/usr/share/doc/fsa/FAQ.html is in fsa 1.15.9+dfsg-3.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 | <HTML>
<HEAD>
<link rel = "stylesheet" href = "pretty.css" type = "text/css">
<TITLE>FSA Frequently Asked Questions</TITLE>
<META NAME="version" content="0.9" />
<META NAME="keywords" CONTENT="FSA, fast statistical alignment, multiple sequence alignment, statistical alignment, bioinformatics, sequence annealing">
<META NAME="description" CONTENT="FSA Frequently Asked Questions">
<META NAME="ROBOTS" CONTENT="ALL">
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<!-- ImageReady Preload Script (faq.tif) -->
</HEAD>
<BODY BGCOLOR=#FFFFFF ONLOAD="preloadImages();">
<h1><a href="http://fsa.sourceforge.net/index.html">FSA</a> Frequently Asked Questions</h1>
<h2>Introduction</h2>
<ul>
<li><a href = "#intro">What is FSA?</a></li>
<li><a href = "#acronym">What does FSA stand for?</a></li>
<li><a href = "#install"> How do I download and install FSA?</a></li>
<li><a href = "#runfsa">How do I run FSA?</a></li>
<li><a href = "#server">Is there a webserver for FSA?</a></li>
<li><a href = "#help"> How do I get a help message to explain the options?</a></li>
</ul>
<h2>Alignments</h2>
<ul>
<li><a href = "#manyseqs"> How do I align many sequences?</a></li>
<li><a href = "#longseqs"> How do I align long sequences?</a></li>
<li><a href = "#genomes"> How do I align genomes?</a></li>
<li><a href = "#manygaps">Why are there so many gaps in my alignment?</a></li>
<li><a href = "#gapfactor">How do I control the sensitivity/specificity tradeoff of my alignment?</a></li>
</ul>
<h2>Visualization</h2>
<ul>
<li><a href="#gui">How do I visualize the alignments produced by FSA?</a></li>
<li><a href = "#reliable">How do I see what parts of my alignment are the most reliable?</a></li>
</ul>
<h2>Output formats and tools</h2>
<ul>
<li><a href = "#stockholm"> How do I parse Stockholm alignments?</a></li>
<li><a href = "#comparealign"> How do I compare alignments?</a></li>
</ul>
<h2>Troubleshooting</h2>
<ul>
<li> <a href = "#logging"> My alignment is taking a long time. How do I see what FSA is doing?</a>
<li> <a href = "#gui_memory">I'm getting memory errors when using the GUI.</a>
<li> <a href = "#malloc">I get out-of-memory (bad alloc) errors when I try to align long sequences.</a>
</ul>
<h2>More information</h2>
<ul>
<li><a href = "#contact"> How do I contact you?</a></li>
<li><a href = "#citation"> How do I cite FSA?</a></li>
<li><a href = "#license"> Under what license is FSA distributed?</a></li>
<li><a href = "#git"> How do I become an FSA developer?</a></li>
<li><a href = "#influences"> What other programs influenced the development of FSA?</a></li>
</ul>
<!-- Introduction -->
<hr>
<h2>Introduction</h2>
<h4><a name = "intro">What is FSA?</a></h4>
FSA is a probabilistic multiple sequence alignment algorithm which uses
a "distance-based" approach to aligning homologous protein, RNA or DNA
sequences. Much as distance-based phylogenetic reconstruction methods
like Neighbor-Joining build a phylogeny using only pairwise divergence
estimates, FSA builds a multiple alignment using only pairwise
estimations of homology. This is made possible by the sequence
annealing technique for constructing a multiple alignment from pairwise
comparisons, developed by Ariel Schwartz in
<a href="http://www.eecs.berkeley.edu/Pubs/TechRpts/2007/EECS-2007-39.html">
"Posterior Decoding Methods for Optimization and Control of Multiple
Alignments</a>."
<p>
FSA brings the high accuracies previously available only for small-scale analyses of proteins or RNAs
to large-scale problems such as aligning thousands of sequences or megabase-long sequences.
FSA introduces several novel methods for constructing better alignments:
<ul>
<li type=square>
FSA uses machine-learning techniques to estimate gap and
substitution parameters on the fly for each set of input sequences.
This "query-specific learning" alignment method makes FSA very robust: it
can produce superior alignments of sets of homologous sequences
which are subject to very different evolutionary constraints.
<li type=square>
FSA is capable of aligning hundreds or even thousands of sequences
using a randomized inference algorithm to reduce the computational
cost of multiple alignment. This randomized inference can be over
ten times faster than a direct approach with little loss of
accuracy.
