Next: MULTALIGN file formats Up: APPENDICES Previous: APPENDICES

Summary of Commands for MULTALIGN

Commands may be entered in mixed case. All commands must include the '=' delimiter. All commands may use a maximum of 80 characters including their arguments, all commands are free format.


output_file=STRING

	STRING is the filename to be used for output of results and error
	messages.  Must be the first command issued.


mode=OPTIONS

	OPTIONS are one of MULTIPLE, PAIRWISE or SCAN
	MULTIPLE is the default - signifies 'normal' multiple alignment and
	tree based alignments may be used.
	PAIRWISE signifies conventional pairwise comparisons to be made for
	all sequences in the seq_file.
	SCAN specifies that an alignment or pattern are to be compared to a
	group of sequences.


matrix_file=STRING

	STRING is the filename of the pairscore matrix chosen (one of MD.MAT,
	UP.MAT, GC.MAT etc)


seq_file=STRING

	STRING is the filename of the file containing sequences.


order_file=STRING 

	STRING gives the name of the order file


constant=INT

	INT is a number added to each element of the matrix defined by
	matrix_file.  (Default=8)


gap_penalty=REAL

	REAL is a number used as the length independent gap penalty.
	(Default=8.0).


iterations=INT

	INT defines the number of iterations performed in single order
	multiple alignment.


print_vertical=

	Switch to specify vertical format output of alignments.


print_horizontal=

	Switch to specify horizontal format output of alignments.


print_horizontal/pattern=STRING

	Specify creation of an output file with a suitable format for program
	PATT.  STRING is the name of the file.


max_horiz=INT

	INT defines the number of residues output per line when the
	print_horizontal flag is set. (Default=100).


show_matrix=

	Flag to specify that the pairscore matrix should be written to the
	output file after any constant has been added.


block_file=STRING,INT

	Specifies the block_file name defined by STRING. INT specifies which
	iteration to read from the block_file. (INT MUST BE PRESENT OR AN 
	ERROR WILL RESULT).


block_file/pattern=STRING,INT

	Specifies a block file with flexible pattern ranges.


consplot=OPTION

	Signals that a conservation profile should be plotted.  Only works if
	the print_vertical command is set.
	OPTIONS are MZ for conservation profiles according to 
	Zvelebil et al (1987).
	GB for a profile that is based upon the N*(N-1)/2 comparisons possible
	between all pairs of residues at a given alignment position.


gap_factor=REAL

	When aligning to a block_file, this command specifies the factor to be 
	applied to the gap_penalty within the defined regions.


conscore=

	Flag to specify that the alignment will be scored using conservation 
	values (a la Zvelebil et al 1987) instead of using the specified 
	pairscore matrix.


show_lookup=

	Flag to specify that the lookup_table should be written out at each
	cycle of the alignment.  The format produced cannot subsequently be 
	read back in by the program.


save_lookup=STRING

	STRING is the name of a file to save the first generated lookup table 
	to.  Once saved, the program stops.  The lookup table saved in this
	way may be read in by the program (see read_lookup) for subsequent
	processing.


s_lookup_comment=STRING

	May be used in conjunction with save_lookup to specify a title
	for the saved lookup Table.


read_lookup=STRING

	Defines the filename of a lookup table to be read in by the program.
	The Table overrides any table generated by pairscore matrices.


pairwise_random=INT1,INT2,INT3

	Specifies the number of randomizations to be performed in pairwise
	per sequence pair in pairwise mode.
	INT1 = initial number of randomizations,
	INT2 = max number of randomizations,
	INT3 = increment.
	Normally one only requires a single value for randomizations - say 100.
	This would be defined thus:  pairwise_random=100,100,1


pairwise_timer=

	Flag, if present then system statistics are recorded for each pairwise
	alignment.


pairwise_self=

	Flag, if present then in pairwise mode a self comparison is also made
	for each pair of sequences.  (eg. for ABC, AvA, BvB etc.).


pairwise_align_file=STRING

	Specifies that alignments produced in pairwise mode be written to a
	file other than the output_file.


tree_file=STRING

	Specifies the name of the file containing the tree to follow
	for tree based multiple alignment.  Also tells the program to
	perform a tree alignment rather than a single order alignment.


numerate=

	If print_vertical has been defined, then this will number each
	aligned position of the alignment.


non_alignment=

	Flag, signals that no alignment is performed.  This can be useful to
	switch formats from vertical to horizontal, calculate conservation
	values or numerate the sequences. Note, if a block file is being read
	in then a seq_file must also be defined).


database,N= STRING

        STRING specifies the name of the database to scan using the SCAN mode.
	N specifies how many randomisations to perform for each database entry
	- normally do not include a value for N (very expensive in CPU and not
	very useful).


gap_char_dash=

	Specifies that all gaps will be output as dashes '-' rather than 
	spaces.


gap_char_dot= 

	Specifies that all gaps will be output as dots '.' rather than 
	spaces.


use_end_pen=

	Weights gaps at the ends of the alignment (overhangs).  Only effective
	in tree_file alignments.



last_it=
	In single order mode with multiple iterations, this requests that only
	the last iteration alignment is output.  (Normally alignments for
	each iteration are displayed).


scanps_output=
	In scan mode, this outputs a more condensed version of the output file
	which only has a score, id pair on each line rather than the
	verbose output produced by default.  This file must be sorted using
	the Unix sort utility rather than SORTER.  (e.g. for a file
	called RESULTS.SCAN type: 
	sort +0 -1 -n -r < RESULTS.SCAN > RESULTS.SORTED
	If you take the top N results from the .SORTED file, you can
	then extract the sequences using the SELECT program.


gjb@bioch.ox.ac.uk