clusteriser

Clusteriser divides input .fasta file into small groups that are more appropriate to analysis.

What is the program input?

FASTA sequence alignment

What is the program output?

timestamp-folder with FASTA alignments. One file per cluster. {fs|ws|we}_{cluster id} - full sized, ws - without start, we - without end.

Parameters

• -h | --help
• -g | --gap - number of sequences which will be put into 'without beginning' or 'without end'. Need both parameters if used.
• -p | --percent - if used, the previous parameter is used in percents. By default, uses number of chars.
• -c | --coeff - similarity coeff. The minimal threshold for sequences to be in the same group.
• -f | --file - path to the input FASTA alignment
• -r | --register - if used, lowercase letters are considered to be gaps.
• -a | --alphabet - which reduced alphabet to use:
  • SE_B_14
  • SE_B_10
  • SE_V_10
  • LI_A_10
  • LI_B_10
  • SOLIS_D_10
  • SOLIS_G_10
  • MURPHY_19
  • SE_B_8
  • SE_B_6
  • DAYHOFF_6
  • REGULAR (non-reduced)
  • (groups, separated by comma, e.g.: QWER,SD,F. Case insensitive.). In case there is a bug in alphabet description, the REGULAR is used.

Only -f is an obligatory parameter.

Default parameter values

• -g = 50 50
• -c = 0.4
• -r = case insensitive by default
• -a = REGULAR

Example run

java -jar clusteriser.jar -g 10 10 -c 0.5 -f testdata.fasta

After the program has finished the job, the result folder is created. Inside there is a folder with timestamp where the results are.