fasta20.doc

序列对齐 Compare a protein sequence to a protein sequence database or a DNA sequence to a DNA sequenc

RELATE    significance program described by Dayhoff (Atlas of
          Protein Sequence and Structure, Vol. 5, Supplement 3).
          Each chunk of 25 residues in one sequence is compared
          to every 25 residue fragment of the second sequence.
          Sequences which are genuinely related will have a large
          number of scores greater than 3 standard deviations
          above the mean score of all of the comparisons.

3.6.  Other analysis programs

AACOMP    calculate the amino acid composition and molecular
          weight of a sequence.

BESTSCOR  calculate the best self-comparison score.

GREASE    Kyte-Doolittle hydropathicity profile

TGREASE   graphic plot of Kyte-Doolittle profile

FROMGB    convert from GenBank LOCUS format (also used by the
          IBI-Pustell programs) to Pearson/FASTA format.

GARNIER   A secondary structure prediction program using the
          method of Garnier, Osgusthorpe, and Robson, J. Mol.
          Biol., (1978) 120:97-120.

3.7.  Options

     These programs have a number of output options, which are
invoked by the environment variables LINLEN, SHOWALL, and MARKX.
Alternatively, these values can be controlled by command line
options.  The number of sequence residues per output line is now
adjustable by setting the environment variable LINLEN, or the
command line option -w.  LINLEN is normally 60, to change it set
LINLEN=80 before running the program or add -w 80 to the command
line.  LINLEN can be set up to 200.  SHOWALL (-a) determines
whether all, or just a portion, of the aligned sequences are
displayed.  Previously, FASTP would show the entire length of
both sequences in an alignment while FASTN would only show the
portions of the two sequences that overlapped. Now the default is
to show only the overlap between the two sequences, to show
complete sequences, set SHOWALL=1, or use the -a option on the
command line.

     The differences between the two aligned sequences can be
highlighted in three different ways by changing the environment
variable MARKX or the -m option.  Normally (MARKX=0) the program
uses ':' do denote identities and '.' to denote conservative
replacements.  If MARKX=1, the program will not mark identities;
instead conservative replacements are denoted by a 'x' and non-
conservative substitutions by a 'X'.  If MARKX=2, the residues in
the second sequence are only shown if they are different from the
first. MARKX=3 displays the aligned library sequences without the
query sequence; these can be used to build a primitive multiple
alignment.  MARKX=4 provides a graphical display of the
boundaries of the alignments. Thus the five options are:


     MARKX=0      MARKX=1       MARKX=2       MARKX=3      MARKX=4

    MWRTCGPPYT   MWRTCGPPYT    MWRTCGPPYT                 MWRTCGPPYT
    ::..:: :::     xx  X       ..KS..Y...    MWKSCGYPYT   ----------
    MWKSCGYPYT   MWKSCGYPYT


(fasta20u4, Feb. 1996) In addition MARKX=10 is a new, parseable
format for use with other programs.  See the file"readme.v20u4"
for a more complete description.

3.8.  Command line options

     It is now possible to specify  several options on the
command line, instead of using environment variables.  The
command line options are preceded by a dash; the following
options are available:

-a        same as showall=1

-A        force Smith-Waterman alignments for DNA sequences and
          TFASA.  By default, only FASTA protein sequence
          comparisons use Smith-Waterman alignments.

-b #      Number of sequence scores to be shown on output.  In
          the absence of this option, fasta (and tfasta and
          ssearch) display all library sequences obtaining
          similarity scores with expectations less than 10.0 if
          optimized score are used, or 2.0 if they are not. The
          -b option can limit the display further, but it will
          not cause additional sequences to be displayed.

-c #      Threshold score for optimization (OPTCUT).  Set "-c 1"
          to optimize every sequence in a database.  (This slows
          the program down about 5-fold).

-E #      Limit the number of scores and alignments shown based
          on the expected number of scores.  Used to override the
          expectation value of 10.0 used by default.  When used
          with -Q, -E 2.0 will show all library sequences with
          scores with an expectation value <= 2.0.

-d #      Number of alignments to be reported by default. (Used
          in conjunction with -Q).  No longer necessary, see "-b"
          above.

-f        Penalty for the first residue in a gap (-12 by default
          for proteins, -16 for DNA or for TFASTA).

-g        Penalty for additional residues in a gap (-2 by default
          for proteins, -4 for DNA and TFASTA ).

-h        Penalty for frameshift (FASTX, TFASTX only).

-H        Omit histogram.

-i        Invert (reverse complement) the query sequence if it is
          DNA.  For TFASTX, search the reverse complement of the
          library sequence only.

-k #      Threshold for joining init1 segments to build an initn
          score (GAPCUT).

-l file   Location of library menu file (FASTLIBS).

-L        Display more information about the library sequence in
          the alignment.

-m #      MARKX = # (0, 1, 2, 3, 4, 10)

-n        Force the query sequence to be treated as a DNA
          sequence.  This is particularly useful for query
          sequences that contain a large number of ambiguous
          residues, e.g. transcription factor binding sites.

-O        Send copy of results to "filename."  Helpful for
          environments without STDOUT.

-o        Turn off default optimization of all scores greater
          than OPTCUT. Sort results by "initn" scores.

