From 86ab6da4331613fa8bf7fa0d6399797541e04978 Mon Sep 17 00:00:00 2001
From: Florian Jochens <fj@andaco.de>
Date: Mon, 29 Nov 2021 12:01:12 +0100
Subject: [PATCH] added updated input files
---
count_journal.py | 40 ++++++++++++
example_input.py | 8 ++-
input/README.md | 19 +++++-
input/__init__.py | 6 --
input/get/__init__.py | 7 ---
input/get/acs.py | 42 +++++++++++--
input/get/journal_fetcher.py | 42 +++++++++++++
input/get/nature.py | 2 +-
input/get/template_.py | 3 +-
input/interface.py | 50 +++++++++------
input/publication.py | 78 ++++++-----------------
input/test/__init__.py | 2 -
input/test/test_acs.py | 119 ++++++++++++++++++-----------------
input/test/test_input.py | 28 ++++++---
input/test_doi.txt | 4 --
15 files changed, 273 insertions(+), 177 deletions(-)
create mode 100755 count_journal.py
mode change 100755 => 100644 input/test/test_acs.py
delete mode 100644 input/test_doi.txt
diff --git a/count_journal.py b/count_journal.py
new file mode 100755
index 0000000..13886a2
--- /dev/null
+++ b/count_journal.py
@@ -0,0 +1,40 @@
+#!/usr/bin/env python3
+
+from input.interface import InputInterface as Input
+
+def count_journals(url: str):
+ inter = Input()
+ pub = inter.get_publication(url)
+
+ if pub.citations:
+ for citation in pub.citations:
+ journal = citation.journal
+ if journal in cit:
+ cit[journal] += 1
+ else:
+ cit[journal] = 1
+
+ if pub.references:
+ for reference in pub.references:
+ journal = reference.journal
+ if journal in cit:
+ cit[journal] += 1
+ else:
+ cit[journal] = 1
+
+if __name__ == "__main__":
+ cit = {}
+
+ count_journals("https://doi.org/10.1021/acs.jcim.1c00203")
+ count_journals("https://doi.org/10.1021/acs.jcim.6b00561")
+ count_journals("https://doi.org/10.1021/acs.jcim.6b00613")
+ count_journals("https://doi.org/10.1021/acs.jcim.1c00917")
+ count_journals("https://doi.org/10.1021/acs.jmedchem.0c01332")
+ #count_journals("https://pubs.acs.org/doi/10.1021/acs.biochem.1c00290")
+ #count_journals("https://pubs.acs.org/doi/10.1021/acsenvironau.1c00007")
+ #count_journals("https://pubs.acs.org/doi/10.1021/acs.biochem.7b01162")
+
+ cit = dict(sorted(cit.items(), key=lambda item: item[1]))
+ for journal in cit:
+ if journal != "":
+ print(f'{journal}: {cit[journal]}')
diff --git a/example_input.py b/example_input.py
index febbc7b..76eede0 100755
--- a/example_input.py
+++ b/example_input.py
@@ -3,8 +3,10 @@
from input.interface import InputInterface as Input
def main(url: str):
- print(Input.get_publication(url))
+ i = Input()
+ print(i.get_publication(url))
+ # print(i.get_supported_fetchers()) Useless because all classes are called the same
if __name__ == "__main__":
- #main("https://doi.org/10.1021/acs.jcim.1c00203")
- main("https://pubs.acs.org/doi/10.1021/acs.jcim.5b00332")
+ #main("https://doi.org/10.1021/acs.jcim.1c0023")
+ main("https://doi.org/10.1021/acs.jcim.5b00332")
diff --git a/input/README.md b/input/README.md
index 7776ee0..110ce69 100644
--- a/input/README.md
+++ b/input/README.md
@@ -9,8 +9,9 @@ from input.interface import InputInterface as Input
from input.publication import Publication
def main(url):
+ inter = Input()
try:
- pub = Input.get_publication(url)
+ pub = inter.get_publication(url)
except Exception as error:
raise error
@@ -18,10 +19,24 @@ def main(url):
pub.title = "Cool new Title"
print(pub)
-if __name__=="__main__":
+if __name__ == "__main__":
main("https://doi.org/10.1021/acs.chemrev.8b00728")
```
+The expected results of calling this methode are:
+| Input-Url | Result |
+|-----------|-----------|
+| supported & correct| A publication Instance |
+| supported & uncorrect| ValueError|
+| not supported | ValueError|
+
+Supported Url are urls, which comply with the url-pattern of supported Journals.
+
+### Supported Journals:
+
+- ACS-Journals
+- (Nature-Journals)
+
## Testing
``` c
diff --git a/input/__init__.py b/input/__init__.py
index 428d906..e69de29 100644
--- a/input/__init__.py
+++ b/input/__init__.py
@@ -1,6 +0,0 @@
-"""
-init.py for Input-Package.
