What it does

KofamScan is a gene function annotation tool based on KEGG Orthology and hidden Markov model.

KofamScan assigns K numbers to the user's sequence data by HMMER/HMMSEARCH against KOfam (a customized HMM database of KEGG Orthologs (KOs)). K number assignments with scores above the predefined thresholds for individual KOs are more reliable than other proposed assignments. Such high score assignments are highlighted with asterisks '*' in the output. The K number assignments facilitate the interpretation of the annotation results by linking the user's sequence data to the KEGG pathways and EC numbers.

Input

• a query file in FASTA format with one or more amino acid sequences. Each sequence must have a unique name. A name of a sequence is a string between the header symbol (">") and the first blank character (whitespace, tab, line break, etc.). Do not put a whitespace right after ">".

  >sp|P00325|ADH1B_HUMAN Alcohol dehydrogenase 1B OS=Homo sapiens GN=ADH1B PE=1 SV=2
  MSTAGKVIKCKAAVLWEVKKPFSIEDVEVAPPKAYEVRIKMVAVGICRTDDHVVSGNLVT
  

• a KO list file of KOfam

  knum    threshold   score_type  profile_type    F-measure   nseq    nseq_used   alen    mlen    eff_nseq    re/pos  definition
  K00001  361.33  domain  trim    0.326825    1601    1149    1538    393 13.33   0.590   alcohol dehydrogenase [EC:1.1.1.1]
  

• KOfam profile files in HMM3 format or a compressed dataset containing HMM3 profiles and HAL filtering files available here.

Output

• KofamScan output
• HMMER results as alignments, run output and tabular summary

References

More information are available on GitHub and the KofamKOALA webserver.