New interfaces (#84)

* Interface (#74) * Add prelude #63 * Make a step plot * Add a new interface, Session * Check dimensionality * #19 Support a new keyword, proceed * #19 Enable to extend reserved keywords * Implement ensemble * session.Result is picklable * Use run_ensemble * Move ecell4.util.viz -> ecell4.plotting * ecell4.util.viz will be deprecated * get_range_of_world and get_range_of_trajectories are moved into _core * Copy deep * Add plotting._plotly * Add a datasource for STRING * Add idmapping using UniProt * Interface2 (#75) * Set method * Add more options * Fix silly mistakes * #11 Enable rich display * Session.run accepts numpy.ndarray as t * Arithmetic representations in y * Enable to set own observable in _plotly * Split elegans-based plot functions into _elegans * Format before assignment * Rename plotting.default -> BACKEND * Enable rule-based descriptions in observables * Enable observables in plotting.plot_number_observer (#77) * Session.run accepts numpy.ndarray as t * Arithmetic representations in y * Enable to set own observable in _plotly * Split elegans-based plot functions into _elegans * Format before assignment * Rename plotting.default -> BACKEND * Enable rule-based descriptions in observables * Load submodules * Skip some submodules * Remove a duplicated file * Remove all * Update version * Import all from submodules * Set __all__ * Rather use the common name * Update the relative module path * Move display_anim into _core * Enable ndiv * Fix for pi * Enable to change figsize in _matplotlib * Add DOI * Add citation * Update the citation outputs * Import citation in prelude * Put a space after a comma * load_world is already imported * Cast to an integer * Enable to fetch an abstract * Remove an unnecessary module * Simplify run_simulation and ensemble_simulations, and deprecate old implementations * The config yaml is loaded if existing * Add plot_trajectory in _plotly * Change the default renderer on Google Colab * Try STL visualization in Plotly * Set opacity * Use default colorscale * Update * Enable to show stl * Enable to check dimensionality of units in a model
ecell · May 2, 2020 · bf39870 · bf39870
1 parent 64438e3
commit bf39870
Show file tree

Hide file tree

Showing 28 changed files with 3,600 additions and 2,655 deletions.
diff --git a/README.md b/README.md
@@ -47,8 +47,7 @@ Run this with Jupyter Notebook
 
 ```python
 %matplotlib inline
-from ecell4_base.core import *
-from ecell4 import *
+from ecell4.prelude import *
 
 with species_attributes():
     M | {'dimension': 2}
@@ -60,7 +59,7 @@ run_simulation(
     0.4, y0={'A': 10}, structures={'M': surface},
     solver='spatiocyte', observers=obs, return_type=None)
 
-viz.plot_trajectory(obs, interactive=False)
+show(obs, interactive=False)
 ```
 
 ![png](./samples/hairball.png)

diff --git a/ecell4/__init__.py b/ecell4/__init__.py
@@ -1,3 +1,3 @@
 from ecell4.util import *
 
-__version__ = '0.1.0'
+__version__ = '1.2.0'
diff --git a/ecell4/datasource/__init__.py b/ecell4/datasource/__init__.py
@@ -1,3 +1,5 @@
+# from . import biocyc, biogrid, ecocyc, pdb, psicquic, pubmed.py, rdf, sbml, sparql, uniprot
+
 def print_descriptions(desc):
     for i, entry in enumerate(desc):
         if i > 0:
@@ -56,4 +58,6 @@ def whereis(entity, collections=None):
     print_descriptions(desc)
 
 
-__all__ = ['description', 'whereis']
+from .pubmed import citation
+
+__all__ = ['description', 'whereis', 'citation']
diff --git a/ecell4/datasource/pubmed.py b/ecell4/datasource/pubmed.py
@@ -1,10 +1,7 @@
 import re
+import numbers
 
-try:
-    from urllib.request import Request, urlopen, HTTPError  # Python3
-except ImportError:
-    from urllib2 import Request, urlopen, HTTPError  # Python2
-
+from urllib.request import Request, urlopen, HTTPError  # Python3
 from xml.dom import minidom
 