-"""
-
-from input.publication import Publication
-from input.interface import InputInterface
diff --git a/input/get/__init__.py b/input/get/__init__.py
index 2a6ddd0..e69de29 100755
--- a/input/get/__init__.py
+++ b/input/get/__init__.py
@@ -1,7 +0,0 @@
-#!/usr/bin/env python3
-"""
-__init__ for journalFetcher-module
-temp file with nothing in it right now
-"""
-from input.publication import Publication
-from input.get.journal_fetcher import JournalFetcher
diff --git a/input/get/acs.py b/input/get/acs.py
index cfc1d6d..892cf98 100755
--- a/input/get/acs.py
+++ b/input/get/acs.py
@@ -27,6 +27,8 @@ class Fetcher(JournalFetcher):
TODO: Support non Doi-urls
"""
matched_url = re.match(r'^(https?://)?(doi.org/|pubs.acs.org/doi/)?(10.(\d{4})/\w+.\S+)', url.strip(". \t\r\n"))
+
+ #Checks if match exists
if matched_url is not None:
return matched_url[4] in Fetcher.SUPPORTED_JOURNALS
else:
@@ -40,8 +42,25 @@ class Fetcher(JournalFetcher):
"""
# Creation of Soup
- soup = JournalFetcher.get_soup(url)
+ try:
+ soup = JournalFetcher.get_soup(url)
+ except Exception as error:
+ raise error
+
+ # Raise Error if re recognizes Pattern, but url isnt correct:
+ # For other Urls
+ if soup.text.strip(" \t\n")=="Missing resource null":
+ raise ValueError("'{}' matches Pattern for 'ACS', but doesnt link to Paper.".format(url))
+
+ # For Dois
+ if soup.title is not None:
+ if soup.title.text == "Error: DOI Not Found":
+ raise ValueError("'{}' matches Pattern for 'ACS', but doesnt link to Paper.".format(url))
+
+
soup_header = soup.select('.article_header')[0]
+
+ #Could be used for more specific search
ref_cit_soup = soup
# Creates Publication
@@ -54,6 +73,11 @@ class Fetcher(JournalFetcher):
journal = soup_header.select(".cit-title")[0].text
+ # Replaces abbreviation with whole name
+ if journal in JournalFetcher.abbrev_dict:
+ journal = JournalFetcher.abbrev_dict[journal]
+
+
published = soup_header.select(".pub-date-value")[0].text
subjects = []
@@ -61,8 +85,6 @@ class Fetcher(JournalFetcher):
for subject in subject_soup.select('a'):
subjects.append(subject.text)
- num_citations = 0
-
references = []
references_soup = ref_cit_soup.select('ol#references')
@@ -77,6 +99,10 @@ class Fetcher(JournalFetcher):
if reference.select('.NLM_article-title') != [] else None
ref_journal = reference.select('i')[0].text\
if reference.select('i') != [] else None
+
+ # Replaces abbreviation with whole name
+ if ref_journal in JournalFetcher.abbrev_dict:
+ ref_journal = JournalFetcher.abbrev_dict[ref_journal]
ref_contributors=[]
for author in reference.select('.NLM_contrib-group'):
@@ -94,9 +120,13 @@ class Fetcher(JournalFetcher):
# No Doi -> No Paper
continue
cit_title = citation.select('.cited-content_cbyCitation_article-title')[0].text\
- if citation.select('.cited-content_cbyCitation_article-title')!= [] else "None"
+ if citation.select('.cited-content_cbyCitation_article-title')!= [] else None
cit_journal = citation.select('.cited-content_cbyCitation_journal-name')[0].text\
- if citation.select('.cited-content_cbyCitation_journal-name') != [] else "None"
+ if citation.select('.cited-content_cbyCitation_journal-name') != [] else None
+
+ # Replaces abbreviation with whole name
+ if cit_journal in JournalFetcher.abbrev_dict:
+ cit_journal = JournalFetcher.abbrev_dict[cit_journal]
cit_contributors =[]
cit_contributors = citation.select('.cited-content_cbyCitation_article-contributors')[0]\
.text.replace("\n", " ").replace("\r", "").split(', ')
@@ -107,4 +137,4 @@ class Fetcher(JournalFetcher):
citations.append(Citation(cit_doi, cit_title, cit_journal, cit_contributors, cit_type = "Citation"))
return Publication(doi_url, title, contributors, journal, published
- , subjects, num_citations, references, citations)
+ , subjects, references, citations)
diff --git a/input/get/journal_fetcher.py b/input/get/journal_fetcher.py
index aad5857..514af1f 100755
--- a/input/get/journal_fetcher.py
+++ b/input/get/journal_fetcher.py
@@ -52,3 +52,45 @@ class JournalFetcher(metaclass=ABCMeta):
Creates a Publication-instance.
"""
raise AttributeError("JournalFetcher for '{}' hasnt implemented 'get_publication()'".format(url))
+
+
+ # A Dictionary, which connects abbreviation to whole journal-name
+ abbrev_dict = {
+ "Nat. Protoc.":"Journal of Natural Products"
+ ,"PLoS Comput. Biol.":"PLoS Computational Biology"
+ ,"PLoS One":"PLoS One"
+ ,"Protein Sci.":"Protein Science"
+ ,"J. Am. Chem. Soc.":"Journal of the American Chemical Society"
+ ,"J. Chem. Phys.":"Journal of Chemical Physics"
+ ,"Appl. Sci.":"Applied Science"
+ ,"Comput. Sci. Eng.":"Computing in Science & Engineering"
+ ,"Beilstein J. Org. Chem.":"Beilstein Journal of Organic Chemistry"
+ ,"Biol. Chem.":"Biological Chemistry"
+ ,"Isr. J. Chem.":"Israel Journal of Chemistry"
+ ,"Nat. Methods":"Nature Methods"
+ ,"Proc. Natl. Acad. Sci. U. S. A.":"Proceedings of the National Academy of Sciences of the United States of America"
+ ,"J. Phys. Chem. B":"Journal of Physical Chemistry B"
+ ,"Carbohydr. Res.":"Carbohydrate Research"
+ ,"J. Chem. Theory Comput.":"Journal of Chemical Theory and Computation"
+ ,"J. Mol. Biol.":"Journal of Molecular Biology"
+ ,"Nucleic Acids Res.":"Nucleic Acids Research"
+ ,"J. Comput. Chem.":"Journal of Computational Chemistry"
+ ,"J. Cheminf.":"Journal of Cheminformatics"
+ ,"J. Med. Chem.":"Journal of Medicinal Chemistry"
+ ,"J. Comput.-Aided Mol. Des.":"Journal of Computer-Aided Molecular Design"
+ ,"J. Chem. Inf. Model.":"Journal of Chemical Information and Modeling"
+ ,"Mol. Cell":"Molecular Cell"
+ ,"J. Cell Biolog.":"Journal of Cell Biology"
+ ,"Mol. Cell Biol.":"Molecular and Cellular Biology"
+ ,"J. Cell Sci.":"Journal of Cell Science"
+ ,"Nat. Cell Biol.":"Nature Cell Biology"
+ ,"J. Aerosol Sci. Technol.":"Aerosol Science and Technology"
+ ,"Mol. Biol. Cell":"Molecular Biology of the Cell"
+ ,"Build. Environ.":"Building and Environment"
+ ,"Sci. Rep.":"Scientific Reports"
+ ,"Nat. Chem.":"Nature Chemistry"
+ ,"Nat. Med.":"Nature Medicine"
