Traxiv
TraXiv is a simple demonstration of the use of the bioRxiv and hypothes.is APIs to link preprints to the complete peer-review process obtained by EMBO Press.
The posted links can be seen in hypothe.is next to preprints that were published in EMBO Press journals. These posts are aggregated on hypothes.is in the EMBO Press publisher group.
Intro
Traxiv identifies EMBO Press papers for which there is a preprint on bioRxiv. It then generates the links to the respective Review Process Files (RPF) hosted by the journals. The links are then inserted in a template to be posted on hypothes.is and linked to the respective preprints.
It illustrates the use of:
- bioRxiv REST API for publishers: https://api.biorxiv.org/publisher
- the python binding to the hypothe.is API: https://github.com/embo-press/hypothepy
- crossref metadata: https://doi.org
Getting started
Install and activate the python virtual environment:
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
Credentials to your hypothes.is account should be stored in an .env file by editing the provided .env.example file.
.env file since it contains your private personal credentials.
Examples
Example 1
Fetching preprints associated with a publisher identified by its doi prefix:
# embo press is identified by its doi prefix 10.15252
$ python -m src.biorxiv 10.15252 --start 2019-08-01 --end 2019-09-01
# Returns:
#
# bioRxiv request: https://api.biorxiv.org/publisher/10.15252/2019-08-01/2019-09-01/0
# biorxiv_doi published_doi category
# 10.1101/605725 10.15252/embj.2019102870 molecular biology
# 10.1101/528380 10.15252/embj.2019101704 cell biology
Example 2
Posting a simple annotation and deleting it using the toolbox module. To run this demonstartion, you have to have a hypothes.is account and have included your hypothes.is username and api_key (see how to obtain an hypothes.is api key at https://web.hypothes.is/developers/) in the .env file as shown in .env.example.
from src import HYPO, HYPOTHESIS_USER # HYPO is an instance of hypothepy.v1.api.HypoApi that has been initialized with the crendentials of the user.
from src.utils import get_groupid # simple utility to retrieve the hyopthe.is groupid from the group name
from src.toolbox import HypoPost, Target, post_one
groupid = get_groupid('test')
print(groupid)
perms = HYPO.helpers.permissions(
read=[f'group:{groupid}'],
update=[f'acct:{HYPOTHESIS_USER}@hypothes.is'],
delete=[f'acct:{HYPOTHESIS_USER}@hypothes.is'],
admin=[f'acct:{HYPOTHESIS_USER}@hypothes.is']
)
# we want to post the annotation "Great work!" on the target page https://www.embopress.org/journal/17444292
# the annotaiton will have the tag "EMBOpress"
target = Target("https://www.embopress.org/journal/17444292", "Molecular Systems Biology")
post = HypoPost(annotation_text="Great work!", tags=['EMBOpress'])
response = post_one(perms, groupid, target, post)
if response.status_code == 200:
hyp_id = response.json()['id']
print(f"posted {hyp_id}")
# checkout https://www.embopress.org/journal/17444292 and open the hypothesis chrome app to see you annotation (after loging in with the same account)
# checkout https://hypothes.is/groups/<groupid>/test
# now we remove the post using its unique identifier hyp_id assigned by hypothes.is upon posting
response = HYPO.annotations.delete(hyp_id)
if response.status_code == 200:
print(f"deleted {hyp_id}")
# verify that the post has disappeared from https://hypothes.is/groups/<groupid>/test
Example 3
Using a template to generate templated posts.
First, extend the base class HypoPost to add a method that generates the text of the annotation with a simple pyhon string.Template object.
from string import Template
from src.toolbox import HypoPost
class HypoPostTemplated(HypoPost):
"""
Extends HypoPost to generate a templated annotation.
"""
def generate(self, msg: str, template: Template):
"""
Generates the text of the post from a Template.
Arguments:
msg (str): the specific message to be inserted.
template (Template): a simple string.Template object to generate the text of the annotation with appropriate substitution
"""
self.annotation_text = template.substitute({'msg': msg})
self.tags = ['EMBOPress']
Define the template with the appropriate $msg substitution variable:
my_template = Template('**This is a simple demo of templated post.**\n$msg\n---\nIsn\'t it nice?') # this would be normally imported from a file
Create an instance of the HypoPostTemplated object and generate the markdoown text:
my_post = HypoPostTemplated()
my_post.generate('### Great work!\n', my_template)
print(my_post.annotation_text)
Let's post it using the permissions perms, groupid and target from Example 2 above.
response = post_one(perms, groupid, target, my_post)
if response.status_code == 200:
hyp_id = response.json()['id']
print(f"posted {hyp_id}")
Check the nice post at https://www.embopress.org/journal/17444292
And purge:
print(f"deleting {hyp_id}")
response = HYPO.annotations.delete(hyp_id)
if response.status_code == 200:
print(f"deleted {hyp_id}")
Example 4
Automated posting of templated annotations links to hypothes.is:
$ python -m src.traxiv --help
usage: postrpf.py [-h] [--prefixes PREFIXES] [--journals JOURNALS]
[--start START] [--end END] [--group GROUP]
Posts RPF files to hypothes.is
optional arguments:
-h, --help show this help message and exit
--prefixes PREFIXES List of publisher's doi prefixes to retrieve
corresponding preprints. (default: ['10.15252',
'10.26508'])
--journals JOURNALS The journals to scan. (default: ['The EMBO Journal',
'EMBO reports', 'EMBO Molecular Medicine', 'Molecular
Systems Biology', 'Life Science Alliance'])
--start START Start date for the search (format YYYY-MM-DD) (default:
2019-01-01)
--end END End date for the search (format YYYY-MM-DD) (default:
2019-08-25)
--group GROUP Name of the hypothesis group (default: )
Example:
python -m src.traxiv --group test --start 2019-08-01
python -m src.purge test