Hello, I am trying to plot a graph from BioPAX file. <p dir="aut

<a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/us

Hello <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-ur

<a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/us

UnicodeDecodeError: 'charmap' codec can't decode byte 0x8d in position 8815866: character maps to <undefined>,about pybel/pybel

Comments (11)

Rajeswara-Rao commented on June 2, 2024 1

@cthoyt Thanks for your support !

from pybel.

cthoyt commented on June 2, 2024

Hi! Could you please paste the entire stack trace, as well as the version of python, pybel, and Indra that you're using?

If I had to guess without seeing that, it would be that the biopax file is using a nonstandard (ie not utf8) encoding.

from pybel.

Rajeswara-Rao commented on June 2, 2024

Hello, Thanks for your quick reply

I am using python 3.8.3,pybel 0.14.1 and indra 1.18.0

import os
import pybel
from urllib.request import urlretrieve
apoptosis_url = 'https://reactome.org/ReactomeRESTfulAPI/RESTfulWS/biopaxExporter/Level2/1643685'
apoptosis_path = os.path.join(os.path.expanduser('~'), 'Downloads', 'WP254_951071.owl')
if not os.path.exists(apoptosis_path):
    urlretrieve(apoptosis_url, apoptosis_path)
graph = pybel.from_biopax(apoptosis_path)

UnicodeDecodeError                        Traceback (most recent call last)
<ipython-input-2-521428bb8621> in <module>
----> 1 graph = pybel.from_biopax(apoptosis_path)

~\Anaconda3\lib\site-packages\pybel\io\indra.py in from_biopax(path, name, version, description, authors, contact, license, copyright, disclaimer)
    167     from indra.sources.biopax import process_owl
    168 
--> 169     model = process_owl(path)
    170 
    171     return from_indra_statements(

~\Anaconda3\lib\site-packages\indra\sources\biopax\api.py in process_owl(owl_filename)
    171         A BiopaxProcessor containing the obtained BioPAX model in bp.model.
    172     """
--> 173     model = model_from_owl_file(owl_filename)
    174     return process_model(model)
    175 

~\Anaconda3\lib\site-packages\pybiopax\api.py in model_from_owl_file(fname)
     39     """
     40     with open(fname, 'r') as fh:
---> 41         owl_str = fh.read()
     42         return model_from_owl_str(owl_str)
     43 

~\Anaconda3\lib\encodings\cp1252.py in decode(self, input, final)
     21 class IncrementalDecoder(codecs.IncrementalDecoder):
     22     def decode(self, input, final=False):
---> 23         return codecs.charmap_decode(input,self.errors,decoding_table)[0]
     24 
     25 class StreamWriter(Codec,codecs.StreamWriter):

UnicodeDecodeError: 'charmap' codec can't decode byte 0x8d in position 8815866: character maps to <undefined>

from pybel.

cthoyt commented on June 2, 2024

You'll notice that this error is actually happening in a transitive dependency, PyBioPAX, because PyBEL defers to INDRA's BioPAX reader, and INDRA defers to PyBioPAX.

It's likely due to an issue with the default encoding (ref: https://stackoverflow.com/questions/9233027/unicodedecodeerror-charmap-codec-cant-decode-byte-x-in-position-y-character). Are you on Windows?

I've made a PR to PyBioPAX at gyorilab/pybiopax#7 that will be the first step in fixing the issue. After that's accepted, we can make a PR on INDRA to also allow the specification of file encoding from the user-facing API, then update PyBEL to do the same.

Because PyBEL is just wrapping the INDRA code for convenience (and similarly, INDRA is wrapping the PyBioPAX code for convenience), I'll provide you with an example that unwraps and re-implements some of this for immediate use.

import os
from urllib.request import urlretrieve

import indra.sources.biopax
import pybel
import pybiopax

apoptosis_url = 'https://reactome.org/ReactomeRESTfulAPI/RESTfulWS/biopaxExporter/Level2/1643685'
apoptosis_path = os.path.join(os.path.expanduser('~'), 'Downloads', 'WP254_951071.owl')
if not os.path.exists(apoptosis_path):
    urlretrieve(apoptosis_url, apoptosis_path)

with open(apoptosis_path, 'r', encoding='utf-8') as file:
    owl_str = file.read()
    biopax_model = pybiopax.model_from_owl_str(owl_str)

biopax_processor = indra.sources.biopax.process_model(biopax_model)

graph = pybel.from_indra_statements(biopax_processor.statements)

If it fails on the file reading, try some different encodings until you get success.

from pybel.

Rajeswara-Rao commented on June 2, 2024

Thanks for your support!!

