Motif Search Tool Tatusov, Altschul, Koonin most most MoST program perl $value CAP 0 CAP CAP: Blast alignment parser - must be run first most most (must be run with a CAP result file) You should run: 1. CAP, 2. MoST Program CAP (Consistent Alignment Parser) may be used to construct blocks acceptable for MoST from BLAST outputs. MoST will scan the indicated database iteratively until convergence, by adding segments selected at each iteration to the original block, constructed by CAP. cap CAP Parameters perl ($most eq "CAP") blast_output Blast Output File perl " $value" 1 perl ($most eq "CAP") capoutfile CAP result file perl " $value" 2 will be used as block file for MoST perl ($most eq "CAP") criterium Criterium perl ($value && $value ne $vdef)? " -t=$value" : "" diri 3 diri diri: rank by information content using Dirichlet priors matrix matrix: rank by the average score per residue square square: rank by the product height * width height height: rank by the 'height' of the block Defines the Threshold used to rank blocks of consistent alignment segments . diri - rank by information content using Dirichlet mixture priors to calculate backgrund probabilities (analogous to the default method of weight matrix construction in MoST) . height - rank by the 'height' of the block, i.e. the number of consistent segment longer than 7 aa . square - rank by the product height x width, i.e. number of segments multiplied by the number of amino acids in each of them . matrix - rank by the average score per residue calculated using the BLOSUM62 matrix Default is t=diri perl ($most eq "CAP") noless min height of motif (%) perl (defined $value && $value != $vdef)? " -l$value\\%" : "" 50 3 -l% - construct only blocks that include no less than certain percentage of the sequences form the BLAST output (default is 50%) it's a percentage perl $value < 0 || $value > 100 perl ($most eq "CAP") blast_score min Blast score perl (defined $value && $value != $vdef)? " s=$value" : "" 3 use only alignments that have BLAST scores equal or greater than this perl ($most eq "CAP") mostopt MoST Parameters perl ($most eq "most") aadb Protein Database perl " $value" uniprot 2 uniprot uniprot genpept genpept: Genbank translation genpept_new genpept_new: genpept updates gpbct gpbct: genpept bacteries gppri gppri gpmam gpmam gprod gprod gpvrt gpvrt gpinv gpinv gppln gppln gprna gprna gpvrl gpvrl gpphg gpphg gpest gpest gpsts gpsts gpsyn gpsyn gppat gppat gpuna gpuna nrprot nrprot: Swissprot + Nrl_3d + PIR nrl3d nrl3d: sequences from PDB sbase sbase: annotated domains sequences perl ($most eq "most") block Block file (generated by CAP) perl " $value" 3 perl ($most eq "most") iter how many iterations perl " \\\\#$value" 1 1 perl ($most eq "most") method Method of computation of position-dependent weight matrix from alignment block perl ($value)? " $value" : "" 4 one one: weighted averaging (Gribskov method) two two: Bayesian prediction (pseudocount) method three three: data-dependent pseudocount method This refers to the method of computation of position-dependent weight matrix from alignment block Default is application of Dirichlet mixture priors to calculate the weight for each amino acid. - one is weighted averaging (Gribskov method). - two is Bayesian prediction (pseudocount) method - three is data-dependent pseudocount method perl ($most eq "most") seg apply seg filter perl (! $value)? " -seg" : " seg$segv" 0 4 default is to apply seg filter for each segment found -seg means omit seg seq# - run seg filter with width parameter #(number) perl ($most eq "most") segv width parameter (if segfilter) perl "" 4 perl ($most eq "most") && $seg identical use only one representative of each group of sequence segments that are #% or more identical perl (defined $value)? " i$value\\%" : "" 4 perl ($most eq "most") tmpfile *.mes remost.all