LALNVIEW : A graphical viewer for pairwise alignments



!!!!!!!!!!!!!!!!!!!!!!!!!!  BETA VERSION  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

		Please report bugs to duret@dim.hcuge.ch

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

                       Version 2.2, July 1996




    Laurent Duret
    Medical Biochemistry Department
    Centre Medical Universitaire
    1211 Geneva 4
    Switzerland

    Electronic mail address: duret@dim.hcuge.ch
    WWW server: http://expasy.hcuge.ch/


    Acknowledgments:

    I thank the GLAXO Institute for Molecular Biology (Geneva) for hardware
    and support, Nicolas Guex and Anne Danckaert for their help in porting
    LALNVIEW to Mac and PC.

----------------
Availability:

Anonymous FTP at:     expasy.hcuge.ch       Directory:   /pub/lalnview

----------------
1- Introduction

LALNVIEW is a graphical program for visualizing local alignments between two 
sequences (protein or nucleic acids). Sequences are represented by colored
rectangles to give an overall picture of the similarities between the two
sequences. Blocs of similarity between the two sequences are colored 
according to the degree of identity between the two segments. 

Look at the picture (GIF format) lalnview_1.4.gif to have an idea of what it 
looks like.

By clicking on a bloc, the user can visualize the corresponding local alignment.
A bloc can be repeated in a sequence: iterative clicking on one block will
successively display all the similar blocks that occur in the other sequence.

----------------
2- Running LALNVIEW

LALNVIEW does not calculate the alignments itself: it uses the output of
local alignment programs. LALNVIEW is able to read alignments from three 
widely used softwares: LFASTA, LALIGN and SIM. LFASTA uses the FASTA heuristic 
to quickly find local regions of similarity between two sequences [Pearson & 
Lipman 1988]. SIM and LALIGN are two different implementations of the rigourous 
algorithm of Huang and Miller [Huang and Miller 1991]. The Huang and Miller 
algorithm guarentees to find the N-best local alignments between two sequences. 
Its main drawback is that it is much slower than LFASTA (more than 100-fold 
slower for the comparison of two sequences of 3000 residues each). If sequences 
are relatively short (less than 1000 residues) or if calculation time is not 
limiting, then the Huang and Miller algorithm should be prefered to LFASTA.


SIM sources, executables and documentation are distributed along with LALNVIEW. 
LFASTA and LALIGN are available by anonymous FTP at 

		ftp.virginia.edu
Directory 	
		/pub/fasta

Or
		ftp.bio.indiana.edu
Directory 	
		/molbio/search

The first step is to run a local alignment software (SIM, LALIGN or LFASTA).
The three programs read sequences in FASTA format :

>seq-name    annotation of any sort
ATCGGAGTCGATGGTCACCGNTGGCAC
GTACGTACCGTTGTCCAAACTGTGCAY
...

For example, to find with SIM 7 best local alignments of segments 
from two sequences in files A and B, and using the default values, use the 
command

 sim  7  A  B  > search.out

Or with LALIGN:

 lalign A  B 7 > search.out

Or with LFASTA:

 lfasta -b 7 A  B  > search.out


Then run LALNVIEW :

lalnview search.out &

NB : lalnview needs the two sequence files "A" and "B" to be located in the same
directory as "search.out".

NB: Mac users simply have to drag and drop the search output file on 
    the LALNVIEW icon.

NB: SIM software has not been compiled for Macintosh or PC. To run the 
    alignments, use the ExPaSy WWW server (protein sequences):

	http://expasy.hcuge.ch/sprot/sim-prot.html
    or 	http://expasy.hcuge.ch/sprot/sim-nucl.html

    or the ACNUC WWW server (nucleic acid sequences):

	http://acnuc.univ-lyon1.fr/lfasta.html

    and declare LALNVIEW as helper application on your WWW browser.

----------------
3- Changing similarity score threshold

LALNVIEW displays all the local alignments with a similarity score greater
than a given threshold value. Change the value in the "Similarity Score 
Threshold" box and type <return> to validate this new value. Increase this 
value to display only the most significant blocks of similarity. If you set 
the threshold to 0, LALNVIEW will display all the local alignments that have 
been reported by the local alignment search software.

