APIR

APIR aggregates peptide-spectrum matches (PSMs) from multiple database search algorithms to increase peptide identification in mass spectrometry (MS) data while controlling the false discovery rate (FDR).

Key Features:

• Universal compatibility: Aggregates outputs from multiple database search algorithms used in MS data analysis.
• Guaranteed FDR control: Provides control of the false discovery rate such that, at a specified FDR threshold, the aggregated set identifies at least as many peptides as any single algorithm.
• Enhanced peptide identification: Increases the number of identified peptides by integrating unique PSMs returned by different algorithms.
• Empirical validation: Validated on a complex proteomics standard where it outperformed individual database search algorithms in peptide identification and FDR control.
• Application to disease research: Enables detection of disease-related proteins and post-translational modifications that may be missed by single algorithms.
• Extensibility: Framework can be extended to aggregate discoveries from multiple algorithms in other high-throughput biomedical analyses such as differential gene expression from RNA sequencing.

Scientific Applications:

• High-throughput proteomics: Combine results from multiple search algorithms to increase depth and accuracy of peptide and protein identification in complex biological samples.
• Biomarker and PTM discovery: Detect disease-related proteins and post-translational modifications that are not consistently reported across individual search algorithms.
• Cross-platform aggregation for other omics: Aggregate algorithmic discoveries in analyses such as differential gene expression using RNA sequencing data.

Methodology:

APIR uses a statistical framework that aggregates PSMs from various database search algorithms and enforces FDR control so that aggregated identifications meet or exceed the peptide identification performance of any individual algorithm under the same FDR threshold.