-Q,-q     Quiet - does not prompt for any input.  Writes scores
          and alignments to the terminal or standard output file.

-r file   Save a results summary line for every sequence in the
          sequence library.  The summary line includes the
          sequence identifier, superfamily number (if available)
          position in the library, and the similarity scores
          calculated.  This option can be used to evaluate the
          sensitivity and selectivity of different search
          strategies (see W. R. Pearson (1991) Genomics 11:635-
          650.)

-s file   SMATRIX is read from file.  Several SMATRIX files are
          provided with the standard distribution.  For protein
          sequences: codaa.mat - based on minimum mutation
          matrix; idnaa.mat - identity matrix; pam250.mat - the
          PAM250 matrix developed by Dayhoff et al (Atlas of
          Protein Sequence and Structure, vol. 5, suppl. 3,
          1978); pam120.mat - a PAM120 matrix.  The default
          scoring matrix is BLOSUM50, PAM250 is available with
          "-s 250", BLOSUM62 ("-s BL62") is also available.

-v        (LINEVAL) values used for line styles in plfasta

-w #      Line length (width) = number (<200)

-x        Specifies offsets for the beginning of the query and
          library sequence.  For example, if you are comparing
          upstream regions for two genes, and the first sequence
          contains 500 nt of upstream sequence while the second
          contains 300 nt of upstream sequence, you might try:

              fasta -x "-500 -300" seq1.nt seq2.nt

          If the -x option is not used, FASTA assumes numbering
          starts with 1.  This option will not work properly with
          the translated library sequence with tfasta.  (You
          should double check to be certain the negative
          numbering works properly.)

-y        Set the width of the band used for calculating
          "optimized" scores.  For proteins and ktup=2, the width
          is 16.  For proteins with ktup=1, the width is 32 by
          default.  For DNA the width is 16.

-z        Turn off statistical calculations.

-1        sort output by init1 score (as FASTP used to do).

-3        (TFASTA, TFASTX only) translate only three forward
          frames


For example:

    fasta -w 80 -a seq1.aa seq.aa

would compare the sequence in seq1.aa to that in seq2.aa and
display the results with 80 residues on an output line, showing
all of the residues in both sequences.  Be sure to enter the
options before entering the file names, or just enter the options
on the command line, and the program will prompt for the file
names.

     Not all of these options are appropriate for all of the
programs.  The options above are used by FASTA and TFASTA. RELATE
uses the -s option, ALIGN uses the -w, -m, and -s options, and
the PRDF program uses -c, -f, -k, and -s.

4.  Environment variable summary

     Environment variables allow you to set search parameters
that will be used frequently when you run a program; for example,
if you prefer to use the PAM250 scoring matrix, you might "set
SMATRIX=250."  Command line parameters, if used, always override
environment variable settings. The following environment
variables are used by this program:

AABANK    the file name  of the default sequence library.

FASTLIBS  the location of the file which contains the list of
          library files to be searched.

GAPCUT    threshold used for joining init1 regions in the second
          step of FASTA.  Normally set based on sequence length
          and ktup.

LIBTYPE   used to specify the format of the library sequence for
          FASTA and TFASTA.

LINLEN    output line length - can go up to 200

LINEVAL   used by plfasta to determine the relationship between
          line style and similarity score (-v).  This should be a
          string of three numbers, e.g.  "200 100 50"

MARKX     symbol for denoting matches, mismatches. Note that this
          symbol is only used across the optimized local region;
          sequences that are outside this region are not marked.

OPTCUT    Set the threshold to be used for optimization in a band
          around the best initial region.  Normally the OPTCUT
          value is calculated from the length of the sequence and
          the ktup value (for a 200 residue sequence, it is about
          28).  If OPTCUT=1, every sequence in the database will
          be optimized.  This is the most sensitive option.

PAMFACT   This version of fasta uses a more sensitive method for
          identifying initial regions. Instead of using a
          constant factor (fact) for each match in a ktup, it
          uses the scoring matrix (PAM) scores.  While this works
          well for protein sequences, it has not been as
          carefully tested for DNA sequences, so by default, this
          modification is used for proteins but not for DNA.
          Setting the PAMFACT environment variable to 1 forces
          the option on; PAMFACT=0 turns it off.

SHOWALL   on output, show the complete sequence instead of just
          the overlap of the two aligned sequences.

SMATRIX   alternative scoring matrix file.

TEKPLOT   (IBM-PC only, Unix and VMS versions generate Tektronix
          graphics by default) Generate Tektronix output.
          Normally, PLFASTA and TGREASE plot graphs using the
          Turbo C graphics library.  Unfortunately, often these
          plots cannot be printed out without special programs.
          However, if you set TEKPLOT=1, tektronix graphics
          commands will be used.  Tektronix commands can be used
          together with the PLOTDEV program, available from
          Microplot Systems.  They no lonter sell this program,
          but it can be downloaded from
          http://iquest.com/~microplt/index1.html.  PLOTDEV also
          allows you to print out graphics on the screen.

As always, please inform me of bugs as soon as possible.

William R. Pearson
Department of Biochemistry
Box 440, Jordan Hall
U. of Virginia
Charlottesville, VA

[e-mail]