----------------
4- Saving the alignments in a file

Click on "Save Alignments" button to write the alignments in a file (by default,
this file is named according to the input filename + the extension ".aln". For 
example "search.aln" ). 

WARNING : Only the local alignments with a similarity score greater than the 
current threshold value will be reported.


----------------
5- Creating a postscript picture 

Click on "Create Post-Script" button to save the current picture in a file 
in Post-Script format (by default this file is named according to the input 
filename + the extension ".ps". For example "search.ps" ). 
You can choose to save either a color or grey-scale picture.

NB: Mac users can use the freeware PS2EPS+ to translate the postscript file
    into a Mac readable picture (EPS, PICT, TIFF, PCX). PS2EPS+ is distributed
    along with LALNVIEW.


----------------
6- Adding "Features"

The button "Add Features" allows you to read in a list of segment
coordinates from a file. These segments are then displayed on the screen as 
coloured boxes. Your file with feature segments should 
be in this format:

SeqNumber  Begin End  Width  Color Label

SeqNumber: 	1 for the bottom sequence, 2 for the upper
Begin:		beginning of the feature
End:		end of the feature
		If "Begin" and "End" are equal, then only a line is displayed.
Height:		Height of the box representing the feature.
		any value between 0.0 and 2.2
Color:		color of the box representing the feature.
		135 colors are available (see at the end of this file).
Label:		any annotation < 80 characters

The file should contain a line "# FEATURES BEGIN" and a line
"# FEATURES END". Only the lines between these two ones are considered.

Example:

# FEATURES BEGIN

1   1       149  1.0   blue     Exon
1   1         1  1.0   blue     Transcription start site

2   700     800  1.0   green    Exon
2   800    1000  0.2   blue     Intron
2   850     950  1.0   yellow   Alu repeat
2   1000   1100  1.0   green    Exon
2   1100   1500  0.2   blue     Intron
2   1500   1600  1.0   green    Exon
# FEATURES END

Click on any "feature" box in the picture to see the corresponding annotation.


WHEN USING LALNVIEW THROUGH THE EXPASY OR ACNUC WWW SERVERS, SEQUENCE
FEATURES ARE AUTOMATICALLY EXTRACTED FROM DATABASE ANNOTATIONS.

    ExPaSy WWW server (protein sequences):

	http://expasy.hcuge.ch/sprot/sim-prot.html
    or 	http://expasy.hcuge.ch/sprot/sim-nucl.html

    ACNUC WWW server (nucleic acid sequences):

	http://acnuc.univ-lyon1.fr/lfasta.html


----------------
7- Changing the colors of the similarity scale

Click on any colored rectangle in the similarity scale to change the color 
affected to the corresponding value.

----------------
8- Reading alignment, sequences and features from a single file

LALNVIEW reads data from 3 (or 4) files:
	- SIM (or LFASTA or LALIGN) output file
	- first sequence file
	- second sequence file
	- features description file [optional]

In some cases, it may be convenient to compile all these data in a single file. 
It is possible to concatenate these 3 (or 4) files in a single file:

	- this file must begin with SIM (or LFASTA or LALIGN) output data
	- the first sequence must be preceded by the line:
		"# SEQUENCES BEGIN 1"
	- the second sequence must be preceded by the line:
		"# SEQUENCES BEGIN 2"
	- the second sequence must be followed by the line:
		"# SEQUENCES END"

In UNIX language, this would be done with the following commands:

cat sim_output_file 		>  my_new_file
echo "# SEQUENCES BEGIN 1"	>> my_new_file
cat first_seq_file		>> my_new_file
echo "# SEQUENCES BEGIN 2"	>> my_new_file
cat second_seq_file		>> my_new_file
echo "# SEQUENCES END"		>> my_new_file
cat feature_file		>> my_new_file

