/usr/lib/rtax/scripts/rtaxSearch.pl is in rtax 0.984-3.
This file is owned by root:root, with mode 0o755.
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 | #!/usr/bin/env perl
# RTAX: Rapid and accurate taxonomic classification of short paired-end
# sequence reads from the 16S ribosomal RNA gene.
#
# David A. W. Soergel 1*, Rob Knight 2, and Steven E. Brenner 1
#
# 1 Department of Plant and Microbial Biology,
# University of California, Berkeley
# 2 Howard Hughes Medical Institute and Department of Chemistry
# and Biochemistry, University of Colorado at Boulder
# * Corresponding author: soergel@cs.umass.edu
#
# http://www.davidsoergel.com/rtax
#
# Version 0.984 (August 7, 2013)
#
# For usage instructions: just run the script with no arguments
#
#
# Copyright (c) 2011 Regents of the University of California
#
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
# * Redistributions of source code must retain the above copyright notice,
# this list of conditions and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution.
# * Neither the name of the University of California, Berkeley nor the
# names of its contributors may be used to endorse or promote products
# derived from this software without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
# PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
# LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
# NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
# SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
use strict;
use warnings;
use Getopt::Long;
use File::Temp qw/ tempdir /;
use FindBin;
use lib "$FindBin::Bin";
use FastaIndex;
### command-line arguments get stored in globals
use vars qw (
$usearch
$slop
$minMaxAccepts
$maxMaxAccepts
$maxPercentDifferenceThreshold
$databaseFile
$readAFileAll
$readBFileAll
$revCompReadA
$revCompReadB
$idRegex
$idList
$singleOK
$singleOKgeneric
);
# just store these as globals for now-- they're basically like command-line args
my $indexA;
my $indexB;
sub init {
$singleOKgeneric = 1; # default value; GetOptions may override
Getopt::Long::Configure("bundling");
GetOptions(
"usearch=s" => \$usearch,
"slop=s" => \$slop,
"minMaxAccepts=s" => \$minMaxAccepts,
"maxMaxAccepts=s" => \$maxMaxAccepts,
"maxPercentDifferenceThreshold=s" => \$maxPercentDifferenceThreshold,
"databaseFile=s" => \$databaseFile,
"idRegex=s" => \$idRegex,
"queryA=s" => \$readAFileAll,
"queryB=s" => \$readBFileAll,
"revcompA" => \$revCompReadA,
"revcompB" => \$revCompReadB,
"idList=s" => \$idList,
"singleOK" => \$singleOK,
"singleOKgeneric!" => \$singleOKgeneric
);
if ( !defined $databaseFile
|| !defined $readAFileAll )
{
print STDERR "Missing required argument!\n\n";
usage();
}
if ( !defined $usearch ) {
$usearch = `whereis usearch`;
chomp $usearch;
if ( !defined $usearch || $usearch eq "" ) {
print STDERR "Could not find usearch.\n\n";
usage();
}
}
if ( !defined $slop ) { $slop = 0.005; }
if ( !defined $minMaxAccepts ) { $minMaxAccepts = 1000; }
if ( !defined $maxMaxAccepts ) { $maxMaxAccepts = 16000; }
if ( !defined $maxPercentDifferenceThreshold ) {
$maxPercentDifferenceThreshold = 0.2;
}
if ( !defined $idRegex || $idRegex eq "" ) { $idRegex = "(\\S+)"; }
print STDERR "Header Regex: $idRegex\n";
# these are redundant between multiple runs, oh well
# but the DBM indexes should persist, in the same directory as the original files
$indexA = FastaIndex->new(); # '-filename' => "A.idx", '-write_flag' => 1 );
if ($revCompReadA) {
$readAFileAll = revcompFile($readAFileAll);
}
$indexA->make_index( $readAFileAll, $idRegex, $readAFileAll );
$indexB = FastaIndex->new();
if ( defined $readBFileAll ) { # '-filename' => "B.idx", '-write_flag' => 1 );
if ($revCompReadB) {
$readBFileAll = revcompFile($readBFileAll);
}
$indexB->make_index( $readBFileAll, $idRegex, $readBFileAll );
}
}
sub usage {
print STDERR << "EOF";
RTAX: Rapid and accurate taxonomic classification of short paired-end
sequence reads from the 16S ribosomal RNA gene.
David A. W. Soergel (1*), Rob Knight (2), and Steven E. Brenner (1)
1 Department of Plant and Microbial Biology, University of California, Berkeley
2 Howard Hughes Medical Institute and Department of Chemistry and Biochemistry,
University of Colorado at Boulder
* Corresponding author: soergel\@berkeley.edu
http://www.davidsoergel.com/rtax
Version 0.984 (August 7, 2013)
OPTIONS:
--usearch <path> location of usearch program
(defaults to result of "whereis usearch")
--slop <number> %id difference from maximum to accept
(default 0.005, i.e. accept hits to a 200nt query that
has 1 more mismatch than the best match, or 2 extra
mismatches out of 400nt.)