+ ,"Nat. Commun.":"Nature Communications"
+ ,"Exp. Cell Res.":"Experimental Cell Research"
+ ,"Nat. Chem. Biol.":"Nature Chemical Biology"
+ }
\ No newline at end of file
diff --git a/input/get/nature.py b/input/get/nature.py
index 4d206f4..c50ea0e 100644
--- a/input/get/nature.py
+++ b/input/get/nature.py
@@ -52,7 +52,7 @@ class Fetcher(JournalFetcher):
for subject in soup.head.findAll(attrs={"name": "dc.subject"}):
_subjects.append(subject.get("content"))
- return Publication(_doi_url, _title, _contributors, _journal, _published, _subjects, 0)
+ return Publication(_doi_url, _title, _contributors, _journal, _published, _subjects)
# TODO: Exceptions-handling
# raise ValueException("Cant Fetch: '{}'".format(error))
diff --git a/input/get/template_.py b/input/get/template_.py
index da200bc..58de023 100755
--- a/input/get/template_.py
+++ b/input/get/template_.py
@@ -46,5 +46,6 @@ class Fetcher(JournalFetcher):
# soup = JournalFetcher.get_soup(url)
# doi,title,contributors[],journal,publication_date,subjects[],references[],citations[]
# TODO: Create new Publication-instance
- # return Publication(doi,title,contributors[],journal,publication_date,subjects[],num_citation=None ,references[],citations[])
+ # return Publication(doi_url, title, contributors = [], journal
+ # , publication_date, subjects = [], references = [], citations = [])
return None
\ No newline at end of file
diff --git a/input/interface.py b/input/interface.py
index c0d6df4..8f0af9d 100755
--- a/input/interface.py
+++ b/input/interface.py
@@ -15,11 +15,30 @@ class InputInterface:
Singleton which dynamically imports and manages fetchers
"""
+ instance = None
get_path = None
fetcher_classes=[]
- @staticmethod
- def get_publication(url: str) -> Publication:
+ # '__new__' is called before '__init__' and gives us an instance
+ def __new__(cls, *args, **kwargs):
+
+ # checks if an instance exists and if it doesnt creates one
+ if cls.instance == None:
+ cls.instance = super(InputInterface, cls).__new__(cls,*args, **kwargs)
+
+ return cls.instance
+
+ def __init__(self):
+ # imports all modules
+
+ if self.fetcher_classes ==[]:
+ self.import_fetcher_classes()
+ if self.fetcher_classes ==[]:
+ raise AttributeError("No specific Fetchers where found at: '{}'"
+ .format(self.get_path))
+
+
+ def get_publication(self, url: str) -> Publication:
"""
The interface-method to get a Publication-instance
@@ -29,24 +48,20 @@ class InputInterface:
:type url: str
:return: Publication instance or None if not supported
"""
- # Initializes 'fetcher_classes', the list of imported modules
- if InputInterface.fetcher_classes ==[]:
- InputInterface.get_fetcher_classes()
- if InputInterface.fetcher_classes ==[]:
- raise AttributeError("No specific Fetchers where found at: '{}'"
- .format(InputInterface.get_path))
# Checks if module supports the 'url' and returns a Publication if it does.
for fetcher_class in InputInterface.fetcher_classes:
if fetcher_class.can_use_url(url):
return fetcher_class.get_publication(url)
-
+
# No Module for given url was found
- return None
+ raise ValueError("'{}' is not supported".format(url))
+ def get_supported_fetchers(self):
+ # print(self.fetcher_classes[0].__name__) Useless right now, because all classes are called the same
+ return [a.__name__ for a in self.fetcher_classes]
- @staticmethod
- def get_fetcher_classes():
+ def import_fetcher_classes(self):
"""
Searches in 'get', if there are [a-z]*.py modules (specific Fetchers)
and tries to import them.
@@ -54,24 +69,21 @@ class InputInterface:
"""
# Path to 'get'-package
- InputInterface.get_path = '{}/get'.format(pathlib.Path(__file__).parent.resolve())
+ self.get_path = '{}/get'.format(pathlib.Path(__file__).parent.resolve())
# Searches for modules with given Pattern
fetcher_file_names=[]
- for file in next(walk(InputInterface.get_path), (None, None, []))[2]:
+ for file in next(walk(self.get_path), (None, None, []))[2]:
if re.match(r'[a-z]+.py', file) is not None:
fetcher_file_names.append(file)
- if fetcher_file_names !=[]:
- print("Found following Modules: {}".format(", ".join(fetcher_file_names)))
-
# Tries to import those modules and saves their 'Fetcher'-class
for file in fetcher_file_names:
try:
fetcher_class = importlib.import_module("input.get.{}".format(file[:-3]))
try:
- InputInterface.fetcher_classes.append(fetcher_class.__getattribute__('Fetcher'))
+ self.fetcher_classes.append(fetcher_class.__getattribute__('Fetcher'))
except Exception as error:
- print("Module '{}' does not have a 'Fetcher'-class".format(file[:-3]))
+ ImportError("Module '{}' does not have a 'Fetcher'-class".format(file[:-3]))
except Exception:
raise ImportError("Module '{}' can not be imported".format(file[:-3]))
diff --git a/input/publication.py b/input/publication.py
index 6de2373..fc512e7 100755
--- a/input/publication.py
+++ b/input/publication.py
@@ -1,13 +1,17 @@
#!/usr/bin/env python3
+# this is needed for typing pre python 3.9, this maybe as an large Overhead
+from typing import Any, List
+
+
class Publication:
"""
Represents a Publications
"""
def __init__(self, doi_url: str, title: str \
- , contributors: list[str], journal: str \
- , publication_date: str, subjects: list[str], num_citations: int = None \
- , references: list[any] = None, citations: list[any] = None ):
+ , contributors: List[str], journal: str \
+ , publication_date: str, subjects: List[str]\
+ , references: List[Any] = None, citations: List[Any] = None ):
"""
Parameters
----------
@@ -20,11 +24,11 @@ class Publication:
:param published: date of release
:type published: str
:param subjects: the subject of the Publication
- :type subjects: list[str]
+ :type subjects: List[str]
:param references: the Citation which is been referenced by this Publication
- :type references: list[any]
+ :type references: List[Any]
:param citations: the Citation which references this Publication
- :type citations: list[any]
+ :type citations: List[Any]
:return: None
"""
self.doi_url = doi_url
@@ -41,11 +45,9 @@ class Publication:
