MPTM: A tool for mining protein post-translational modifications from literature

Area of Science:

• Biomedical Informatics
• Computational Biology
• Molecular Biology

Background:

• Post-translational modifications (PTMs) are crucial in human health and disease.
• The rapid growth of biomedical literature presents challenges for PTM information retrieval.
• Existing PTM text mining tools are limited in scope and methodology.

Purpose of the Study:

• To develop an efficient text mining tool, MPTM, for extracting PTM information from biomedical literature.
• To provide comprehensive PTM data including substrates, enzymes, diseases, and Gene Ontology terms.
• To support researchers and database curators in PTM knowledge discovery.

Main Methods:

• Developed MPTM, a text mining tool utilizing dependency parse trees and a heuristic algorithm.
• Extracted information on 11 common PTMs from PubMed abstracts.
• Integrated protein name recognition, association with UniProtKB, and retrieval of related biological context (diseases, GO terms, organisms).
• First web server for mining hydroxylation, myristoylation, and GPI-anchor modifications.

Main Results:

• MPTM successfully extracts and organizes valuable PTM knowledge.
• Identified novel PTM information through large-scale text analysis.
• Provides access to an online database (MPTMDB) and a downloadable package (MPTM Lite).

Conclusions:

• MPTM offers an efficient solution for PTM literature mining.
• The tool facilitates the discovery of new PTM-related publications and insights.
• MPTM enhances PTM research by providing accessible and organized information.