TL;DR
The dbPTM database has been significantly updated to provide a more comprehensive resource for exploring protein post-translational modifications (PTMs). This latest version integrates a vast amount of new PTM site data, regulatory information, and functional annotations, making it a powerful tool for understanding the complex roles of PTMs in cellular processes and disease.
What Problem Does This Paper Address?
Protein post-translational modifications (PTMs) are crucial for regulating nearly all aspects of cellular life, from protein activity and localization to signaling pathways and disease development. While the importance of PTMs is well-established, the sheer volume of experimental data generated by advanced mass spectrometry (MS) techniques has outpaced the ability of researchers to effectively integrate and analyze this information. Existing databases often lack comprehensive coverage of PTM types, regulatory relationships, and functional annotations, hindering a holistic understanding of PTM networks. This paper addresses the need for an updated, integrated, and user-friendly resource that consolidates this rapidly expanding knowledge base, enabling deeper insights into PTM mechanisms and their implications.
Methods
The dbPTM 2022 update builds upon its previous version by employing a multi-pronged approach to data acquisition and integration:
- Data Curation and Integration: The core of the update involves manually curating literature and integrating data from numerous public biological databases. This includes extracting experimentally verified PTM sites and their corresponding sequences.
- Regulatory Network Construction: To capture the upstream regulation of PTMs, the researchers compiled information on kinases and E3 ligases that modify specific PTM sites. This data was integrated into a protein-protein interaction (PPI) network.
- Functional Annotation Extraction: A text-mining program was developed to extract functional descriptions of PTM sites from literature. This involved sentence tokenization and pattern matching to identify sentences describing PTM functions.
- Database Updates and Resource Integration: Existing PTM-related resources, including annotation databases and prediction tools, were renewed and integrated into a comprehensive resource portal.
Datasets
The dbPTM 2022 update incorporates data from:
- Public Biological Databases: Over 41 published databases (an increase from 30 in the previous version) were integrated. Key examples include UniProtKB/Swiss-Prot, PhosphoSitePlus, and ActiveDriverDB.
- Literature Mining: Approximately 82,444 research articles were manually reviewed to extract PTM sites and regulatory information. This represents a significant increase in the literature coverage compared to the previous version.
- Upstream Regulatory Data: Data from UbiNet 2.0, UniProtKB/Swiss-Prot, GPS 5.0, and PhosphoSitePlus were used to compile kinase-specific phosphorylation sites and E3 ligase-substrate interactions.
Key Insights and Ideas
- The Expanding Landscape of PTMs: The sheer number of newly identified PTM sites and types highlights the dynamic and complex nature of cellular regulation. The inclusion of 42 new modification types underscores the ongoing discovery in this field.
- PTM Crosstalk is Key: The database's enhanced ability to investigate PTM crosstalk (co-occurrences of different PTMs on the same protein) is crucial for understanding how multiple modifications can synergistically or antagonistically regulate protein function.
- Bridging PTMs and Disease: The increased number of disease-associated PTM sites and their integration with genetic data (SNPs, GWAS) provides a powerful resource for understanding the molecular basis of diseases.
- Visualizing Regulatory Networks: The integration of PTM regulatory networks within a PPI framework offers a more intuitive way to visualize how upstream regulators influence PTM events and downstream signaling cascades.
Key Results
- Massive Data Expansion: The updated dbPTM database contains over 2,777,771 PTM sites, a substantial increase of more than 1,520,000 sites from the previous version.
- Experimentally Verified Data: Of the total PTM sites, over 2,235,664 are experimentally verified, providing a high level of confidence in the data.
- Increased PTM Types: The database now covers 76 PTM types, with 42 new types added in this update.
- Extensive Regulatory Information: The database includes 44,753 upstream regulatory relationships, linking modifying enzymes to their PTM substrates.
- Functional Annotations: Over 12,079 records of functional annotations for PTM sites have been integrated, offering insights into the biological roles of specific modifications.
- Disease Association Growth: The number of disease-associated PTM sites has increased significantly, from 350 to 2846.
- PTM Crosstalk Data: 370 PTM pairs exhibiting co-occurrence are now available, facilitating the study of PTM crosstalk.
Limitations and Future Work
- Data Heterogeneity: While efforts are made to map PTM sites to UniProtKB entries, the heterogeneity of data from different sources can still pose challenges.
- Manual Curation Effort: The reliance on manual literature curation, while ensuring accuracy, is a time-consuming process. Future work might explore more advanced automated curation methods.
- Continued Expansion: The authors emphasize that dbPTM will be maintained and updated quarterly, indicating a commitment to continuously incorporating new findings and expanding the database's scope. Future work will likely focus on further enriching the functional annotations, regulatory networks, and disease associations. The development of text-mining-based models for future PTM analysis is also a promising avenue.
Summary updated