--minMaxAccepts <number>
The initial maxaccepts value to pass to USEARCH,
controlling the maximum number of hits that will be
returned in the first iteration. If this number of hits
is reached, the search is repeated with a doubled
maxaccepts value. Thus, a smaller value causes the
first iteration to run faster, but increases the
probability of needing more iterations.
(Default 1000)
--maxMaxAccepts <number>
The highest escalated maxaccepts value to allow. The
maxaccepts value is doubled on every iteration, so this
value is effectively rounded down to a power of two
factor of minMaxAccepts. USEARCH runs with large
maxaccepts values are very slow, so this controls when
RTAX gives up on a query because there are too many
hits (i.e., a query that is too short and/or too highly
conserved to be informative). (Default 16000)
--maxPercentDifferenceThreshold <number>
The largest percent difference to allow between a query
(considering the read pair jointly) and a database hit.
(Default 0.2, i.e. reference sequences of less than 80%
identity with the query sequence are never matched)
--databaseFile <file>
A FASTA file containing a reference database.
--queryA <file> A FASTA file containing one set of query reads.
--queryB <file> A FASTA file containing the other set of query reads
(if any). Must be provided in the forward sense
unless --revcompB is used.
--revcompB Reverse-complement read B. Required if the queryB file
is provided in the reverse sense, as is typical with
paired-end experiments.
--singleOK Classify a sequence based on only one read when the
other read is missing. Required when queryB is absent.
Default: false, so sequences present in only one of the
two input files are dropped. When enabled, paired-end
sequences are classified using both reads as usual, but
any remaining single-ended reads are also classified.
--singleOKgeneric Classify a sequence based on only one read when the
other read is present but uninformative. This occurs
when one read is so generic (i.e., short, or highly
conserved) that more than maxMaxAccepts hits are
found. Default: true; use --nosingleOKgeneric to
disable.
--idRegex <regex> A regular expression for extracting IDs from fasta
headers. The first capture group (aka \$1) will be
used as the sequence ID; these should be unique.
This is useful when the ID needed to match mate
pairs is embedded in the header in some way.
"^>" is automatically prepended to the regex
provided here. Take care that double escaping may
be required. Default: "(\\S+)" (the first field)
--idList <file> A file containing a list of IDs to process (one per
line). By default all IDs found in the query files
are used.
Note that the two query files must provide mate-paired reads with exactly
matching identifiers (though they need not be in the same order). Any ids
present in one file but not the other are classified in single-ended mode.
Alternate naming schemes for the paired reads (e.g., "SOMEID.a" paired with
"SOMEID.b", and so forth) are handled via the --idRegex option.
Various indexes and derived FASTA files will be created in a temporary
directory and are cleaned up afterwards.
The output is tab-delimited text, one line per query pair, in the form
<query ID> <%id> <list of reference IDs>
where the second column gives the %id between the query and the best match,
and the reference IDs provided are all those matches within "slop" of this
best value.
This output is suitable for piping directly into rtaxVote for taxonomy
assignment.
EOF
exit;
}
sub main {
init();
my ( $pairedReadAFile, $pairedReadBFile, $pairedBothCount, $singleReadAFile, $singleReadACount, $singleReadBFile, $singleReadBCount, $allSingleIDs ) =
partitionReadFiles();
processPairs( $pairedReadAFile, $pairedReadBFile, $pairedBothCount );
if ($singleOK) {
processSingle( $singleReadAFile, $indexA, $singleReadACount );
processSingle( $singleReadBFile, $indexB, $singleReadBCount );
}
else {
for my $queryLabel (@$allSingleIDs) {
print "$queryLabel\t\tNOMATEPAIR\n";