self.citations = []
else:
self.citations = citations
- if num_citations is None:
- self.num_citations = len(self.citations)
- else:
- self.num_citations = num_citations # braucht man nicht einfach len(citations)
-
+
+ # For the 'Verarbeitungsgruppe'
+ self.group = None
def __str__(self) -> str:
return ("Title: {}\n"
@@ -55,7 +57,7 @@ class Publication:
"Published on: {}\n"
"Subjects: {}\n"
"References: \n{}\n"
- "Citations: \n{}\n")\
+ "Citations: \n{}")\
.format(self.title, self.doi_url, ", ".join(self.contributors)
, self.journal, self.publication_date
, ", ".join(self.subjects)
@@ -101,39 +103,14 @@ class Publication:
def __eq__(self, other) -> bool:
""" Compares the unique doi_url of two Publications"""
- return self.doi_url == other.doi_url
-
- def print_pub(self):
- print(f'''Article title: {self.title}
-Publication date: {self.publication_date}
-DOI-URL: {self.doi_url}
-
-Subjects:''')
- print(*(self.subjects), sep = ", ")
- print('\nContributors:')
- print(*(self.contributors), sep = ", ")
-
- if int(self.num_citations) > 0:
- if int(self.num_citations) == 1:
- print(f'\nThis publication is cited by the following publication:\n')
- else:
- print(f'\nThis publication is cited by the following {self.num_citations} publications:\n')
- for citation in self.citations:
- print(f'''
- Title: {citation.title}
- Journal: {citation.journal}
- Contributors: {citation.contributors}
- DOI-URL: {citation.doi_url}
- ''')
- else:
- print('\nThis publication is not cited by any other publication.')
-
-
+ if type(self)==type(other):
+ return self.doi_url == other.doi_url
+ return False
class Citation:
def __init__(self, doi_url: str, title: str \
- , journal: str, contributors: list[str] \
+ , journal: str, contributors: List[str] \
, cit_type: str = "Citation"):
"""
Parameters
@@ -143,7 +120,7 @@ class Citation:
:param title: title of the publication
:type title: str
:param contributors: list of all contributors
- :type contributors: list[str]
+ :type contributors: List[str]
:param cit_type: Specifies if Reference or Citation
:type cit_type: str
:return: None
@@ -164,20 +141,3 @@ class Citation:
, self.cit_type, self.doi_url
, self.cit_type, self.journal
, self.cit_type, ", ".join(self.contributors))
-
-
-# This is just a replica of Citations
-class Reference:
- def __init__(self, doi_url: str, title: str, journal: str, contributors: list[str]):
- self.title = title
- self.doi_url = doi_url
- self.journal = journal
- self.contributors = contributors
-
- def __str__(self) -> str:
- return ("\tReferences-Title: {}\n"
- "\tReferences-Doi: {}\n"
- "\tReferences-Journal: {}\n"
- "\tReferences-Contributors: {}")\
- .format(self.title, self.doi_url
- , self.journal, ", ".join(self.contributors))
diff --git a/input/test/__init__.py b/input/test/__init__.py
index 897ddc5..e69de29 100644
--- a/input/test/__init__.py
+++ b/input/test/__init__.py
@@ -1,2 +0,0 @@
-from input.test.test_input import *
-from input.publication import Publication
\ No newline at end of file
diff --git a/input/test/test_acs.py b/input/test/test_acs.py
old mode 100755
new mode 100644
index b187140..e3dfe84
--- a/input/test/test_acs.py
+++ b/input/test/test_acs.py
@@ -10,68 +10,71 @@ class AcsTestCase(FetcherTestCase):
Methods with test_* will be detected by unittest and run.
"""
- def test_url(self):
+ def test_acs_url(self):
# Positive Testing
- self.can_use_url_test(Acs, "https://doi.org/10.1021/acs.jcim.1c00203", True)
- self.can_use_url_test(Acs, "doi.org/10.1021/acs.jcim.1c00203", True)
- self.can_use_url_test(Acs, "10.1021/acs.jcim.1c00203", True)
- self.can_use_url_test(Acs, " 10.1021/acs.jcim.1c00203", True)
- self.can_use_url_test(Acs, "10.1021/acs.jcim.1c00203 ", True)
- self.can_use_url_test(Acs, "\t 10.1021/acs.jcim.1c00203 \t\n", True)
- self.can_use_url_test(Acs, "https://pubs.acs.org/doi/10.1021/acs.jcim.1c00203", True)
+ self.can_use_url_test(Acs, "https://doi.org/10.1021/acs.jcim.1c00203" , True)
+ self.can_use_url_test(Acs, "doi.org/10.1021/acs.jcim.1c00203" , True)
+ self.can_use_url_test(Acs, "10.1021/acs.jcim.1c00203" , True)
+ self.can_use_url_test(Acs, " 10.1021/acs.jcim.1c00203" , True)
+ self.can_use_url_test(Acs, "10.1021/acs.jcim.1c00203 " , True)
+ self.can_use_url_test(Acs, "\t 10.1021/acs.jcim.1c00203 \t\n" , True)
+ self.can_use_url_test(Acs, "https://pubs.acs.org/doi/10.1021/acs.jcim.1c00203" , True)
# Negative Testing
- self.can_use_url_test(Acs, "", False)
- self.can_use_url_test(Acs, "https://doi.org/10.1038/219021a0", False)
- self.can_use_url_test(Acs, "https://www.nature.com/articles/219021a0", False)
+ self.can_use_url_test(Acs, "" , False)
+ self.can_use_url_test(Acs, "https://doi.org/10.1038/219021a0" , False)
+ self.can_use_url_test(Acs, "https://www.nature.com/articles/219021a0" , False)
self.can_use_url_test(Acs, "https://pubs.acs.org/doi/doi.org/10.1021/acs.jcim.1c00203", False)
- def test_publication(self):
+ def test_acs_publication(self):
url = "https://doi.org/10.1021/acs.jcim.1c00203"
self.get_publication_test(Acs, url, self.expectedPubs[url])
-
+ def test_acs_exceptions(self):
+ test_url= "https://doi.org/10.1021/acs.jcim.1c002"
+ self.get_publication_exception_test(Acs, test_url)
+
# Dictionary of Expected Results, with url
expectedPubs = {
"https://doi.org/10.1021/acs.jcim.1c00203":
Publication(
doi_url = "https://doi.org/10.1021/acs.jcim.1c00203",
title = "AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings",
- contributors = ["Jerome Eberhardt","Diogo Santos-Martins", "Andreas F. Tillack", "Stefano Forli"],
- journal="J. Chem. Inf. Model.",
+ contributors = ["Jerome Eberhardt", "Diogo Santos-Martins", "Andreas F. Tillack", "Stefano Forli"],
+ journal="Journal of Chemical Information and Modeling",
publication_date = "July 19, 2021",
subjects = ["Algorithms","Ligands","Molecules","Receptors","Macrocycles"],
references = [
Citation(doi_url = "https://doi.org/10.1002/jcc.21334"
, title ="AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading"
- , journal="J. Comput. Chem."
