FAQ

My search was so big that mokapot ran out of memory

You can replace the mokapot with percolator.

1. Download the standalone version: percolator-noxml.zip
2. Unzip the percolator binary in the folder with your andromeda.tab
3. Run percolator with the --subset-max-train function:

   ./percolator --weights andromeda_weights.csv -Y \
    --testFDR 0.01 \
    --trainFDR 0.01 \
    --results-psms andromeda.mokapot.psms.txt \
    --decoy-results-psms andromeda.mokapot.decoys.psms.txt \
    --results-peptides andromeda.mokapot.peptides.txt \
    --decoy-results-peptides andromeda.mokapot.decoys.peptides.txt \
    --subset-max-train 10000000 \
    andromeda.tab 2> andromeda.log
   

Can I use a custom mapping of peptides to proteins?

Yes, you can specify a mapping file with the --peptide_protein_map <mapping_file> flag. This is a tab separated file without a header row containing peptide sequences (without modifications) in the first column and a semicolon separated list of protein identifiers in the second column. Note that this file should also contain a mapping of peptides for the decoy proteins.

Here is an example file:

GQEGIR  Q6ZSK4
VSSLQGDFPPGGFWALSNDSALSLPPLSLPHPHPLR    Q6ZSK4
PLTSCYHK        REV__Q6ZSK4
PYFCTQFVSAAPHLPPHLGQTFENVGLGPR  REV__Q6ZSK4
MTSGATR Q9Y263;E5RIM3
GLVCCAYPPGAFVSVSR       Q9Y263;H0YAU9;E5RIM3