}
}
}
sub partitionReadFiles {
my @idList = ();
if ($idList) {
open( IDLIST, "$idList" ) or die "Could not open $idList";
while (<IDLIST>) { chomp; push @idList, $_; }
close IDLIST;
}
my @idsA = keys %{ $indexA->startpos() };
my @idsB =
defined $indexB->startpos() ? keys %{ $indexB->startpos() } : (); # in the single-read case, there should still be an empty index
my @bothAandB = ();
my @aOnly = ();
my @bOnly = ();
# encode which of the input files contain which IDs in three bits
my %count = ();
foreach my $element (@idsA) {
if ( ( $element =~ /[\t ]/ ) > 0 ) # $indexA->db() should return parsed IDs with no whitespace
{
die "Invalid FASTA id: $element\n";
}
$count{$element} += 1;
}
foreach my $element (@idsB) {
if ( ( $element =~ /[\t ]/ ) > 0 ) # $indexA->db() should return parsed IDs with no whitespace
{
die "Invalid FASTA id: $element\n";
}
$count{$element} += 2;
}
foreach my $element (@idList) {
if ( ( $element =~ /[\t ]/ ) > 0 ) { die "Invalid FASTA id: $element\n"; }
$count{$element} += 4;
}
if ($idList) {
foreach my $element ( keys %count ) {
$count{$element} -= 4;
}
}
foreach my $element ( keys %count ) {
if ( !( $element =~ /^__/ ) ) {
if ( $count{$element} == 1 ) { push @aOnly, $element }
elsif ( $count{$element} == 2 ) { push @bOnly, $element }
elsif ( $count{$element} == 3 ) { push @bothAandB, $element }
else {
# no problem; these are sequences not in the idList
}
}
}
print STDERR "file a only = " . scalar(@aOnly) . " sequences\n";
print STDERR "file b only = " . scalar(@bOnly) . " sequences\n";
print STDERR "both = " . scalar(@bothAandB) . " sequences\n";
my $pairedReadAFile = extractFasta( $indexA, \@bothAandB );
my $pairedReadBFile = extractFasta( $indexB, \@bothAandB );
my $singleReadAFile = extractFasta( $indexA, \@aOnly );
my $singleReadBFile = extractFasta( $indexB, \@bOnly );
my @allSingleIDs = (@aOnly, @bOnly);
return ( $pairedReadAFile, $pairedReadBFile, scalar(@bothAandB), $singleReadAFile, scalar(@aOnly), $singleReadBFile, scalar(@bOnly), \@allSingleIDs );
}
sub processPairs {
my ( $pairedReadAFile, $pairedReadBFile, $pairedBothCount ) = @_;
if ( $pairedBothCount == 0 ) { return; }
my $nohitQueryIds = [];
push @$nohitQueryIds, "ALL";
my $tooManyHitQueryIds = [];
my $percentDifferenceThreshold = 0.005; # this gets doubled to 1% before being used the first time
while ( @$nohitQueryIds > 0 ) {
# double the allowed %diff on every round
$percentDifferenceThreshold *= 2;
if ( $percentDifferenceThreshold > $maxPercentDifferenceThreshold ) {
last;
}
if ( $nohitQueryIds->[0] ne "ALL" ) {
# prepare input files with the remaining sequences
$pairedReadAFile = extractFasta( $indexA, $nohitQueryIds );
$pairedReadBFile = extractFasta( $indexB, $nohitQueryIds );
}
# within doPairSearch we escalate maxAccepts, so if a queryLabel is still marked tooManyHits at this point,
# that means that there are more than maxMaxAccepts hits for this threshold--
# so there's really no point in testing this query again at an even lower threshold
my $tooManyHitQueryIdsThisRound;
my $numRemaining = ( $nohitQueryIds->[0] eq "ALL" ) ? $pairedBothCount : scalar(@$nohitQueryIds);
print STDERR
"doPairSearch $pairedReadAFile, $pairedReadBFile: $numRemaining query sequences remaining\n searching with pair %id "
. $percentDifferenceThreshold
. " and maxAccepts "
. $minMaxAccepts . "\n";
( $nohitQueryIds, $tooManyHitQueryIdsThisRound ) =
doPairSearch( $pairedReadAFile, $pairedReadBFile, $percentDifferenceThreshold, $minMaxAccepts );
print STDERR "MAIN: Finished round at threshold $percentDifferenceThreshold; "
. scalar(@$nohitQueryIds)
. " NOHIT, "
. scalar(@$tooManyHitQueryIdsThisRound)
. " TOOMANYHIT.\n";
push @$tooManyHitQueryIds, @$tooManyHitQueryIdsThisRound;
}
print STDERR "MAIN: " . scalar(@$nohitQueryIds) . " query sequences remaining with NOHIT\n";
for my $queryLabel (@$nohitQueryIds) {
print "$queryLabel\t\tNOHIT\n";
}
print STDERR "MAIN: " . scalar(@$tooManyHitQueryIds) . " query sequences remaining with TOOMANYHITS\n";