+ , journal="Journal of Computational Chemistry"
, contributors=["Trott, O.", "Olson, A. J."]
, cit_type="Reference")
, Citation(doi_url = "https://doi.org/10.1038/nprot.2016.051"
, title ="Computational protein-ligand docking and virtual drug screening with the AutoDock suite"
- , journal="Nat. Protoc."
+ , journal="Journal of Natural Products"
, contributors=["Forli, S.","Huey, R.","Pique, M. E.","Sanner, M. F.","Goodsell, D. S.","Olson, A. J."]
, cit_type="Reference")
, Citation(title = "A semiempirical free energy force field with charge-based desolvation"
, doi_url = "https://doi.org/10.1002/jcc.20634"
- , journal="J. Comput. Chem."
+ , journal="Journal of Computational Chemistry"
, contributors=["Huey, R.","Morris, G. M.","Olson, A. J.","Goodsell, D. S."]
, cit_type="Reference")
, Citation(title="Accelerating autodock4 with gpus and gradient-based local search"
, doi_url="https://doi.org/10.1021/acs.jctc.0c01006"
- , journal="J. Chem. Theory Comput."
+ , journal="Journal of Chemical Theory and Computation"
, contributors=["Santos-Martins, D.","Solis-Vasquez, L.","Tillack, A. F.","Sanner, M. F.","Koch, A.","Forli, S."]
, cit_type="Reference")
, Citation(title="AutoDockFR: Advances in Protein-Ligand Docking with Explicitly Specified Binding Site Flexibility"
, doi_url="https://doi.org/10.1371/journal.pcbi.1004586"
- , journal="PLoS Comput. Biol."
+ , journal="PLoS Computational Biology"
, contributors=["Ravindranath, P. A.","Forli, S.","Goodsell, D. S.","Olson, A. J.","Sanner, M. F."]
, cit_type="Reference")
, Citation(title="Docking flexible cyclic peptides with AutoDock CrankPep"
, doi_url="https://doi.org/10.1021/acs.jctc.9b00557"
- , journal="J. Chem. Theory Comput."
+ , journal="Journal of Chemical Theory and Computation"
, contributors=["Zhang, Y.","Sanner, M. F."]
, cit_type="Reference")
, Citation(title="Fast, accurate, and reliable molecular docking with QuickVina 2"
@@ -81,17 +84,17 @@ class AcsTestCase(FetcherTestCase):
, cit_type="Reference")
, Citation(title="Lessons learned in empirical scoring with smina from the CSAR 2011 benchmarking exercise"
, doi_url="https://doi.org/10.1021/ci300604z"
- , journal="J. Chem. Inf. Model."
+ , journal="Journal of Chemical Information and Modeling"
, contributors=["Koes, D. R.","Baumgartner, M. P.","Camacho, C. J."]
, cit_type="Reference")
, Citation(title="Vina-Carb: Improving Glycosidic Angles during Carbohydrate Docking"
, doi_url="https://doi.org/10.1021/acs.jctc.5b00834"
- , journal="J. Chem. Theory Comput."
+ , journal="Journal of Chemical Theory and Computation"
, contributors=["Nivedha, A. K.","Thieker, D. F.","Makeneni, S.","Hu, H.","Woods, R. J."]
, cit_type="Reference")
, Citation(title="AutoDock VinaXB: implementation of XBSF, new empirical halogen bond scoring function, into AutoDock Vina"
, doi_url="https://doi.org/10.1186/s13321-016-0139-1"
- , journal="J. Cheminf."
+ , journal="Journal of Cheminformatics"
, contributors=["Koebel, M. R.","Schmadeke, G.","Posner, R. G.","Sirimulla, S."]
, cit_type="Reference")
, Citation(title="Vinardo: A Scoring Function Based on Autodock Vina Improves Scoring, Docking, and Virtual Screening"
@@ -101,32 +104,32 @@ class AcsTestCase(FetcherTestCase):
, cit_type="Reference")
, Citation(title="Lennard-Jones potential and dummy atom settings to overcome the AUTODOCK limitation in treating flexible ring systems"
, doi_url="https://doi.org/10.1021/ci700036j"
- , journal="J. Chem. Inf. Model."
+ , journal="Journal of Chemical Information and Modeling"
, contributors=["Forli, S.","Botta, M."]
, cit_type="Reference")
, Citation(title="AutoDock4Zn: an improved AutoDock force field for small-molecule docking to zinc metalloproteins"
, doi_url="https://doi.org/10.1021/ci500209e"
- , journal="J. Chem. Inf. Model."
+ , journal="Journal of Chemical Information and Modeling"
, contributors=["Santos-Martins, D.","Forli, S.","Ramos, M. J.","Olson, A. J."]
