/predector

Effector prediction pipeline based on protein properties.

Primary LanguageDockerfileApache License 2.0Apache-2.0

predector

Predict effectors in your proteomes using the Predector!

What's the plan?

The pipeline will run several common tools (and some less common ones) for effector candidate prediction based on protien properties.

We plan to include.

  • Signalp 3, 4, and 5.
  • TMHMM.
  • Phobius.
  • Deepsig.
  • Deeploc and possibly targetp.
  • EffectorP 1, and 2.
  • ApoplastP
  • LOCALIZER
  • Protein similarity searches against effector databases using MMSeqs2 and/or HHblits.
  • Protein structural prediction tools from the RaptorX toolkit, and comparison with structural features of known effectors.
  • Integration of existing experimental custom scores (e.g. RNAseq, Proteomics) to inform candidate weighting.

Ultimately, the pipeline should output a series of effector candidate clusters and scores to prioritise experimental functional characterisation efforts.

Also we intend to inject as many Predator references as possible because predector kind of sounds like predator.

Stay tuned!