for my $queryLabel (@$tooManyHitQueryIds) {
print "$queryLabel\t\tTOOMANYHITS\n";
}
}
sub processSingle {
my ( $singleReadFile, $singleIndex, $singleReadCount ) = @_;
if ( !defined $singleReadFile || $singleReadFile eq "" || $singleReadCount == 0 ) { return; }
my $nohitQueryIds = [];
push @$nohitQueryIds, "ALL";
my $tooManyHitQueryIds = [];
my $singlePercentDifferenceThreshold = 0.005; # this gets doubled to 1% before being used the first time
while ( @$nohitQueryIds > 0 ) {
# double the allowed %diff on every round
$singlePercentDifferenceThreshold *= 2;
if ( $singlePercentDifferenceThreshold > $maxPercentDifferenceThreshold ) {
last;
}
if ( $nohitQueryIds->[0] ne "ALL" ) {
# prepare input files with the remaining sequences
$singleReadFile = extractFasta( $singleIndex, $nohitQueryIds );
}
my $numRemaining = ( $nohitQueryIds->[0] eq "ALL" ) ? $singleReadCount : scalar(@$nohitQueryIds);
# within doSearch we escalate maxAccepts, so if a queryLabel is still marked tooManyHits at this point,
# that means that there are more than maxMaxAccepts hits for this threshold--
# so there's really no point in testing this query again at an even lower threshold
my $tooManyHitQueryIdsThisRound;
print STDERR "doSingleSearch $singleReadFile: $numRemaining query sequences remaining\n searching with %id "
. $singlePercentDifferenceThreshold
. " and maxAccepts "
. $minMaxAccepts . "\n";
( $nohitQueryIds, $tooManyHitQueryIdsThisRound ) =
doSingleSearch( $singleReadFile, $singleIndex, $singlePercentDifferenceThreshold, $minMaxAccepts );
print STDERR "Finished round at threshold $singlePercentDifferenceThreshold; "
. scalar(@$nohitQueryIds)
. " NOHIT, "
. scalar(@$tooManyHitQueryIdsThisRound)
. " TOOMANYHIT.\n";
push @$tooManyHitQueryIds, @$tooManyHitQueryIdsThisRound;
}
print STDERR scalar(@$nohitQueryIds) . " query sequences remaining with NOHIT\n";
for my $queryLabel (@$nohitQueryIds) {
print "$queryLabel\t\tNOHIT\n";
}
print STDERR scalar(@$tooManyHitQueryIds) . " query sequences remaining with TOOMANYHITS\n";
for my $queryLabel (@$tooManyHitQueryIds) {
print "$queryLabel\t\tTOOMANYHITS\n";
}
}
sub doSingleSearch {
my $singleReadFile = shift;
my $singleIndex = shift;
my $singlePercentDifferenceThreshold = shift;
my $maxAccepts = shift;
my $singlePercentIdThreshold = 1. - $singlePercentDifferenceThreshold;
my $nohitQueryIds = [];
my $tooManyHitQueryIds = [];
# open the USEARCH streams
# print STDERR "$usearch --quiet --global --iddef 2 --query $singleReadFile --db $databaseFile --uc /dev/stdout --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject\n";
# open( UCA,
# "$usearch --quiet --global --iddef 2 --query $singleReadFile --db $databaseFile --uc /dev/stdout --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject |"
# ) || die "can't fork usearch: $!";
my $dir = tempdir(CLEANUP => 1);
if ( system( 'mknod', "$dir/a", 'p' ) && system( 'mkfifo', "$dir/a" ) ) { die "mk{nod,fifo} $dir/a failed"; }
if ( !fork() ) {
# see paired end case for explanation
open( UCAW, ">$dir/a" ) || die("Couldn't write named pipe: $dir/a");
print UCAW "# pipe open!\n";
close UCAW;
my $cmd =
"$usearch --quiet --global --iddef 2 --query $singleReadFile --db $databaseFile --uc $dir/a --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject";
print STDERR $cmd . "\n";
exec $cmd || die "can't fork usearch: $!";
}
open( UCA, "$dir/a" ) || die("Couldn't read named pipe from usearch: $dir/a");
#print STDERR "Reading named pipe from usearch: $dir/a\n";
# Load the first non-comment line from each stream
my $nextLineA;
my $pipeARetryCount = 0;
while ( !defined $nextLineA ) {
# keep trying to read from the pipe even if the writer disconnects
while (<UCA>) {
if (/^\s*#/) { next; }
if (/^\s*$/) { next; }
$nextLineA = $_;
last;
}
#print STDERR "Waiting for named pipe $dir/a\n";
sleep 1;
$pipeARetryCount++;
if ( $pipeARetryCount > 10 ) { die "Named pipe communication with usearch failed: $dir/a\n"; }
}
# read synchronized blocks from each stream
while (1) {
