how to handle pre-digested peptides in fasta

[heejongkim]

Hi,

I got into a situation where I have a fasta file with the following format:

>PEPTIDE
PEPTIDE

and I set no enzyme with 0 miscleavage with the following command:

java -jar fdrbench-0.0.1/fdrbench-0.0.1.jar -I2L -level peptide -db fasta/first10k.fasta -o fasta/first10k_entrapment_pep.txt -diann -fix_nc c -enzyme 0 -miss_c 0 -debug

and I got the following log with empty entrapment_pep.txt file:

2025-06-24 00:14:22 [INFO ] main.java.FDR.DBGear[getEnzymeByIndex:106] - Use enzyme:NoEnzyme
2025-06-24 00:14:22 [INFO ] main.java.FDR.FDREval[generate_peptide_file:607] - Protein sequences:5000, total unique peptide sequences:0
Fix C terminal amino acid
2025-06-24 00:14:22 [INFO ] main.java.FDR.FDREval[generate_peptide_file:837] - Time used for protein digestion:0 s.
================== parameters ==================
2025-06-24 00:14:22 [INFO ] main.java.FDR.DBGear[getEnzymeByIndex:106] - Use enzyme:NoEnzyme
Enzyme: NoEnzyme
Max allowed missed cleavages: 0
Peptide length range: 7 - 35
Clip N-terminal M: false
Fix N/C terminal amino acid: Only fix C terminal amino acid
Add decoy: false
Convert I to L: true
N fold:1
No shared peptides between target and entrapment proteins when using foreign species: false
Reverse order of generated random peptides:false
Max number of tries for generating decoy peptides: 20
For DIA-NN: false
Use Uniprot format protein ID: false
Global random seed: 2000
================================================

Somehow it still recognizes each entry as a protein and emits 0 unique peptide sequences.
It would be awesome if you can guide me what I did wrong and what's the best arguments/parameters I should use to move forward.
Thank you.

best,
heejong

[wenbostar]

The input fasta file is a peptide level database?

[heejongkim]

Yes!

>PEPTIDE
PEPTIDE

I’m flexible to change the format to meet criteria.

[wenbostar]

Could you please try this new version of FDRBench: https://drive.google.com/file/d/1sc9ZPV9Vc5m4rJkXfAsMXrKa1PcXuxz2/view?usp=sharing?

java -jar fdrbench-0.0.3.jar -I2L -level peptide -db fasta/first10k.fasta -o fasta/first10k_entrapment_pep.txt -fix_nc c -enzyme NoCut

[heejongkim]

thank you so much.
I tested and got the following log:

2025-06-24 12:38:35 [INFO ] main.java.FDR.DBGear[getEnzymeIndexByName:74] - Use enzyme:NoCut
2025-06-24 12:38:35 [INFO ] main.java.FDR.DBGear[getEnzymeByIndex:56] - Use enzyme:NoCut
2025-06-24 12:38:35 [INFO ] main.java.FDR.FDREval[generate_peptide_file:707] - Protein sequences:5000, total unique peptide sequences:4995
Fix C terminal amino acid
Ignore peptide: RRRRRRR, 0
Ignore peptide: RRRRRRS, 0
Ignore peptide: ALLLLLA, 1
Ignore peptide: PPPPPPP, 0
Ignore peptide: PPPPPPY, 0
2025-06-24 12:38:35 [INFO ] main.java.FDR.FDREval[generate_peptide_file:937] - Time used for protein digestion:0 s.
================== parameters ==================
2025-06-24 12:38:35 [INFO ] main.java.FDR.DBGear[getEnzymeByIndex:56] - Use enzyme:NoCut
Enzyme: NoCut
Max allowed missed cleavages: 1
Peptide length range: 7 - 35
Clip N-terminal M: false
Fix N/C terminal amino acid: Only fix C terminal amino acid
Add decoy: false
Convert I to L: true
N fold:1
No shared peptides between target and entrapment proteins when using foreign species: false
Reverse order of generated random peptides:false
Max number of tries for generating decoy peptides: 20
For DIA-NN: false
Use Uniprot format protein ID: false
Global random seed: 2000
================================================

seemingly it's working and I have a couple of questions related to this log.

1. Even if I set -diann, I'm still getting For DIA-NN: false. Is is something I should be concerned about? I'm thinking of feeding this into DiaNN with no enzymatic digestion and circle back to this.
2. what's the criteria that caused "Ignore peptide" above?
   Thank you so much.

best,
heejong

[wenbostar]

1. Even if I set -diann, I'm still getting For DIA-NN: false. Is is something I should be concerned about? I'm thinking of feeding this into DiaNN with no enzymatic digestion and circle back to this.

Since your input is a peptide fasta file, it should work fine without adding -diann. This option works when the input is a protein level fasta file from uniprot.

2. what's the criteria that caused "Ignore peptide" above?

An original target peptide for which we fail to generate r (r=1 in this case) distinct shuffles after 20 + r attempts is removed from the original target database as described in the method section of "Entrapment database generation via random sequences" in our paper.

[heejongkim]

Great!
Thank you so much for your clarification.
I will try the fully fasta file with FDRbench and DiaNN and hopefully get back to you soon.
Thanks!

best,
heejong