The application uses one of two methods to calculate the FDRs for the PSMs identified in a search:
- Separated calculation: Calculates the FDRs by using the rank 1 target PSM and the rank 1 decoy PSM for each spectrum. Sorts all target and decoy PSMs by score and determines the score threshold by successively decreasing the score threshold value until the quotient of decoy and target PSMs reaches the specified target FDR.
- Concatenated calculation: Concatenates all target and decoy PSMs of a spectrum in one list sorted by score and assigns the rank of 1 to the best match. Displays this concatenated rank in a separate column on the PSMs page of the results report. With this method, the target and decoy PSMs of a spectrum compete for the rank of 1. Using the concatenated rank, decreases the number of rank 1 decoy PSMs with a high score. The application determines FDR thresholds by using the concatenated calculation method. Because of the smaller number of high-scoring decoy PSMs, the FDR thresholds are lower, and the application identifies more PSMs for the same target FDR.
You can select these methods by using the Concatenated FDR Calculation parameter of the Target Decoy PSM Validator node in the processing workflow or the Peptide Validator node in the consensus workflow.