Do we want our data raw? Including binary mass spectrometry data in public proteomics data repositories

Area of Science:

• Proteomics
• Bioinformatics
• Data Science

Background:

• The Human Plasma Proteome Project (PPP) pilot phase has concluded, marking a significant milestone in proteomics research.
• The large volume of data generated necessitates a robust, centralized data dissemination mechanism.
• Existing proteomics data repositories and infrastructure development are crucial for managing this data.

Purpose of the Study:

• To address the debate on storing raw mass spectrometry (MS) data versus processed peak lists within the PPP.
• To detail the advantages and disadvantages of centralized storage for both raw and processed proteomics data.
• To provide recommendations for immediate and future MS data storage in public repositories.

Main Methods:

• Analysis of data management strategies employed during the PPP pilot phase.
• Evaluation of the merits and caveats of storing raw binary MS data.
• Assessment of the benefits and drawbacks of storing processed peak lists.

Main Results:

• The PPP pilot phase generated a substantial amount of proteomics data, underscoring data management challenges.
• A centralized data gathering infrastructure was developed at the University of Michigan, Ann Arbor.
• The European Bioinformatics Institute established a protein identifications database as a general proteomics data repository.

Conclusions:

• The choice between storing raw data and peak lists has significant implications for proteomics data accessibility and future research.
• Centralized storage of proteomics data, whether raw or processed, is essential for the scientific community.
• Recommendations are proposed for optimizing MS data storage in public repositories for long-term scientific benefit.