One input, five filters. Submit raw protein sequences (whole proteome is fine) — the pipeline applies the exact funnel published for Ciboria shiraiana effector prediction: SignalP (N-terminal signal peptide) → TMHMM (drop internal TM helices) → TargetP (drop mitochondrial) → GPI-anchor filter → EffectorP-fungi 3.0 (final call). Each protein is reported with the stage it passed or the stage that removed it.
Run mode: Model:

Protein sequences (FASTA; one or more)

Loading limits…
Parameters: organismeuk / non-plant SP-region TM slack+5 aa modelfungi threshold0.5 (fixed)
Running pipeline…
0 s

Funnel

proteins surviving each stage

Per-protein verdicts

Run log 

    

  


  

    ⚖
    Submit protein sequences to run the five-stage effector funnel and get a per-protein verdict with evidence.
  







  
What each stage does

  
    
    
  1. SignalP 6.0 (-t euk -m fast) — keeps proteins carrying an N-terminal Sec/SPI signal peptide.
        Proteins without one cannot enter the secretory pathway and are removed immediately (typically >85% of a proteome).
    2. 
    
  2. TMHMM 2.0 — predicts transmembrane topology per residue. A helix that falls inside the signal peptide
        (up to the cleavage site + 5 aa) is not treated as an internal TM; any helix further downstream removes the protein.
    3. 
    
  3. TargetP 2.0 (-org nonpl) — removes proteins whose best target is the mitochondrion (mTP),
        since those are organellar rather than secreted.
    4. 
    
  4. GPI-anchor filter — removes GPI-anchored proteins, which stay tethered to the plasma membrane.
        The published pipeline used PredGPI; that program has no downloadable package, so a rule-based equivalent is used
        and labelled as such in the results.
    5. 
    
  5. EffectorP-fungi 3.0 — the final call: apoplastic, cytoplasmic or dual-localised effectors versus non-effectors,
        from an ensemble of 30 fungal-specific WEKA models.
    6. 
  






  
Citations

  
Teufel F, et al. SignalP 6.0 predicts all five types of signal peptides using protein language models.
    Nature Biotechnology 2022;40:1023–1025.
    doi:10.1038/s41587-021-01156-3

  
Almagro Armenteros JJ, et al. SignalP 5.0 improves signal peptide predictions using deep neural networks.
    Nature Biotechnology 2019;37:420–423.
    doi:10.1038/s41587-019-0036-z

  
Krogh A, Larsson B, von Heijne G, Sonnhammer ELL. Predicting transmembrane protein topology with a hidden
    Markov model: application to complete genomes. Journal of Molecular Biology 2001;305:567–580.

  
Almagro Armenteros JJ, et al. Detecting sequence signals in targeting peptides using deep learning.
    Life Science Alliance 2019;2:e201900429.
    doi:10.26508/lsa.201900429

  
Pierleoni A, Martelli PL, Casadio R. PredGPI: a GPI-anchor predictor.
    BMC Bioinformatics 2008;9:392.
    doi:10.1186/1471-2105-9-392

  
Sperschneider J, Dodds PN. EffectorP 3.0: prediction of apoplastic and cytoplasmic effectors in fungi and oomycetes.
    Molecular Plant-Microbe Interactions 2022;35(2):94–99.
    doi:10.1094/MPMI-08-21-0201-R

  

    Server-side: SignalP 6.0, TMHMM 2.0d and TargetP 2.0 run under academic licences from DTU Health Tech (non-commercial,
    academic use only); EffectorP 3.0 runs with WEKA 3.8.4 on OpenJDK 17. Targets are fungal proteins —
    the pipeline is not applicable to bacterial, nematode or insect effectors.