3.2.1 align2model - create a multiple alignment of sequences to an existing model R. Hughey, A. Krogh R. Hughey and A. Krogh., SAM: Sequence alignment and modeling software system. Technical Report UCSC-CRL-96-22, University of California, Santa Cruz, CA, September 1996. http://www.cse.ucsc.edu/research/compbio/ismb99.tutorial.html http://www.cse.ucsc.edu/research/compbio/sam.html align2model align2model seqlab align2model perl "align2model" 0 run Run name perl " $value" test 1 db Sequences to align (-db) perl " -db $value" 2 model_file Model (-i) perl ($value)? " -i $value" : "" 2 sam_model perl 1 a2m *.a2m readseq_ok_alig perl 1 input Input options 2 id Sequence identifier(s) selection (separated by commas) (-id) perl ($value && ($value !~ /,/ || $value =~ s/,/ -id /g) ) ? " -id $value " : "" nscoreseq Maximum number of sequences to be read (-nscoreseq) perl (defined $value && $value != $vdef)? " -nscoreseq $value" : "" 100000 control Control options 3 adpstyle Dynamic programming style (-adpstyle perl ($value != $vdef) ? " -adpstyle $value" : "" 1 1 1: Viterbi 4 4: posterior_decoded alignment on transitions and character emissions 5 5: posterior_decoded alignment on only character emissions Alignment methods 4 and 5 are much slower and memory consuming than standard Viterbi alignment. SW Sequence scoring (-SW) perl ($value)? " -SW $value":"" 0 0 0: Global 1 1: semi-local (fragments) 2 2: local (fragments) 3 3: domains perl $adpstyle != 1 . 1: start matching the model at any location (rather than only the begin node) and end at any location (rather than only the end node). This will improve alignment for short sequences that match a segment of the model. . 2: this is similar to Smith and Waterman method of sequence comparison, which will find the best alignment for any pair of subsequences within two sequences. When this is done, sequences can jump from the initial module into the match state of any module in the model, and can also jump out of the match state of any module within the model to the delete state of the next-to-last node. The first and next-to-last module are assumed to be FIMs, hence the rational is that a sequence will use the FIM for some period of time to consume characters that do not match the model, then the sequence will jump to the model node corresponding to the start of the fragment, use several model nodes, and then jump to the ending FIM to consume the rest of the sequence. 3: use for domains. It matchs part of a model to all of a sequence. auto_fim Add FIMs automatically (-auto_fim) perl ($value) ? "" : " -auto_fim 1" 1 jump_in_prob Probability cost of jumping into the center of the model (-jump_in_prob) perl (defined $value && $value != $vdef) ? " -jump_in_prob $value" : "" 1.0 perl $SW jump_out_prob Probability cost of jumping out the center of the model (-jump_out_prob) perl (defined $value && $value != $vdef) ? " -jump_out_prob $value" : "" 1.0 perl $SW output Output options 4 a2mdots Print dots to fill space need for other sequences' insertions (-a2mdots) perl ($value) ? "" : " -a2mdots 0" 1 dump_parameters (-dump_parameters) perl ($value && $value ne $vdef)? " -dump_parameters $value":"" 0 0 0: only modified parameters are printed 1 1: all parameters are printed 2 2: print dump parameters and exit