, cit_type="Reference")
, Citation(title="A force field with discrete displaceable waters and desolvation entropy for hydrated ligand docking"
, doi_url="https://doi.org/10.1021/jm2005145"
- , journal="J. Med. Chem."
+ , journal="Journal of Medicinal Chemistry"
, contributors=["Forli, S.","Olson, A. J."]
, cit_type="Reference")
, Citation(title="Directional phosphorylation and nuclear transport of the splicing factor SRSF1 is regulated by an RNA recognition motif"
, doi_url="https://doi.org/10.1016/j.jmb.2016.04.009"
- , journal="J. Mol. Biol."
+ , journal="Journal of Molecular Biology"
, contributors=["Serrano, P.","Aubol, B. E.","Keshwani, M. M.","Forli, S.","Ma, C.-T.","Dutta, S. K.","Geralt, M.","Wüthrich, K.","Adams, J. A."]
, cit_type="Reference")
, Citation(title="Covalent docking using autodock: Two-point attractor and flexible side chain methods"
, doi_url="https://doi.org/10.1002/pro.2733"
- , journal="Protein Sci."
+ , journal="Protein Science"
, contributors=["Bianco, G.","Forli, S.","Goodsell, D. S.","Olson, A. J."]
, cit_type="Reference")
, Citation(title="Consensus docking: improving the reliability of docking in a virtual screening context"
, doi_url="https://doi.org/10.1021/ci300399w"
- , journal="J. Chem. Inf. Model."
+ , journal="Journal of Chemical Information and Modeling"
, contributors=["Houston, D. R.","Walkinshaw, M. D."]
, cit_type="Reference")
, Citation(title="DockBench: an integrated informatic platform bridging the gap between the robust validation of docking protocols and virtual screening simulations"
@@ -136,7 +139,7 @@ class AcsTestCase(FetcherTestCase):
, cit_type="Reference")
, Citation(title="A new force field for molecular mechanical simulation of nucleic acids and proteins"
, doi_url="https://doi.org/10.1021/ja00315a051"
- , journal="J. Am. Chem. Soc."
+ , journal="Journal of the American Chemical Society"
, contributors=["Weiner, S. J.","Kollman, P. A.","Case, D. A.","Singh, U. C.","Ghio, C.","Alagona, G.","Profeta, S.","Weiner, P."]
, cit_type="Reference")
, Citation(title="AutoDock Bias: improving binding mode prediction and virtual screening using known protein-ligand interactions"
@@ -146,17 +149,17 @@ class AcsTestCase(FetcherTestCase):
, cit_type="Reference")
, Citation(title="Inhomogeneous Fluid Approach to Solvation Thermodynamics. 1. Theory"
, doi_url="https://doi.org/10.1021/jp9723574"
- , journal="J. Phys. Chem. B"
+ , journal="Journal of Physical Chemistry B"
, contributors=["Lazaridis, T."]
, cit_type="Reference")
, Citation(title="Inhomogeneous fluid approach to solvation thermodynamics. 2. Applications to simple fluids"
, doi_url="https://doi.org/10.1021/jp972358w"
- , journal="J. Phys. Chem. B"
+ , journal="Journal of Physical Chemistry B"
, contributors=["Lazaridis, T."]
, cit_type="Reference")
, Citation(title="Grid inhomogeneous solvation theory: Hydration structure and thermodynamics of the miniature receptor cucurbit[7]uril"
, doi_url="https://doi.org/10.1063/1.4733951"
- , journal="J. Chem. Phys."
+ , journal="Journal of Chemical Physics"
, contributors=["Nguyen, C. N.","Young, T. K.","Gilson, M. K."]
, cit_type="Reference")
, Citation(title="AutoDock-GIST: Incorporating Thermodynamics of Active-Site Water into Scoring Function for Accurate Protein-Ligand Docking"
@@ -166,117 +169,117 @@ class AcsTestCase(FetcherTestCase):
, cit_type="Reference")
, Citation(title="ZINC20—A Free Ultralarge-Scale Chemical Database for Ligand Discovery"
, doi_url="https://doi.org/10.1021/acs.jcim.0c00675"
- , journal="J. Chem. Inf. Model."
+ , journal="Journal of Chemical Information and Modeling"
, contributors=["Irwin, J. J.","Tang, K. G.","Young, J.","Dandarchuluun, C.","Wong, B. R.","Khurelbaatar, M.","Moroz, Y. S.","Mayfield, J.","Sayle, R. A."]
, cit_type="Reference")
, Citation(title="Structural biology-inspired discovery of novel KRAS–PDEδ inhibitors"
, doi_url="https://doi.org/10.1021/acs.jmedchem.7b01243"
- , journal="J. Med. Chem."
+ , journal="Journal of Medicinal Chemistry"
, contributors=["Jiang, Y.","Zhuang, C.","Chen, L.","Lu, J.","Dong, G.","Miao, Z.","Zhang, W.","Li, J.","Sheng, C."]
, cit_type="Reference")
, Citation(title="D3R grand challenge 2015: evaluation of protein–ligand pose and affinity predictions"
, doi_url="https://doi.org/10.1007/s10822-016-9946-8"
- , journal="J. Comput.-Aided Mol. Des."
+ , journal="Journal of Computer-Aided Molecular Design"
, contributors=["Gathiaka, S.","Liu, S.","Chiu, M.","Yang, H.","Stuckey, J. A.","Kang, Y. N.","Delproposto, J.","Kubish, G.","Dunbar, J. B.","Carlson, H. A.","Burley, S. K.","Walters, W. P.","Amaro, R. E.","Feher, V. A.","Gilson, M. K."]
, cit_type="Reference")
, Citation(title="D3R grand challenge 4: blind prediction of protein–ligand poses, affinity rankings, and relative binding free energies"
, doi_url="https://doi.org/10.1007/s10822-020-00289-y"
- , journal="J. Comput.-Aided Mol. Des."
