Searching for TFBS with PFM

Task Name: pfm-search

This task searches for transcription factor binding sites (TFBS) using position weight matrices (PWM) converted from input position frequency matrices (PFM) and saves the identified regions as annotations.

Parameters:

• seq — A semicolon-separated list of input sequence files to search for TFBS. [String, Required]

• matrix — A semicolon-separated list of the input PFM files. [String, Required]

• out — The output Genbank file.

• name — The name of the annotated regions. [String, Optional, Default: “misc_feature”]

• type — The type of the matrix. [Boolean, Optional, Default: false]

  The following values are available:

  • true (dinucleic type)
  • false (mononucleic type)

  Dinucleic matrices provide more detail, while mononucleic matrices are more useful for smaller input datasets.

• algo — The algorithm used to convert a PFM to a PWM. [String, Optional, Default: “Berg and von Hippel”]

  The following values are available:

  • Berg and von Hippel
  • Log-odds
  • Match
  • NLG

• score — The minimum percentage score to detect TFBS. [Number, Optional, Default: 85]

• strand — The strands to search in. [Number, Optional, Default: 0]

  The following values are available:

  • 0 (both strands)
  • 1 (direct strand)
  • -1 (complement strand)

Example:

ugene pfm-search --seq=in.fa --matrix=MA0265.1.pfm;MA0266.1.pfm --out=res.gb