my ( $queryLabelA, $idsA );
#print STDERR "reading next block...\n";
# idsA is a reference to a hash from targetIds to %ids
( $queryLabelA, $idsA, $nextLineA ) = collectIds( *UCA, $nextLineA );
if ( ( scalar keys %$idsA ) >= $maxAccepts ) {
push @$tooManyHitQueryIds, $queryLabelA;
}
elsif ( !reconcileSingleHitsAndPrint( $queryLabelA, $idsA, $singlePercentIdThreshold ) ) {
push @$nohitQueryIds, $queryLabelA;
}
if ( !$nextLineA ) {
#print STDERR "End of stream, close\n";
last;
}
}
close(UCA) || die "can't close usearch: $!";
#print STDERR "Closed usearch stream.\n";
# for the TOOMANYHITS cases, escalate maxAccepts and try again
# Note this recurses, so no need to iterate here
if ( scalar(@$tooManyHitQueryIds) && ( $maxAccepts * 2 <= $maxMaxAccepts ) ) {
my $nohitQueryIdsB;
print STDERR "Escalating maxAccepts to " . ( $maxAccepts * 2 ) . " for " . scalar(@$tooManyHitQueryIds) . " sequences.\n";
# prepare input files with the remaining sequences
$singleReadFile = extractFasta( $singleIndex, $tooManyHitQueryIds );
( $nohitQueryIdsB, $tooManyHitQueryIds ) =
doSingleSearch( $singleReadFile, $singleIndex, $singlePercentDifferenceThreshold, $maxAccepts * 2 );
if (@$nohitQueryIdsB) {
die "A TOOMANYHITS case can't turn into a NOHITS case";
}
}
print STDERR
"doSingleSearch $singleReadFile: Finished at pair threshold $singlePercentDifferenceThreshold and maxAccepts $maxAccepts\n";
print STDERR " NOHITS: " . scalar(@$nohitQueryIds) . "\n";
print STDERR " TOOMANYHITS: " . scalar(@$tooManyHitQueryIds) . "\n";
# print STDERR " NOHITS: " . ( join ", ", @$nohitQueryIds ) . "\n";
# print STDERR " TOOMANYHITS: " . ( join ", ", @$tooManyHitQueryIds ) . "\n";
# any tooManyHitQueryIds that remain had more than maxMaxAccepts hits
return ( $nohitQueryIds, $tooManyHitQueryIds );
}
sub reconcileSingleHitsAndPrint {
my ( $queryLabel, $idsA, $singlePercentIdThreshold ) = @_;
my $idsByIdentity = {};
for my $targetLabel ( keys %$idsA ) {
my $percentIdA = $idsA->{$targetLabel};
if ( !defined $idsByIdentity->{$percentIdA} ) {
$idsByIdentity->{$percentIdA} = [];
}
push @{ $idsByIdentity->{$percentIdA} }, $targetLabel;
}
if ( !%$idsByIdentity ) {
#print STDERR "$queryLabel -> no reconciled hits at $singlePercentIdThreshold%\n";
# this is the NOHIT case, but we'll back off and try again
return 0;
}
else {
#print STDERR "$queryLabel -> printing reconciled hits at $singlePercentIdThreshold%\n";
printLine( $queryLabel, $idsByIdentity );
return 1;
}
}
sub doPairSearch {
my $readAFile = shift;
my $readBFile = shift;
my $pairPercentDifferenceThreshold = shift;
my $maxAccepts = shift;
my $pairPercentIdThreshold = 1. - $pairPercentDifferenceThreshold;
# because we're going to average the two %ids, we have to search each read with twice the %diff for the pair
my $singlePercentDifferenceThreshold = $pairPercentDifferenceThreshold * 2;
my $singlePercentIdThreshold = 1. - $singlePercentDifferenceThreshold;
my $nohitQueryIds = [];
my $tooManyHitQueryIds = [];
# open the USEARCH streams
# print STDERR
#"$usearch --quiet --global --iddef 2 --query $readAFile --db $databaseFile --uc /dev/stdout --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject\n";
# open( UCA,
#"$usearch --quiet --global --iddef 2 --query $readAFile --db $databaseFile --uc /dev/stdout --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject |"
# ) || die "can't fork usearch: $!";
# print STDERR
#"$usearch --quiet --global --iddef 2 --query $readBFile --db $databaseFile --uc /dev/stdout --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject\n";
# open( UCB,
#"$usearch --quiet --global --iddef 2 --query $readBFile --db $databaseFile --uc /dev/stdout --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject |"
# ) || die "can't fork usearch: $!";
my $dir = tempdir(CLEANUP => 1);
if ( system( 'mknod', "$dir/a", 'p' ) && system( 'mkfifo', "$dir/a" ) ) { die "mk{nod,fifo} $dir/a failed"; }
if ( system( 'mknod', "$dir/b", 'p' ) && system( 'mkfifo', "$dir/b" ) ) { die "mk{nod,fifo} $dir/b failed"; }