+ , journal="Journal of Computer-Aided Molecular Design"
, contributors=["Parks, C. D.","Gaieb, Z.","Chiu, M.","Yang, H.","Shao, C.","Walters, W. P.","Jansen, J. M.","McGaughey, G.","Lewis, R. A.","Bembenek, S. D.","Ameriks, M. K.","Mirzadegan, T.","Burley, S. K.","Amaro, R. E.","Gilson, M. K."]
, cit_type="Reference")
, Citation(title="D3R Grand Challenge 4: prospective pose prediction of BACE1 ligands with AutoDock-GPU"
, doi_url="https://doi.org/10.1007/s10822-019-00241-9"
- , journal="J. Comput.-Aided Mol. Des."
+ , journal="Journal of Computer-Aided Molecular Design"
, contributors=["Santos-Martins, D.","Eberhardt, J.","Bianco, G.","Solis-Vasquez, L.","Ambrosio, F. A.","Koch, A.","Forli, S."]
, cit_type="Reference")
, Citation(title="Comparison of affinity ranking using AutoDock-GPU and MM-GBSA scores for BACE-1 inhibitors in the D3R Grand Challenge 4"
, doi_url="https://doi.org/10.1007/s10822-019-00240-w"
- , journal="J. Comput.-Aided Mol. Des."
+ , journal="Journal of Computer-Aided Molecular Design"
, contributors=["El Khoury, L.","Santos-Martins, D.","Sasmal, S.","Eberhardt, J.","Bianco, G.","Ambrosio, F. A.","Solis-Vasquez, L.","Koch, A.","Forli, S.","Mobley, D. L."]
, cit_type="Reference")
, Citation(title="Macrocycle modeling in ICM: benchmarking and evaluation in D3R Grand Challenge 4"
, doi_url="https://doi.org/10.1007/s10822-019-00225-9"
- , journal="J. Comput.-Aided Mol. Des."
+ , journal="Journal of Computer-Aided Molecular Design"
, contributors=["Lam, P. C.-H.","Abagyan, R.","Totrov, M."]
, cit_type="Reference")
, Citation(title="Directory of useful decoys, enhanced (DUD-E): better ligands and decoys for better benchmarking"
, doi_url="https://doi.org/10.1021/jm300687e"
- , journal="J. Med. Chem."
+ , journal="Journal of Medicinal Chemistry"
, contributors=["Mysinger, M. M.","Carchia, M.","Irwin, J. J.","Shoichet, B. K."]
, cit_type="Reference")
, Citation(title="Evaluation of AutoDock and AutoDock Vina on the CASF-2013 benchmark"
, doi_url="https://doi.org/10.1021/acs.jcim.8b00312"
- , journal="J. Chem. Inf. Model."
+ , journal="Journal of Chemical Information and Modeling"
, contributors=["Gaillard, T."]
, cit_type="Reference")
, Citation(title="Autodock vina adopts more accurate binding poses but autodock4 forms better binding affinity"
, doi_url="https://doi.org/10.1021/acs.jcim.9b00778"
- , journal="J. Chem. Inf. Model."
+ , journal="Journal of Chemical Information and Modeling"
, contributors=["Nguyen, N. T.","Nguyen, T. H.","Pham, T. N. H.","Huy, N. T.","Bay, M. V.","Pham, M. Q.","Nam, P. C.","Vu, V. V.","Ngo, S. T."]
, cit_type="Reference")
, Citation(title="Development and validation of a genetic algorithm for flexible docking"
, doi_url="https://doi.org/10.1006/jmbi.1996.0897"
- , journal="J. Mol. Biol."
+ , journal="Journal of Molecular Biology"
, contributors=["Jones, G.","Willett, P.","Glen, R. C.","Leach, A. R.","Taylor, R."]
, cit_type="Reference")
, Citation(title="Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy"
, doi_url="https://doi.org/10.1021/jm0306430"
- , journal="J. Med. Chem."
+ , journal="Journal of Medicinal Chemistry"
, contributors=["Friesner, R. A.","Banks, J. L.","Murphy, R. B.","Halgren, T. A.","Klicic, J. J.","Mainz, D. T.","Repasky, M. P.","Knoll, E. H.","Shelley, M.","Perry, J. K."]
, cit_type="Reference")
, Citation(title="Surflex: fully automatic flexible molecular docking using a molecular similarity-based search engine"
, doi_url="https://doi.org/10.1021/jm020406h"
- , journal="J. Med. Chem."
+ , journal="Journal of Medicinal Chemistry"
, contributors=["Jain, A. N."]
, cit_type="Reference")
, Citation(title="A fast flexible docking method using an incremental construction algorithm"
, doi_url="https://doi.org/10.1006/jmbi.1996.0477"
- , journal="J. Mol. Biol."
+ , journal="Journal of Molecular Biology"
, contributors=["Rarey, M.","Kramer, B.","Lengauer, T.","Klebe, G."]
, cit_type="Reference")
, Citation(title="EDock: blind protein–ligand docking by replica-exchange monte carlo simulation"
, doi_url="https://doi.org/10.1186/s13321-020-00440-9"
- , journal="J. Cheminf."
+ , journal="Journal of Cheminformatics"
, contributors=["Zhang, W.","Bell, E. W.","Yin, M.","Zhang, Y."]
, cit_type="Reference")
, Citation(title="DOCK 6: Impact of new features and current docking performance"
, doi_url="https://doi.org/10.1002/jcc.23905"
- , journal="J. Comput. Chem."
+ , journal="Journal of Computational Chemistry"
, contributors=["Allen, W. J.","Balius, T. E.","Mukherjee, S.","Brozell, S. R.","Moustakas, D. T.","Lang, P. T.","Case, D. A.","Kuntz, I. D.","Rizzo, R. C."]
, cit_type="Reference")
, Citation(title="Improving scoring-docking-screening powers of protein–ligand scoring functions using random forest"
, doi_url="https://doi.org/10.1002/jcc.24667"
- , journal="J. Comput. Chem."
+ , journal="Journal of Computational Chemistry"
, contributors=["Wang, C.","Zhang, Y."]
, cit_type="Reference")
, Citation(title="ID-Score: a new empirical scoring function based on a comprehensive set of descriptors related to protein–ligand interactions"
, doi_url="https://doi.org/10.1021/ci300493w"
- , journal="J. Chem. Inf. Model."
