OpenEBench publications enricher

This is the code repository of the OpenEBench publications enricher used in OpenEBench.

Depending on the chosen flags, it fetches from OpenEBench technical toolbox REST API (which follows next JSON Schema, source from repo here) the list of tools, along with their registered PubMed Id , DOI or PubMedCentral Id. Only tools with such information are considered.

The extracted publication identifiers are validated against the enabled publication repositories (currently supported PubMed, Offline PubMed, EuropePMC and WikiData), gathering for the valid identifiers additional information, like the journal, year, authors, references and citations.

Installation requirements

This program was initially written for Python 3.5 and later. The installation procedure is in INSTALL.md.

Usage

Once the program is installed, and its environment activated, you can see the different options using -h flag:

usage: pubEnricher.py [-h] [--log-file LOGFILENAME] [--log-format LOGFORMAT] [-q] [-v] [-d] [-F] [--fully-annotated]
                      [-b {europepmc,pubmed,wikidata,offline_pubmed,meta}] [-C CONFIG_FILENAME] [--save-opeb SAVE_OPEB_FILENAME]
                      [--use-opeb LOAD_OPEB_FILENAME] (-D RESULTS_DIR | -f RESULTS_FILE | -p RESULTS_PATH)
                      [--format {single,multiple,flat}]
                      [cacheDir]

positional arguments:
  cacheDir              The optional cache directory, to be reused

options:
  -h, --help            show this help message and exit
  --log-file LOGFILENAME
                        Store messages in a file instead of using standard error and standard output
  --log-format LOGFORMAT
                        Format of log messages (for instance %(asctime)-15s - [%(levelname)s] %(message)s)
  -q, --quiet           Only show engine warnings and errors
  -v, --verbose         Show verbose (informational) messages
  -d, --debug           Show debug messages, including URLs (use with care, as it could potentially disclose sensitive contents)
  -F, --full            Return the full gathered citation results, not the citation stats by year
  --fully-annotated     Return the reference and citation results fully annotated, not only the year
  -b {europepmc,pubmed,wikidata,offline_pubmed,meta}, --backend {europepmc,pubmed,wikidata,offline_pubmed,meta}
                        Choose the enrichment backend
  -C CONFIG_FILENAME, --config CONFIG_FILENAME
                        Config file to pass setup parameters to the different enrichers
  --save-opeb SAVE_OPEB_FILENAME
                        Save the OpenEBench content to a file
  --use-opeb LOAD_OPEB_FILENAME
                        Use the OpenEBench content from a file instead of network
  -D RESULTS_DIR, --directory RESULTS_DIR
                        Store each separated result in the given directory
  -f RESULTS_FILE, --file RESULTS_FILE
                        The results file, in JSON format
  -p RESULTS_PATH, --path RESULTS_PATH
                        The path to the results. Depending on the format, it may be a file or a directory
  --format {single,multiple,flat}
                        The output format to be used

The chosen output format may change the way the results are recovered and some flags implemented.

The most prominent change has been the flat format, which implies writing a separate file for each searched tool and found publication, avoiding duplications in the original, nested format. It also generates a manifest.json file, describing the generated files.

Although a config file is not needed to run the program, it is needed to customize its behaviour. A sample config file is available at sample-config.ini, with embedded descriptions.