Then run LALNVIEW with:

	lalnview my_new_file &

Exemple:
--------------- begin of file ------------------------------------
#:lav
 
d {
  "SIM output with parameters:
  Used PAM200 matrix
  O = 12, E = 4"
}
s {
  "seq_file1" 1 110
  "seq_file2" 1 115
}
k {
  "% match"
}
a {
  s 168.0
  b 11 13
  e 110 115
  l 11 13 52 54 52.4
  l 53 56 64 67 25.0
  l 65 69 87 91 4.3
  l 88 93 110 115 78.3
}
a {
  s 24.0
  b 1 1
  e 24 24
  l 1 1 24 24 33.3
}

# SEQUENCES BEGIN 1
>INS_HUMAN      
MALWMRLLPLLALLALWGPDPAAAFVNQHLCGSHLVEALYLVCGERGFFY
TPKTRREAEDLQVGQVELGGGPGAGSLQPLALEGSLQKRGIVEQCCTSIC
SLYQLENYCN
# SEQUENCES BEGIN 2
>INS_MYXGL
MALSPFLAAVIPLVLLLSRAPPSADTRTTGHLCGKDLVNALYIACGVRGF
FYDPTKMKRDTGALAAFLPLAYAEDNESQDDESIGINEVLKSKRGIVEQC
CHKRCSIYDLENYCN
# SEQUENCES END

# FEATURES BEGIN
1 1 24 0.50 yellow SIGNAL
1 25 54 0.50 yellow CHAIN B CHAIN.
1 57 87 0.50 yellow PROPEP C PEPTIDE.
2 96 97 0.10 black TURN
2 102 108 0.50 blue HELIX
2 109 109 0.50 pink STRAND
# FEATURES END
--------------- end of file ------------------------------------

##########################################################################
Known Bugs:

UNIX :

LALNVIEW first opens a window named "Lost Window" this is just to get the
features of the screen and thus adjust the size of the LALNVIEW window that
is opened afterward. This "Lost Window" remains on the screen until one
exits LALNVIEW.

HP : for some stupid reasons, LALNVIEW produces many error messages 
     when running on HP. To avoid these useless messages run lalnview with
     the following command: 

	lalnview search.out >& /dev/null


##########################################################################


################################
#                              #
#    Available Colors          #
#                              #
################################
#
#	aliceblue	green			navajowhite
#	antiquewhite	greenyellow		navy
#	aquamarine	grey			navyblue
#	azure		honeydew		oldlace
#	beige		hotpink			olivedrab
#	bisque		indianred		orange
#	black		ivory			orangered
#	blanchedalmond	khaki			orchid
#	blue		lavender		palegoldenrod
#	blueviolet	lavenderblush		palegreen
#	brown		lawngreen		paleturquoise
#	burlywood	lemonchiffon		palevioletred
#	cadetblue	lightblue		papayawhip
#	chartreuse	lightcoral		peachpuff
#	chocolate	lightcyan		pink
#	coral		lightgoldenrod		plum
#	cornflowerblue	lightgoldenrodyellow	powderblue
#	cornsilk	lightgray		purple
#	cyan		lightgrey		red
#	darkgoldenrod	lightpink		rosybrown
#	darkgreen	lightsalmon		royalblue
#	darkkhaki	lightseagreen		saddlebrown
#	darkolivegreen	lightskyblue		salmon
#	darkorange	lightslateblue		sandybrown
#	darkorchid	lightslategray		seagreen
#	darksalmon	lightslategrey		seashell
#	darkseagreen	lightsteelblue		sienna
#	darkslateblue	lightyellow		skyblue
#	darkslategray	limegreen		slateblue
#	darkslategrey	linen			slategray
#	darkturquoise	magenta 		slategrey
#	darkviolet	maroon			snow
#	deeppink	mediumaquamarine	springgreen
#	deepskyblue	mediumblue		steelblue
#	dimgray		mediumorchid		tan
#	dimgrey		mediumpurple		thistle
#	dodgerblue	mediumseagreen		tomato
#	firebrick	mediumslateblue		turquoise
#	floralwhite	mediumspringgreen	violet
#	forestgreen	mediumturquoise		violetred
#	gainsboro	mediumvioletred		wheat
#	ghostwhite	midnightblue		white
#	gold		mintcream		whitesmoke
#	goldenrod	mistyrose		yellow
#	gray		moccasin		yellowgreen