# try to avoid mysterious intermittent condition where opening a named pipe blocks forever
#while ( !-p "$dir/a" ) {
# print STDERR "Waiting for named pipe $dir/a\n";
# sleep 1;
#}
#while ( !-p "$dir/b" ) {
# print STDERR "Waiting for named pipe $dir/b\n";
# sleep 1;
#}
if ( !fork() ) {
# there is a mysterious intermittent condition where opening a named pipe blocks forever.
# I think what is happening may be:
# if the reader side of the named pipe is not already connected when usearch tries to write to it, usearch gets confused and never writes anything
# thus when the reader side does connect, it blocks.
# one hack is just to sleep here for a while in hopes that the reader thread gets all hooked up
# sleep 5;
# let's try writing to it, so we block here until we know it works, and then continue on to usearch
open( UCAW, ">$dir/a" ) || die("Couldn't write named pipe: $dir/a");
print UCAW "# pipe open!\n";
close UCAW;
my $cmd =
"$usearch --quiet --global --iddef 2 --query $readAFile --db $databaseFile --uc $dir/a --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject";
print STDERR $cmd . "\n";
exec $cmd || die("can't fork usearch: $!");
}
if ( !fork() ) {
open( UCBW, ">$dir/b" ) || die("Couldn't write named pipe: $dir/b");
print UCBW "# pipe open!\n";
close UCBW;
my $cmd =
"$usearch --quiet --global --iddef 2 --query $readBFile --db $databaseFile --uc $dir/b --id $singlePercentIdThreshold --maxaccepts $maxAccepts --maxrejects 128 --nowordcountreject";
print STDERR $cmd . "\n";
exec $cmd || die("can't fork usearch: $!");
}
open( UCA, "$dir/a" ) || die("Couldn't read named pipe from usearch: $dir/a");
#print STDERR "Reading named pipe from usearch: $dir/a\n";
open( UCB, "$dir/b" ) || die("Couldn't read named pipe from usearch: $dir/b");
#print STDERR "Reading named pipe from usearch: $dir/b\n";
# Load the first non-comment line from each stream
my $nextLineA;
my $pipeARetryCount = 0;
while ( !defined $nextLineA ) {
# keep trying to read from the pipe even if the writer disconnects
while (<UCA>) {
if (/^\s*#/) { next; }
if (/^\s*$/) { next; }
$nextLineA = $_;
last;
}
#print STDERR "Waiting for named pipe $dir/a\n";
sleep 1;
$pipeARetryCount++;
if ( $pipeARetryCount > 10 ) { die("Named pipe communication with usearch failed: $dir/a\n"); }
}
my $nextLineB;
my $pipeBRetryCount = 0;
while ( !defined $nextLineB ) {
# keep trying to read from the pipe even if the writer disconnects
while (<UCB>) {
if (/^\s*#/) { next; }
if (/^\s*$/) { next; }
$nextLineB = $_;
last;
}
#print STDERR "Waiting for named pipe $dir/b\n";
sleep 1;
$pipeBRetryCount++;
if ( $pipeBRetryCount > 10 ) { die("Named pipe communication with usearch failed: $dir/b\n"); }
}
# read synchronized blocks from each stream
while (1) {
my ( $queryLabelA, $idsA, $queryLabelB, $idsB );
# idsA is a reference to a hash from targetIds to %ids
( $queryLabelA, $idsA, $nextLineA ) = collectIds( *UCA, $nextLineA );
( $queryLabelB, $idsB, $nextLineB ) = collectIds( *UCB, $nextLineB );
if ( !( $queryLabelA eq $queryLabelB ) ) {
die("Usearch results desynchronized: $queryLabelA neq $queryLabelB");
}
my $numHitsA = ( scalar keys %$idsA );
my $numHitsB = ( scalar keys %$idsB );
# if either read got NOHITS, then it's definitely NOHITS for the pair. This trumps TOOMANYHITS for the other read.
# don't bother trying to reconcile hits in this case
if ( ( $numHitsA == 0 ) || ( $numHitsB == 0 ) ) {
push @$nohitQueryIds, $queryLabelA;
}
# if both reads got TOOMANYHITS, then it's definitely TOOMANYHITS for the pair.
elsif ( ( $numHitsA >= $maxAccepts ) && ( $numHitsB >= $maxAccepts ) ) {
push @$tooManyHitQueryIds, $queryLabelA;
}
# if neither read got TOOMANYHITS, then we're good to go
elsif ( ( $numHitsA < $maxAccepts ) && ( $numHitsB < $maxAccepts ) ) {
if ( !reconcilePairedHitsAndPrint( $queryLabelA, $idsA, $idsB, $pairPercentIdThreshold ) ) {
push @$nohitQueryIds, $queryLabelA;
}
}
# if only one read got TOOMANYHITS...
else {
# escalate if possible
if ( $maxAccepts < $maxMaxAccepts ) {
push @$tooManyHitQueryIds, $queryLabelA;