+ , journal="Journal of Chemical Information and Modeling"
, contributors=["Li, G.-B.","Yang, L.-L.","Wang, W.-J.","Li, L.-L.","Yang, S.-Y."]
, cit_type="Reference")
, Citation(title="Further development and validation of empirical scoring functions for structure-based binding affinity prediction"
, doi_url="https://doi.org/10.1023/a:1016357811882"
- , journal="J. Comput.-Aided Mol. Des."
+ , journal="Journal of Computer-Aided Molecular Design"
, contributors=["Wang, R.","Lai, L.","Wang, S."]
, cit_type="Reference")
, Citation(title="A knowledge-based energy function for protein- ligand, protein- protein, and protein- DNA complexes"
, doi_url="https://doi.org/10.1021/jm049314d"
- , journal="J. Med. Chem."
+ , journal="Journal of Medicinal Chemistry"
, contributors=["Zhang, C.","Liu, S.","Zhu, Q.","Zhou, Y."]
, cit_type="Reference")
, Citation(title="DLIGAND2: an improved knowledge-based energy function for protein–ligand interactions using the distance-scaled, finite, ideal-gas reference state"
, doi_url="https://doi.org/10.1186/s13321-019-0373-4"
- , journal="J. Cheminf."
+ , journal="Journal of Cheminformatics"
, contributors=["Chen, P.","Ke, Y.","Lu, Y.","Du, Y.","Li, J.","Yan, H.","Zhao, H.","Zhou, Y.","Yang, Y."]
, cit_type="Reference")
, Citation(title="Comparing AutoDock and Vina in ligand/decoy discrimination for virtual screening"
, doi_url="https://doi.org/10.3390/app9214538"
- , journal="Appl. Sci."
+ , journal="Applied Science"
, contributors=["Vieira, T. F.","Sousa, S. F."]
, cit_type="Reference")
, Citation(title="Benchmark of four popular virtual screening programs: construction of the active/decoy dataset remains a major determinant of measured performance"
, doi_url="https://doi.org/10.1186/s13321-016-0167-x"
- , journal="J. Cheminf."
+ , journal="Journal of Cheminformatics"
, contributors=["Chaput, L.","Martinez-Sanz, J.","Quiniou, E.","Rigolet, P.","Saettel, N.","Mouawad, L."]
, cit_type="Reference")
, Citation(title="Array programming with NumPy"
@@ -286,7 +289,7 @@ class AcsTestCase(FetcherTestCase):
, cit_type="Reference")
, Citation(title="Matplotlib: A 2D graphics environment"
, doi_url="https://doi.org/10.1109/mcse.2007.55"
- , journal="Comput. Sci. Eng."
+ , journal="Computing in Science & Engineering"
, contributors=["Hunter, J. D."]
, cit_type="Reference")
], citations = [
diff --git a/input/test/test_input.py b/input/test/test_input.py
index 86fe6ae..b2ca55f 100755
--- a/input/test/test_input.py
+++ b/input/test/test_input.py
@@ -1,6 +1,6 @@
import unittest
from input.get.journal_fetcher import JournalFetcher
-
+from input.interface import InputInterface
from input.publication import Publication
"""
@@ -24,7 +24,18 @@ Publication 6: 'https://doi.org/10.1021/acs.jmedchem.0c01332' # Paper is a PDF
class InterfaceTestCase(unittest.TestCase):
- pass
+ def setUp(self):
+ self.assertEqual(InputInterface.instance, None)
+ self.interface = InputInterface()
+
+ def test_singleton(self):
+ # interface should already be made in setUp()
+ self.assertNotEqual(self.interface.instance, None)
+ new_interface = InputInterface()
+ self.assertEqual(self.interface, new_interface)
+
+ # def test_imported_modules(self):
+ # fetchers = self.interface.get_supported_fetchers
class FetcherTestCase(unittest.TestCase):
@@ -45,10 +56,9 @@ class FetcherTestCase(unittest.TestCase):
self.assertEqual(actual_res.journal, expected_res.journal)
self.assertEqual(actual_res.publication_date, expected_res.publication_date)
self.assertEqual(actual_res.subjects, expected_res.subjects)
- # self.assertEqual(actual_res.num_citations, expected_res.num_citations)
# Checking for all references
- self.assertEquals(len(actual_res.references), len(expected_res.references))
+ self.assertEqual(len(actual_res.references), len(expected_res.references))
num_references = len(expected_res.references)
for i in range(num_references):
self.assertEqual(actual_res.references[i].doi_url, expected_res.references[i].doi_url)
@@ -57,7 +67,7 @@ class FetcherTestCase(unittest.TestCase):
self.assertEqual(actual_res.references[i].cit_type, expected_res.references[i].cit_type)
# Checking for all citations
- self.assertEquals(len(actual_res.citations), len(expected_res.citations))
+ self.assertEqual(len(actual_res.citations), len(expected_res.citations))
num_citations = len(expected_res.citations)
for i in range(num_citations):
self.assertEqual(actual_res.citations[i].doi_url, expected_res.citations[i].doi_url)
@@ -66,7 +76,7 @@ class FetcherTestCase(unittest.TestCase):
self.assertEqual(actual_res.citations[i].cit_type, expected_res.citations[i].cit_type)
-
-if __name__=="__main__":
- print("test")
- unittest.main()
\ No newline at end of file
+ def get_publication_exception_test(self, fetcher: JournalFetcher, test_url: str):
+ # Ckecks
+ with self.assertRaises(ValueError):
+ fetcher.get_publication(test_url)
\ No newline at end of file
diff --git a/input/test_doi.txt b/input/test_doi.txt
deleted file mode 100644
index ced8c84..0000000
--- a/input/test_doi.txt
+++ /dev/null
@@ -1,4 +0,0 @@
-https://doi.org/10.1021/acs.jcim.1c00203
-https://doi.org/10.1021/acs.jcim.1c00917
-https://doi.org/10.1021/acs.jmedchem.0c01332
-10.1093/bioinformatics/btaa190
--
GitLab