 
@@ -31,6 +28,7 @@ def description(entity):
         entry.append(('Author(s)', ', '.join(src.data['AuthorList'])))
         entry.append(('Source', src.data['Source']))
         entry.append(('SO', src.data['SO']))
+        entry.append(('DOI', 'https://doi.org/{}'.format(src.data['DOI'])))
         entry.append(('URL', src.link(entity_id)))
         return [entry]
 
@@ -47,8 +45,16 @@ def __init__(self, entity=None):
             data = self.parse_esummary(read_url(url))
             assert len(data) == 1
             self.data = data[0]
+            url = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&id={}&rettype=abstract".format(entity_id)
+            abstract = self.parse_abstract(read_url(url))
+            if len(abstract) == 0:
+                self.abstract = ""
+            else:
+                assert len(abstract) == 1
+                self.abstract = abstract[0]
         else:
             self.data = None
+            self.abstract = ""
 
     @classmethod
     def parse_entity(cls, entity):
@@ -57,6 +63,8 @@ def parse_entity(cls, entity):
         idpttrn = r'\d+'
         uri1 = r'https://www.ncbi.nlm.nih.gov/pubmed/(?P<id>{})'.format(idpttrn)
         uri2 = r'http://identifiers.org/pubmed/(?P<id>{})'.format(idpttrn)
+        if isinstance(entity, numbers.Integral) and entity >= 0:
+            entity = str(entity)
         if isinstance(entity, str):
             if re.match(r'^{}$'.format(idpttrn), entity) is not None:
                 return entity
@@ -87,13 +95,55 @@ def parse_esummary(cls, esummary):
             entry['ID'] = entry_node.getElementsByTagName('Id')[0].firstChild.data
             for item in entry_node.getElementsByTagName('Item'):
                 name = item.getAttribute('Name')
-                if name in ('Title', 'Volume', 'Issue', 'Pages', 'Source', 'PubDate', 'SO'):
+                if name in ('Title', 'Volume', 'Issue', 'Pages', 'Source', 'PubDate', 'SO', 'DOI', 'FullJournalName'):
                     entry[name] = item.firstChild.data
                 elif name == 'AuthorList':
                     entry['AuthorList'] = [author.firstChild.data for author in item.getElementsByTagName('Item') if author.getAttribute('Name') == 'Author']
             retval.append(entry)
         return retval
 
+    @classmethod
+    def parse_abstract(cls, efetch):
+        retval = []
+        doc = minidom.parseString(efetch)
+        for node in doc.getElementsByTagName('AbstractText'):
+            retval.append(node.firstChild.data)
+        return retval
+
+class Formatter(object):
+
+    def __init__(self, entity):
+        entity_id = PubMedDataSource.parse_entity(entity)
+        if entity_id is None:
+            self.src = None
+        else:
+            self.src = PubMedDataSource(entity)
+
+    @property
+    def abstract(self):
+        return self.src.abstract
+
+    def __str__(self):
+        if self.src is None:
+            return None
+        authors = ', '.join(self.src.data['AuthorList'])
+        year = self.src.data['PubDate'].strip().split(' ')[0]
+        return "{Authors}, {Title} {FullJournalName}, {Issue}({Volume}), {Pages}, {Year}. {DOI}. PubMed MPID: {ID}.".format(Authors=authors, Year=year, **self.src.data)
+
+    def _ipython_display_(self):
+        if self.src is None:
+            return
+        from IPython.display import display, Markdown
+        authors = ', '.join(self.src.data['AuthorList'])
+        year = self.src.data['PubDate'].strip().split(' ')[0]
+        doi_url = 'https://doi.org/{}'.format(self.src.data['DOI'])
+        url = self.src.link(self.src.data['ID'])
+        text = "{Authors}, {Title} *{FullJournalName}*, **{Issue}**({Volume}), {Pages}, {Year}. [{DOI}]({DOI_URL}). PubMed PMID: [{ID}]({URL}).".format(Authors=authors, Year=year, DOI_URL=doi_url, URL=url, **self.src.data)
+        display(Markdown(text))
+
+def citation(entity, formatter=Formatter):
+    return formatter(entity)
+
 
