PPI Finder: A Mining Tool for Human Protein-Protein Interactions
2009

PPI Finder: A Tool for Mining Protein-Protein Interactions

Sample size: 944 publication Evidence: moderate

Author Information

Author(s): He Min, Wang Yi, Li Wei

Primary Institution: Key Laboratory of Molecular and Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences

Hypothesis

Hybrid methods combining co-occurrence and rule-based approaches may perform better in predicting protein-protein interactions (PPIs).

Conclusion

PPI Finder provides a useful tool for biologists to uncover potential novel PPIs.

Supporting Evidence

  • Only 28% of co-occurred pairs in PubMed abstracts appeared in known PPI databases.
  • 69% of known PPIs in HPRD showed co-occurrences in the literature.
  • PPI Finder uses a hybrid text mining approach combining statistical and computational linguistic methods.

Takeaway

PPI Finder helps scientists find connections between proteins by looking at research papers and finding words that suggest they interact.

Methodology

Developed a web-based tool to mine human PPIs from PubMed abstracts using co-occurrence and interaction words, validated against known PPI databases.

Limitations

The data used are offline and need to be updated regularly; the tool currently only processes literature related to Homo sapiens.

Digital Object Identifier (DOI)

10.1371/journal.pone.0004554

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication