Understand the source comparison
Concatenated versus separate searches
Two strategies exist for implementing target-decoy searches, and the choice affects your FDR calculations. Understanding both helps interpret results from different software packages. Concatenated search combines target and decoy databases into a single search
No winner is assigned.
This page preserves a source comparison for education. It does not add a rating, recommendation or clinical judgment.
- Two strategies exist for implementing target-decoy searches, and the choice affects your FDR calculations. Understanding both helps interpret results from different software packages.
- Concatenated search combines target and decoy databases into a single search. Each spectrum competes against all sequences simultaneously. If a decoy peptide scores higher than all target peptides for a given spectrum, the decoy wins and gets reported. This competition means some spectra that might have matched targets will instead match decoys.
- The competition in concatenated searches slightly reduces sensitivity. Some correct matches get beaten by slightly higher-scoring decoy matches. However, this approach provides more accurate FDR estimates because it directly models the competition between true and false matches that occurs in the target database.
- Separate search runs target and decoy searches independently. Each spectrum gets a best match from targets and a best match from decoys. Researchers then compare these parallel results to estimate FDR. This approach preserves all potential target matches but may overestimate FDR slightly.
- Most modern software uses concatenated searches with appropriate corrections. MaxQuant, Proteome Discoverer, and similar platforms implement variations of this approach. The specific implementation matters for absolute FDR values but does not change the fundamental relationship between peptide and protein FDR levels.