 if __name__ == "__main__":
     print(description("8752322"))
diff --git a/ecell4/datasource/string.py b/ecell4/datasource/string.py
@@ -0,0 +1,60 @@
+import urllib
+import time
+
+import requests
+
+
+class StringDataSource:
+    """
+    See http://string-db.org/cgi/help.pl
+
+    """
+
+    wait = 1.0
+
+    def __init__(self):
+        pass
+
+    @classmethod
+    def network(cls, entity, output_format='highres_image', **optional_parameters):
+        """
+        Display STRING network image.
+
+        Parameters
+        ----------
+        output_format : str, optional
+            'image' or 'highres_image'.
+
+        """
+        if isinstance(entity, str):
+            identifiers = [entity]
+        else:
+            raise TypeError(f'The given identifier [{entity}] is invalid.')
+
+        parameters = {'identifiers': '\r'.join(identifiers)}
+        parameters.update(optional_parameters)
+
+        url = "https://string-db.org/api/{}/network?{}".format(
+                output_format, urllib.parse.urlencode(parameters))
+        response = requests.get(url)
+        time.sleep(cls.wait)
+
+        from IPython.display import Image, display_png
+        display_png(Image(response.content))
+
+    @classmethod
+    def interaction_partners(cls, entity, **optional_parameters):
+        if isinstance(entity, str):
+            identifiers = [entity]
+        else:
+            raise TypeError(f'The given identifier [{entity}] is invalid.')
+
+        parameters = {'identifiers': '\r'.join(identifiers)}
+        parameters.update(optional_parameters)
+
+        output_format = 'json'
+        url = "https://string-db.org/api/{}/interaction_partners?{}".format(
+                output_format, urllib.parse.urlencode(parameters))
+        response = requests.get(url)
+        time.sleep(cls.wait)
+        return response.json()
diff --git a/ecell4/datasource/uniprot.py b/ecell4/datasource/uniprot.py
@@ -1,4 +1,6 @@
 import re
+import urllib.parse
+import urllib.request
 
 from rdflib import Namespace
 from rdflib.namespace import RDF, RDFS, SKOS
@@ -55,6 +57,36 @@ def whereis(entity):
                 desc.extend(description(topology))
     return desc
 
+def idmapping(query, source='ACC+ID', target='ACC'):
+    """https://www.uniprot.org/help/api_idmapping"""
+
+    if isinstance(query, str):
+        query = [query]
+    query = ' '.join(query)
+
+    url = 'https://www.uniprot.org/uploadlists/'
+    params = {
+            'from': source,
+            'to': target,
+            'format': 'tab',
+            'query': query,
+            }
+    data = urllib.parse.urlencode(params)
+    data = data.encode('utf-8')
+    req = urllib.request.Request(url, data)
+    with urllib.request.urlopen(req) as f:
+        response = f.read()
+    content = response.decode('utf-8')
+
+    import csv
+    import io
+    buffer = io.StringIO(content)
+    reader = csv.reader(buffer, delimiter='\t')
+    next(reader)
+    ret = [tuple(row) for row in reader]
+    buffer.close()
+    return ret
+
 class UniProtDataSourceBase(rdf.RDFDataSourceBase):
 
     GRAPH = {}
@@ -221,6 +253,10 @@ def link(cls, entity):
         assert entity_id is not None
         return 'http://www.uniprot.org/uniprot/{}'.format(entity_id)
 
+    @property
+    def identifier(self):
+        return self.parse_entity(self.url)
+
     def mnemonic(self):
         return [str(obj) for obj in self.graph.objects(predicate=self.UNIPROT.mnemonic)]  #XXX: Specify its subject