<li type=square>
FSA can quickly align very long sequences using the "anchor
annealing" technique for resolving anchors and projecting them with
transitive anchoring. It then stitches together the alignment
between the anchors using the methods described above.
<li type=square>
The included GUI, MAD (Multiple Alignment Display), can display
the intermediate alignments produced by FSA, where each character
is colored according to the probability that it is correctly
aligned.
</ul>
<h4><a name = "acronym">What does FSA stand for?</a></h4>
Fast statistical alignment: We use machine-learning techniques to quickly
re-estimate parameters for each alignment problem.
<p>
Fast sequence annealing: We build a multiple alignment from pairwise comparisons
with the sequence annealing technique.
<p>
Functional statistical alignment: We implicitly use functional information when
constructing alignments.
<p>
<a href="http://www.fullspeedahead.com/">Full Speed Ahead</a><p>
<p>
...and more...
<h4><a name = "install">How do I download and install FSA?</a></h4>
FSA is hosted by <a href="http://sourceforge.net">SourceForge</a>.
You can download the latest version from the <a href="http://sourceforge.net/projects/fsa/">SourceForge project page</a>.
<p>
FSA is built and installed by running the following commands: <br>
<p><kbd>
tar xvzf fsa-X.X.X.tar.gz <br>
cd fsa-X.X.X <br>
./configure <br>
make <br>
make install <br>
</kbd>
<p>
(Substitute <var>fsa-X.X.X.tar.gz</var> with the name of the file
that you downloaded.)
<p>
The FSA executables can then be found in your system's standard
binary directory (e.g., /usr/local/bin). Alternatively, you may
just run FSA from the src/main subdirectory in which it is built
(which does not require running the <kbd>make install</kbd> step).
If you wish to install the FSA binaries in a location other than
your system's standard directories (which usually requires root
permissions), specify the top-level installation directory with
the <kbd>--prefix</kbd> option to configure. For example,
<p><kbd>./configure --prefix=$HOME</kbd>
<p>
specifies that binaries should be installed in <var>$HOME/bin</var>, libraries in
<var>$HOME/lib</var>, etc. <br>
<p>
If you wish to align long sequences, then you must download and install <a href="http://mummer.sourceforge.net/">MUMmer</a>,
which FSA calls to get candidate anchors between sequences.
When running <kbd>./configure</kbd>, either have the MUMmer executable in your path
or specify the executable with the <kbd>--with-mummer</kbd> option to <kbd>./configure</kbd>.
<p>
FSA can also call <a href="http://www.ebi.ac.uk/~guy/exonerate/">exonerate</a>
to obtain anchors. If you wish to use exonerate, then as with MUMmer,
when running <kbd>./configure</kbd>, you must either have the exonerate executable in your path
or specify the executable with the <kbd>--with-exonerate</kbd> option to <kbd>./configure</kbd>.
<p>
See the <var>README</var> file and <a href = "#longseqs">How do I align long sequences?</a>
for more information.
<p>
Please contact us if you have any build problems.
<h4><a name = "runfsa"> How do I run FSA?</a></h4>
FSA accepts FASTA-format input files and outputs multi-FASTA
alignments by default. The most basic usage is:
<p><kbd>fsa <mysequences.fa> >myalignedsequences.mfa</kbd>
<p>
or
<p><kbd>fsa --stockholm <mysequences.fa> >myalignedsequences.stk</kbd>
<p>
<h4><a name = "server"> Is there a webserver for FSA?</a></h4>
There is a webserver hosted <a href="http://orangutan.math.berkeley.edu/fsa/">here</a>
which you can submit alignment jobs to.
You will be emailed when the alignment is completed.
<h4><a name = "help"> How do I get a help message to explain the options?</a></h4>
Run
<p><kbd>fsa --help</kbd>
<h4><a name = "example_files"> Are there example sequence files and alignments?</a></h4>
Yes. Please see the <var>examples/</var> directory.
<!-- Alignments -->
<hr>
<h2>Alignments</h2>
<h4><a name = "manyseqs"> How do I align many sequences?</a></h4>
FSA can align thousands or tens of thousands of sequences.
Try running FSA with the <kbd>--fast</kbd> option. You can get finer-grained control with
the <kbd>--alignment-number <int></kbd> option, which controls the total number of pairwise
comparisons which FSA uses to build a multiple alignment. If you want to align
N sequences, then you can set <kbd>--alignment-number</kbd> to as low as (N - 1) or as
high as (N choose 2) == (N * (N - 1) / 2).
<h4><a name = "longseqs"> How do I align long sequences?</a></h4>
FSA can align long sequences (megabases or tens of megabases) with the
"anchor annealing" technique. It uses the program MUMmer to find maximal
unique matches between pairs of sequences to be aligned, resolves
inconsistencies with anchor annealing, and then pieces together the alignment
between anchored regions using its standard inference method.
<p>
FSA can also use the program exonerate to detect remote homology.
<p>
Please use the <kbd>--with-mummer</kbd> and <kbd>--with-exonerate</kbd> options to <kbd>./configure</kbd>
before compilation as explained in <a href="#runfsa">How do I run FSA?</a>.
<p>
You can read about the MUMmer and exonerate programs in:
<p>
S. Kurtz, A. Phillippy, A.L. Delcher, M. Smoot, M. Shumway, C. Antonescu, and S.L. Salzberg.
<a href="http://genomebiology.com/2004/5/2/R12">Versatile and open software for comparing large genomes</a>. Genome Biology. 2004, 5:R12.
<p>
G. S. Slater and E. Birney.
<a href="http://www.biomedcentral.com/1471-2105/6/31">Automated generation of heuristics for biological sequence comparison</a>. BMC Bioinformatics. 2005, 6:31.
<h4><a name = "genomes"> How do I align genomes?</a></h4>
If the genomes which you want to align have few rearrangements, then
you can run FSA directly on them. If they have rearrangments, then
you must first use a program such as Colin Dewey's <a href="http://www.biostat.wisc.edu/~cdewey/mercator/">Mercator</a>
to construct a homology map for the genomes and then run FSA on the
homologous segments. FSA can directly use the constraint information
produced by Mercator to inform its multiple alignment
(use the <kbd>--mercator</kbd> option to specify the Mercator constraint file).
<p>
You can read about Mercator in:
<p>
C. Dewey. <a href="http://www.eecs.berkeley.edu/Pubs/TechRpts/2006/EECS-2006-104.html">Whole-genome alignments and polytopes for comparative genomics</a>.
Ph.D. thesis, University of California, Berkeley. 2006.
<h4><a name = "logging"> My alignment is taking a long time. How do I see what FSA is doing?</a></h4>
FSA has an extensive logging system. Try running with the <kbd>--log 7</kbd> option
to see progress of the DP algorithm, and <kbd>--log 6</kbd> to see progress of anchoring
(when aligning long sequences).
Log levels from 0 to 10 are permitted,
where lower numbers are more verbose.
<h4><a name = "manygaps">Why are there so many gaps in my alignment?</a></h4>
Most alignment programs attempt to maximize sensitivity, even at
the expense of specificity, leading to over-alignment (alignment of
non-homologous sequence). FSA, in contrast, maximizes the expected
accuracy of the alignment with a measure which rewards sensitivity
but penalizes over-alignment. If FSA cannot reliably detect homology,
then it will leave characters unaligned (gapped).
<h4><a name = "gapfactor">How do I control the sensitivity/specificity tradeoff of my alignment?</a></h4>
By default FSA stops aligning characters when the probability that
a character is aligned is equal to the probability that it is gapped.
Use the <kbd>--maxsn</kbd> option for maximum sensitivity.
<p>
You can get finer-grained control with the <kbd>--gapfactor <int></kbd> option.
By default FSA runs at <kbd>--gapfactor 1</kbd>.
Use <kbd>--gapfactor 0</kbd> for highest sensitivity (this is equivalent to <kbd>--maxsn</kbd>)
and gap factors > 1 for higher specificity.
<!-- Visualization -->
<hr>
<h2>Visualization</h2>
<h4><a name = "gui"> How do I visualize the alignments produced by FSA?</a></h4>
Run FSA with the command-line option <kbd>--gui</kbd>. If the input alignment file
is <var>myseqs.fasta</var>, then FSA will write the files <var>myseqs.fasta.gui</var> and
'myseqs.fasta.probs'. Invoke the MAD (Multiple Alignment Display) GUI
as
<p><kbd>java -jar display/mad.jar myseqs.fasta</kbd>
<p>
Please be patient if it takes a while to load.
<p>
Characters in the multiple alignment are colored according to the
probability that they are correctly aligned.
<p>
You can see what a typical FSA alignment looks like by running
on one of the provided example alignments, such as
<p><kbd>java -jar display/mad.jar examples/tRNA.aln1.fasta</kbd>
<p>
You can also use the GUI to compare a FSA alignment with another alignment,
such as one which you have edited by hand. Invoke the GUI as
<p><kbd>java -jar display/mad.jar examples/tRNA.aln1.fasta myalignment.fa</kbd>
The alternate alignment <kbd>myalignment.mfa</kbd> must be in multi-FASTA format.
<h4><a name = "reliable">How do I see what parts of my alignment are the most reliable?</a></h4>
The accuracy estimates produced by FSA and the GUI are useful for
downstream analyses, for example allowing biologists to restrict
their analyses to the most reliable portions of the alignment
or edit unreliable parts of the alignment by hand.
The accuracy measures also allow you to visualize how FSA works.
If you switch to the "Specificity" coloring and watch the animation from the beginning,
you will see FSA first aligns characters whose homology it is most sure of (red),
and only later aligns characters of unclear homology (blue).
Similarly, you can visualize the sensitivity/specificity tradeoff: Near the beginning
of the alignment the specificity is very high, but the sensitivity is low.
The specificity decreases and the sensitivity increases as the alignment progresses.
The GUI displays five different accuracy measures for each position in the multiple alignment.
These are:
<ul>
<li type=square><u>Accuracy</u>: What characters or gaps of the multiple alignment are the most accurate?
<blockquote>
Accuracy is the per-character estimated Alignment Metric Accuracy,
which measures the fidelity of both aligned characters and unaligned characters (gaps).
It can be thought of as a single measure encompassing the sensitivity/specificity tradeoff.
</blockquote>
<li type=square><u>Sensitivity</u>: What characters are aligned with the greatest sensitivity?
<blockquote>
Sensitivity is the estimated number of correctly-aligned character pairs divided by the
true number of aligned character pairs.
<br>
Sensitivity is defined as the expectation of (True positives) / (True positives + False negatives).
This definition is equivalent to <a href="http://en.wikipedia.org/wiki/Precision_and_recall">recall</a> as used in classification problems.
</blockquote>
<li type=square><u>Specificity</u>: What characters are aligned with the greatest specificity?
<blockquote>
Specificity is the estimated fraction of character pairs which are aligned correctly.
<br>
Specificity is defined as the expectation of (True positives) / (True positives + False positives).
This definition is equivalent to <a href="http://en.wikipedia.org/wiki/Precision_and_recall">precision</a> as used in classification problems.
It is also frequently called <a href="http://en.wikipedia.org/wiki/Positive_predictive_value">Positive Predictive Value</a> in the literature.
</blockquote>
<li type=square><u>Certainty</u>: Was there a better place to align this character?
<blockquote>
Certainty measures whether a character or gap is aligned correctly.
</blockquote>
<li type=square><u>Consistency</u>: What parts of the multiple alignment are optimal on a pairwise level?
<blockquote>
Consistency measures the extent to which the posterior probabilities from pairwise comparisons
are optimized by the multiple alignment. If a multiple alignment is perfectly consistent,
then each pairwise alignment implied by the multiple alignment corresponds perfectly to the pairwise alignment
which you would obtain by aligning only those two sequences.
</blockquote>
</ul>
<p>
Please see the manuscript for precise definitions of these reliability measures.
<p>
Notice that the accuracy scores tend to
decrease near gaps, reflecting the difficulty of precisely
resolving gap boundaries.
<!-- Output formats and tools -->
<hr>
<h2>Output formats and tools</h2>
<h4><a name = "stockholm"> How do I parse Stockholm alignments?</a></h4>
Use the <kbd>--stockholm</kbd> output option to tell FSA
to produce a Stockholm-format alignment. The alignment is marked up with
a per-column accuracy annotation which is identical to the values
reported by the <a href="#gui">GUI</a>.
FSA includes tools for working with Stockholm alignments, such as
<kbd>prettify.pl</kbd> for making Stockholm-format alignments human-
readable, in the <var>perl/</var> directory.
<h4><a name = "comparealign"> How do I compare alignments?</a></h4>
The included script <kbd>cmpalign.pl</kbd> will compare two alignments
and report accuracies measures including Accuracy (AMA), Sensitivity and Specificity.
It can parse Stockholm, multi-FASTA, MSF and CLUSTAL format alignments.
<!-- Troubleshooting -->
<hr>
<h2>Troubleshooting</h2>
<h4><a name = "gui_memory"> I'm getting memory errors when using the GUI.</a></h4>
If you're getting errors which say something like
<p><kbd>Exception in thread ... java.lang.OutOfMemoryError: Java heap space</kbd>
<p>
then try increasing the memory allowed with with <kbd>-Xmx</kbd> option, ie
<p><kbd>java -Xmx256m jar display/mad.jar examples/tRNA.aln1.fasta</kbd>
<p>
You can increase it up to the maximum allowed by your machine.