}
# if we're already at maxMaxAccepts and we're allowed, fall back to the info provided by the other read.
# This is a little tricky: which percent ID threshold do we use?
# For consistency, I think we have to assume that the overly generic read is a 100% match.
# the cleanest way to say this is to say that the "overly generic" read just matches everything, so
# reconcilePairHitsAndPrint( $queryLabelA, $idsA, $allIds, $pairPercentIdThreshold )
# but that would require a hash %allIds mapping every ID to 100, just because reconcilePairHitsAndPrint says $idsB->{$targetLabel}
# it's equivalent to just use reconcileSingleHitsAndPrint with singlePercentIdThreshold.
elsif ($singleOKgeneric) {
if ( $numHitsA < $maxAccepts ) {
if ( !reconcileSingleHitsAndPrint( $queryLabelA, $idsA, $singlePercentIdThreshold ) ) {
push @$nohitQueryIds, $queryLabelA;
}
}
elsif ( $numHitsB < $maxAccepts ) {
if ( !reconcileSingleHitsAndPrint( $queryLabelA, $idsB, $singlePercentIdThreshold ) ) {
push @$nohitQueryIds, $queryLabelA;
}
}
else { die("impossible"); }
}
# if we're already at maxMaxAccepts, but not allowed to rely on the other read, just report TOOMANYHITS for the pair
else {
push @$tooManyHitQueryIds, $queryLabelA;
}
}
if ( !$nextLineA || !$nextLineB ) {
if ( !( !$nextLineA && !$nextLineB ) ) {
die("Usearch results desynchronized at end:\nA: $nextLineA\nB: $nextLineB");
}
last;
}
}
close(UCA) || die "can't close usearch: $!";
close(UCB) || die "can't close usearch: $!";
# for the TOOMANYHITS cases, escalate maxAccepts and try again
# Note this recurses, so no need to iterate here
if ( scalar(@$tooManyHitQueryIds) && ( $maxAccepts * 2 <= $maxMaxAccepts ) ) {
my $nohitQueryIdsB;
print STDERR "doPairSearch $readAFile, $readBFile: Escalating maxAccepts to "
. ( $maxAccepts * 2 ) . " for "
. scalar(@$tooManyHitQueryIds)
. " sequences.\n";
# prepare input files with the remaining sequences
my $readAFileEsc = extractFasta( $indexA, $tooManyHitQueryIds );
my $readBFileEsc = extractFasta( $indexB, $tooManyHitQueryIds );
( $nohitQueryIdsB, $tooManyHitQueryIds ) =
doPairSearch( $readAFileEsc, $readBFileEsc, $pairPercentDifferenceThreshold, $maxAccepts * 2 );
# A TOOMANYHITS case can certainly turn into a NOHITS case:
# once one read is no longer TOOMANYHITS, it may turn out that nothing can be reconciled with the other read.
#if (@$nohitQueryIdsB) {
# die( "A TOOMANYHITS case can't turn into a NOHITS case: "
# . ( join ", ", @$nohitQueryIdsB )
# . " at pair threshold $pairPercentDifferenceThreshold and maxAccepts "
# . ( $maxAccepts * 2 ) );
# }
push @$nohitQueryIds, @$nohitQueryIdsB;
}
print STDERR
"doPairSearch $readAFile, $readBFile: Finished at pair threshold $pairPercentDifferenceThreshold and maxAccepts $maxAccepts\n";
print STDERR " NOHITS: " . scalar(@$nohitQueryIds) . "\n";
print STDERR " TOOMANYHITS: " . scalar(@$tooManyHitQueryIds) . "\n";
# print STDERR " NOHITS: " . ( join ", ", @$nohitQueryIds ) . "\n";
# print STDERR " TOOMANYHITS: " . ( join ", ", @$tooManyHitQueryIds ) . "\n";
# any tooManyHitQueryIds that remain had more than maxMaxAccepts hits
return ( $nohitQueryIds, $tooManyHitQueryIds );
}
sub reconcilePairedHitsAndPrint {
my ( $queryLabel, $idsA, $idsB, $pairPercentIdThreshold ) = @_;
my $idsByIdentity = {};
for my $targetLabel ( keys %$idsA ) {
my $percentIdA = $idsA->{$targetLabel};
my $percentIdB = $idsB->{$targetLabel};
if ($percentIdB) {
# both reads hit the same target
# compute the average percent id
my $pairPercentId = ( $percentIdA + $percentIdB ) / 2.0;
if ( $pairPercentId >= $pairPercentIdThreshold ) {
if ( !defined $idsByIdentity->{$pairPercentId} ) {
$idsByIdentity->{$pairPercentId} = [];
}
push @{ $idsByIdentity->{$pairPercentId} }, $targetLabel;
}
}
}
if ( !%$idsByIdentity ) {
#print STDERR "$queryLabel -> no reconciled hits at $pairPercentIdThreshold%\n";
# this is the NOHIT case, but we'll back off and try again
return 0;
}
else {
#print STDERR "$queryLabel -> printing reconciled hits at $pairPercentIdThreshold%\n";
printLine( $queryLabel, $idsByIdentity );
return 1;
}
}
##### Read a block of target IDs for one query from a USEARCH stream
sub collectIds {
# assume that the records for a given label are all grouped together, so we can collect the hits for each label and process them in turn