The error has resolved now.

from pybel.

cthoyt commented on June 2, 2024

Great! We're working on updating our first-class WikiPathways -> BEL converter which needs a bit of maintenance, and I'm sure it will get better results than the WikiPathways -> Reactome -> INDRA -> BEL loop here once it's ready. Until then, happy coding!

from pybel.

Rajeswara-Rao commented on June 2, 2024

Hello @cthoyt , I was trying to plot a graph using PyBEL after reading a BioPAX file.
I am getting an empty graph by using below code

import os
from urllib.request import urlretrieve
from pybel_jupyter import to_jupyter
import indra.sources.biopax
import pybel
import pybiopax

apoptosis_url = 'https://reactome.org/ReactomeRESTfulAPI/RESTfulWS/biopaxExporter/Level2/1643685'
apoptosis_path = os.path.join(os.path.expanduser('~'), 'Downloads', 'WP254_951071.owl')
if not os.path.exists(apoptosis_path):
    urlretrieve(apoptosis_url, apoptosis_path)

with open(apoptosis_path, 'r', encoding='utf-8') as file:
    owl_str = file.read()
    biopax_model = pybiopax.model_from_owl_str(owl_str)

biopax_processor = indra.sources.biopax.process_model(biopax_model)
graph = pybel.from_indra_statements(biopax_processor.statements)

to_jupyter(graph)

While I am trying to get summary of graph I am getting as

graph.summarize()

indra v64396028-494f-47d8-b3b8-cdfec1102634
Number of Nodes: 0
Number of Edges: 0
Number of Citations: 0
Number of Authors: 0
Network Density: 0.00E+00
Number of Components: 0
Number of Warnings: 0

Please provide me a solution to resolve this problem.
Thanks!

from pybel.

cthoyt commented on June 2, 2024

Unfortunately, problem is not originating from PyBEL, but rather due to the content from Reactome itself. BioPAX is only an upper level ontology and it does not prescribe how information should be encoded. Since Reactome does not use the schema that PyBioPAX expects, PyBioPAX can not parse anything useful from Reactome content. Therefore, no information makes it into INDRA statements via biopax_processor = indra.sources.biopax.process_model(biopax_model) and none into the BEL graph via
graph = pybel.from_indra_statements(biopax_processor.statements).

You can check that this is the case with:

>>> import pybiopax
>>> model = pybiopax.model_from_owl_url('https://reactome.org/ReactomeRESTfulAPI/RESTfulWS/biopaxExporter/Level2/1643685')
>>> len(model.objects)
0

You can also do a check that there are no INDRA statements parsed from the empty PyBioPAX model

>>> len(biopax_processor.statements)
0

You can see that content returned from Pathway Commons does have useful information, though:

>>> import pybiopax
>>> model = pybiopax.model_from_pc_query('pathsfromto', ['MAP2K1'], ['MAPK1'])
>>> len(model.objects)
13022

I would suggest moving questions about BioPAX processing over to the https://github.com/indralab/pybiopax (its maintainer, Ben Gyori, can likely write a better explanation on what's going on with the Reactome content) or to directly contact the Reactome team for support on their BioPAX content

from pybel.

cthoyt commented on June 2, 2024

After looking at the URL, you're using BioPAX level 2. Switch your URL to https://reactome.org/ReactomeRESTfulAPI/RESTfulWS/biopaxExporter/Level3/1643685 and you get something looking much more useful.

I tried it myself and ran into an INDRA-related error with the processing of the resulting BioPAX content. I've posted the issue here: sorgerlab/indra#1184

from pybel.

cthoyt commented on June 2, 2024

@Rajeswara-Rao as of 8202006, it should now be possible to directly specify the file encoding from the PyBEL function pybel.from_biopax with no need to use the wrapped INDRA or PyBioPAX functions. Your use case can now be more simply written as:

import os
from urllib.request import urlretrieve
from pybel_jupyter import to_jupyter
import pybel

apoptosis_url = 'https://reactome.org/ReactomeRESTfulAPI/RESTfulWS/biopaxExporter/Level3/1643685'
apoptosis_path = os.path.join(os.path.expanduser('~'), 'Downloads', 'WP254_951071.owl')
if not os.path.exists(apoptosis_path):
    urlretrieve(apoptosis_url, apoptosis_path)

graph = pybel.from_biopax(apoptosis_path, encoding='utf-8')

to_jupyter(graph)

Caveat This relies on an unmerged PR that's under review on INDRA with PyBEL v15 support, so please sit tight :) then you'll have to be on the development version of both repositories until their next releases

from pybel.

Rajeswara-Rao commented on June 2, 2024

@cthoyt Thanks for the update!

from pybel.

UnicodeDecodeError: 'charmap' codec can't decode byte 0x8d in position 8815866: character maps to <undefined> about pybel HOT 11 CLOSED

Comments (11)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent