me a couple of lines of your GTF? It seems like you might have mismatches between the transcript IDs reported in your GTF and reported in your abundance matrix.

from swan_vis.

Could this link work for you?https://drive.google.com/drive/folders/1mxvjWqdYfzzdTNJPP9aOiJHnnFoWwMNv?usp=sharing
Thanks a lot

from swan_vis.

Nope, I don't have access to that. Just the first few lines copied here should work.

from swan_vis.

I figure out the reason of that error I think. When SQANTI3 generate the abundance matrix, the third column is p value. I removed it using excel, when I forgot to change the format. When I use the corrected tab separate file, the problem solved. Thank you so much for the wonderful script. The figure is very beautiful.

from swan_vis.

Oh perfect, glad it worked for you! Please don't hesitate to ask if you have any more questions :)

from swan_vis.