# need to jump through some hoops to save the first line of the next block (since we can't rewind a pipe with seek()).
my $fh = shift;
my $firstLine = shift;
my %hits = ();
# process the first line (cached from the previous call)
#print STDERR "Processing firstLine: $firstLine\n";
my ( $typeF, $clusterF, $sizeF, $percentIdF, $strandF, $queryStartF, $targetStartF, $alignmentF, $queryLabelF, $targetLabelF ) =
split /\t/, $firstLine;
chomp $targetLabelF;
$queryLabelF =~ /$idRegex/;
$queryLabelF = $1;
#$queryLabelF =~ s/\s.*//; # primary ID is only the portion before whitespace
if ( $typeF eq "N" ) {
#print STDERR "$queryLabelF -> N\n";
# NOHIT represented as empty hash
# note we still have to cache the next line, after filtering comments
while (<$fh>) {
if (/^\s*#/) { next; }
if (/^\s*$/) { next; }
my $line = $_;
chomp $line;
#print STDERR "$line\n";
return ( $queryLabelF, \%hits, $line );
}
}
elsif ( $typeF eq "H" ) {
$hits{$targetLabelF} = $percentIdF;
}
# process the remaining lines
while (<$fh>) {
if (/^\s*#/) { next; }
if (/^\s*$/) { next; }
my $line = $_;
chomp $line;
#print STDERR "$line\n";
my ( $type, $cluster, $size, $percentId, $strand, $queryStart, $targetStart, $alignment, $queryLabel, $targetLabel ) =
split /\t/;
chomp $targetLabel;
$queryLabel =~ /$idRegex/;
$queryLabel = $1;
#$queryLabel =~ s/\s.*//; # primary ID is only the portion before whitespace
if ( $queryLabel ne $queryLabelF ) {
# advanced past the end of the current block
#print STDERR "End of block $queryLabelF -> block of " . scalar( keys %hits ) . " hits\n";
#print STDERR "End of block, return\n";
return ( $queryLabelF, \%hits, $line );
}
else {
if ( $type eq "N" ) {
die "impossible: $queryLabel reports no hit after it already had a hit";
}
elsif ( $type eq "H" ) {
$hits{$targetLabel} = $percentId;
}
}
}
# end of the stream (the last block)
#print STDERR "End of stream $queryLabelF -> block of " . scalar( keys %hits ) . " hits\n";
#print STDERR "End of stream, return\n";
return ( $queryLabelF, \%hits, "" );
}
my $fastaNum;
sub extractFasta {
my $index = shift;
my $ids = shift;
my $name = $fastaNum++;
# print STDERR "Extracting " . scalar(@$ids) . " to file $name\n";
open( OUT, ">$name" );
# my $out = Bio::SeqIO->new( '-format' => 'Fasta', '-fh' => \*OUT );
for my $id (@$ids) {
my $seqobj = $index->fetch($id);
if ( !defined $seqobj ) {
print STDERR "Extracting from " . $index->fastaFileName() . ": Undefined: $id\n";
}
# elsif ($seqobj->primary_id() ne $id)
# {
# print STDERR "Extracting from " . $index->filename() . ": ID problem: " . ($seqobj->primary_id()) . " ne $id\n";
# }
else {
#$out->write_seq($seqobj);
print OUT $seqobj;
}
}
close OUT;
return $name;
}
sub printLine {
my ( $label, $idsByIdentity ) = @_;
my @ids = ();
my @pcids = sort { $b <=> $a } keys %$idsByIdentity;
my $bestPcid = $pcids[0];
my $acceptPcid = $bestPcid - $slop;
for my $pcid (@pcids) {
if ( $pcid < $acceptPcid ) { last; }
push @ids, @{ $idsByIdentity->{$pcid} };
}
my $idString = ( join "\t", @ids );
print "$label\t$bestPcid\t" . $idString . "\n";
# print STDERR "$label\t$bestPcid\t" . $idString . "\n";
}
sub revcompFile {
my $infile = shift;
my $outfile = $infile . ".rc";
open( IN, $infile ) or die "Can't read $infile\n";
open( OUT, ">$outfile" ) or die "Can't write $outfile\n";
while (<IN>) {
my $a = $_;
chomp $a;
if ( !( $a =~ /^>/ ) ) {
$a = reverse $a;
$a =~ tr/ACGTacgt/TGCAtgca/;
}
print OUT $a . "\n";
}
close IN;
close OUT;
return $outfile;
}
main();
|