<h4><a name = "malloc"> I get out-of-memory (bad alloc) errors when I try to align long sequences.</a></h4>
This occurs when FSA is unable to find good anchors between your sequences
to restrict the complexity of inference. This can occur if you use the option
<kbd>--noanchored</kbd> to prevent anchoring, if your sequences are very diverged,
or if they have many simple repeats. Use the <kbd>--maxram</kbd> option to
prevent FSA from attempting to perform exhaustive inference when it can't
find good anchors. It will leave sequence for which it can't find
sufficiently-good anchors unaligned.
<!-- More information -->
<hr>
<h2>More information</h2>
<h4><a name = "contact"> How do I contact you?</a></h4>
You can reach the FSA team at <a href="mailto:fsa@math.berkeley.edu">fsa@math.berkeley.edu</a> with any questions, feedback, etc.
<h4><a name = "citation"> How do I cite FSA?</a></h4>
Bradley RK, Roberts A, Smoot M, Juvekar S, Do J, Dewey C, Holmes I, Pachter L (2009) <a href="http://www.ploscompbiol.org/article/info%3Adoi%2F10.1371%2Fjournal.pcbi.1000392">Fast Statistical Alignment</a>. PLoS Computational Biology. 5:e1000392.
<p>The FSA manuscript can also be found in the <var>doc/</var>
directory of the FSA source code distribution.
<h4><a name = "license">Under what license is FSA distributed?</a></h4>
FSA is licensed under version 3 of the <a href="http://www.gnu.org/licenses/gpl.html">GNU General Public License</a>.
Please see the files <var>LICENSE</var> and <var>COPYING</var> for further information.
<h4><a name = "git"> How do I become an FSA developer?</a></h4>
FSA is designed to be modular and there are many aspects of the program that can be improved;
we welcome your help! The code is under <a href="http://git.or.cz/">Git</a> version control.
Please contact us at <a href="mailto:fsa@math.berkeley.edu">fsa@math.berkeley.edu</a> for information.
The source code is set up for use with <a href="http://www.doxygen.org/">Doxygen</a>,
a system for automated building of documentation.
<h4><a name = "influences"> What other programs influenced the development of FSA?</a></h4>
Source code in <var>seq/</var> and <var>util/</var> is from Ian Holmes's DART library [1],
which is used for input and output routines.
<p>
FSA's DP code was generated by HMMoC by Gerton Lunter [2]. The
aligner example distributed with HMMoC, which implements a
learning procedure for gap parameters, was an inspiration for FSA's
learning strategies. FSA's banding code is taken directly from the
aligner example.
<p>
The sequence annealing technique for constructing a multiple
alignment from pairwise comparisons was developed by Ariel Schwartz.
The implementation of sequence annealing in FSA is a modified version
of the original implementation in AMAP by Ariel Schwartz and Lior Pachter [3,4].
<p>
The anchor annealing approach used in FSA is modeled after the recursive
anchoring strategy used in MAVID by Nicolas Bray and Lior Pachter [5].
<p>
The MAD GUI interface to FSA was written by Adam Roberts based on a preliminary
version developed by Michael Smoot.
<p>
Please see:
<p>[1] I. Holmes and R. Durbin. <a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=9773345">Dynamic Programming Alignment Accuracy</a>. Journal of Computational Biology. 1998, 5 (3):493-504.
<p>[2] G.A. Lunter. <a href="http://bioinformatics.oxfordjournals.org/cgi/content/abstract/23/18/2485">HMMoC - a Compiler for Hidden Markov Models</a>. Bioinformatics. 2007, 23 (18):2485-2487.
<p>[3] A.S. Schwartz. <a href="http://www.eecs.berkeley.edu/Pubs/TechRpts/2007/EECS-2007-39.html">Posterior Decoding Methods for Optimization and Control of Multiple Alignments</a>. Ph.D. Thesis, UC Berkeley. 2007.
<p>[4] A.S. Schwartz and L. Pachter. <a href="http://bioinformatics.oxfordjournals.org/cgi/content/abstract/23/2/e24?etoc">Multiple Alignment by Sequence Annealing</a>. Bioinformatics. 2007, 23 (2):e24-e29.
<p>[5] N. Bray and L. Pachter. <a href="http://www.genome.org/cgi/content/abstract/14/4/693">MAVID: Constrained Ancestral Alignment of Multiple Sequences</a>. Genome Research. 2004, 14:693-699.
<hr>
<a href="http://sourceforge.net/projects/fsa">Get FSA at SourceForge.net. Fast, secure and Free Open Source software downloads"</a>
</BODY>
</HTML>
|