Skip to content

Source code and data files for manuscript titled "Pharmaceutical patent landscaping: A novel approach to understand patents from the drug discovery perspective"

License

Notifications You must be signed in to change notification settings

Fraunhofer-ITMP/Pharmaceutical-patent-landscaping

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Pharmaceutical patent landscaping: A novel approach to understand patents from the drug discovery perspective

DOI:10.1101/2023.02.10.527980 DOI:10.1016/j.ailsci.2023.100069

The repository contains the code base for generating and analysing patent corpora for drug discovery aspects.

Publication

To understand in depth the applicability of patent landscaping in drug research and discovery, we recommend reading our publication titled Pharmaceutical patent landscaping: A novel approach to understand patents from the drug discovery perspective.

Pharmaceutical patent landscaping: A novel approach to understand patents from the drug discovery perspective. Yojana Gadiya, Philip Gribbon, Martin Hofmann-Apititus, Andrea Zaliani. bioRxiv 2023.02.10.527980; doi:https://doi.org/10.1101/2023.02.10.527980

If you have found this work useful, please consider citing us.

Resource overview

We utilized OrphaNet, an open-source network focused on rare diseases, and the Human Brain Pharmacome (HBP), a biomedical knowledge graph representing the Alzheimer's disease landscape, as the sources for our data for patent corpora generation.

For generation of the patent corpora, we made use of the PEMT tool.

Code reproducibility

To re-run the pipeline for corpus generation, you can use the alzheimer.py and rare_disease.py.

To perform and validate the analysis done, we have deposited a number of notebooks based on the section number found in the manuscript. You can find these notebooks in the notebooks folder.

Analysis overview

With the patent corpora generated, we performed landscaping at various aspects. The overall patent landscaping from a drug discovery perspective can be performed in the following modalities:

  1. Looking into specific patent class annotations to understand what is the scope of the patent within in the corpora ( analysis found in 3_1_patent_overview.ipynb)
  2. Looking into top patent inventors to understand the competitive landscape for the disease of interest (analysis found in 3_2_inventor_overview.ipynb)
  3. Looking into target importance from a market perspective and understanding the prioritisation of future ranking for drug targets (analysis found in 3_3_target_overview.ipynb)
  4. Looking into the drug repurposing scenarios from the patent drug-target relations and possible with legal assistance can be leverage as a resource for "freedom-of-operation" within the patenting world (analysis found in 3_4_clinical_phase_overview.ipynb)

For each of the cases mentioned above, the conclusion on the drug R&D utility would require an additive analysis mentioned in detail in our manuscript.

Issues

If you have difficulties using the codebase, please open an issue here.

Repository structure

|-- README.md
|-- data
|   |-- figures
|   |   |-- figure_1.jpg
|   |   |-- figure_2.png
|   |   |-- figure_3.png
|   |   |-- figure_4.png
|   |   |-- figure_5.png
|   |   |-- figure_6.png
|   |   |-- figure_7.png
|   |   |-- supplementary_figure_1.png
|   |   |-- supplementary_figure_2.png
|   |   |-- supplementary_figure_3A.png
|   |   `-- supplementary_figure_3B.png
|   |-- mapper
|   |   |-- ad_chem_smile_mapper.json
|   |   |-- ad_chemical_max_phase.tsv
|   |   |-- ad_chemical_names.json
|   |   |-- ad_gene_to_chemicals.json
|   |   |-- assignee
|   |   |   |-- README.md
|   |   |   |-- acquired.tsv
|   |   |   |-- individual.tsv
|   |   |   `-- organization.tsv
|   |   |-- orphanet_chem_smile_mapper.json
|   |   |-- orphanet_chemical_max_phase.tsv
|   |   |-- orphanet_chemical_names.json
|   |   `-- orphanet_gene_to_chemicals.json
|   |-- patent_dumps
|   |   |-- ad_chemicals.tsv
|   |   |-- ad_gene_enumerated_patent_data.tsv
|   |   |-- ad_normalized_patent_data.tsv
|   |   |-- ad_patent_data.tsv
|   |   |-- orphanet_chemicals.tsv
|   |   |-- orphanet_gene_enumerated_patent_data.tsv
|   |   |-- orphanet_normalized_patent_data.tsv
|   |   `-- orphanet_patent_data.tsv
|   `-- raw
|       |-- ad_proteins.csv
|       |-- en_product6.xml
|       `-- orphanet_2021.tsv
|-- notebooks
|   |-- 2_1_data_overview.ipynb
|   |-- 3_1_patent_overview.ipynb
|   |-- 3_2_inventor_overview.ipynb
|   |-- 3_2_inventor_overview_academia.ipynb
|   |-- 3_3_target_overview.ipynb
|   |-- 3_4_clinical_phase_overview.ipynb
|   `-- 3_4_modulator_overview.ipynb
|-- requirements.txt
`-- src
    |-- __init__.py
    |-- alzheimers.py
    |-- constants.py
    |-- normalizer.py
    |-- rare_disease.py
    `-- util.py

Funding

This project has been funded by EOSC-Life which has received funding from the European Union's Horizon 2020 programme under grant agreement number 824087.

About

Source code and data files for manuscript titled "Pharmaceutical patent landscaping: A novel approach to understand patents from the drug discovery perspective"

Topics

Resources

License

Stars

Watchers

Forks