Mercurial > repos > okorol > itsx
view ITSx @ 9:3610c6312b85 draft default tip
Uploaded
author | okorol |
---|---|
date | Fri, 05 Jun 2015 12:05:37 -0400 |
parents | f82c70f54bd7 |
children |
line wrap: on
line source
#!/usr/bin/perl # ITSx ITS Extractor $app_title = "ITSx -- Identifies ITS sequences and extracts the ITS region"; $app_author = "Johan Bengtsson-Palme et al., University of Gothenburg"; $app_version = "1.0.11"; $app_message = ""; # ----------------------------------------------------------------- # # License information $license = " ITSx - ITS Extractor -- Identifies ITS sequences and extracts the ITS region\ Copyright (C) 2012-2014 Johan Bengtsson-Palme et al.\ \ This program is free software: you can redistribute it and/or modify\ it under the terms of the GNU General Public License as published by\ the Free Software Foundation, either version 3 of the License, or\ (at your option) any later version.\ \ This program is distributed in the hope that it will be useful,\ but WITHOUT ANY WARRANTY; without even the implied warranty of\ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\ GNU General Public License for more details.\ \ You should have received a copy of the GNU General Public License\ along with this program, in a file called 'license.txt'\ If not, see: http://www.gnu.org/licenses/.\ "; ## BUGS: $bugs = "New features in this version ($app_version):\ - None\ \ Fixed bugs in this version ($app_version):\ - Fixed a bug causing newline characters to be occasionally skipped in the 'its1.full_and_partial.fasta' FASTA output file when the '--anchor' option was used\ \ Known bugs in this version ($app_version):\ - None\ "; ## OPTIONS: $options = "\ -i {file} : DNA FASTA input file to investigate\ -o {file} : Base for the names of output file(s)\ -p {directory} : A path to a directory of HMM-profile collections representing ITS conserved regions, default is in the same directory as ITSx itself\ --date {T or F} : Adds a date and time stamp to the output directory, off (F) by default\ --reset {T or F} : Re-creates the HMM-database before ITSx is run, off (F) by default\ Sequence selection options:\ -t {character code} : Profile set to use for the search, see the User's Guide (comma-separated), default is all\ -E {value} : Domain E-value cutoff for a sequence to be included in the output, default = 1e-5\ -S {value} : Domain score cutoff for a sequence to be included in the output, default = 0\ -N {value} : The minimal number of domains that must match a sequence before it is included, default = 2\ --selection_priority {sum, domains, eval, score} : Selects what will be of highest priority when determining the origin of the sequence, default is sum\ --search_eval {value} : The E-value cutoff used in the HMMER search, high numbers may slow down the process, cannot be used with the --search_score option, default is 0.01\ --search_score {value} : The score cutoff used in the HMMER search, low numbers may slow down the process, cannot be used with the --search_eval option, default is to used E-value cutoff, not score\ --allow_single_domain {e-value,score or F} : Allow inclusion of sequences that only find a single domain, given that they meet the given E-value and score thresholds, on with parameters 1e-9,0 by default\ --allow_reorder {T or F} : Allows profiles to be in the wrong order on extracted sequences, off (F) by default\ --complement {T or F} : Checks both DNA strands against the database, creating reverse complements, on (T) by default\ --cpu {value} : the number of CPU threads to use, default is 1\ --multi_thread {T or F} : Multi-thread the HMMER-search, on (T) if number of CPUs (--cpu option > 1), else off (F) by default\ --heuristics {T or F} : Selects whether to use HMMER's heuristic filtering, off (F) by default\ Output options:\ --summary {T or F} : Summary of results output, on (T) by default\ --graphical {T or F} : 'Graphical' output, on (T) by default\ --fasta {T or F} : FASTA-format output of extracted ITS sequences, on (T) by default\ --preserve {T or F} : Preserve sequence headers in input file instead of printing out ITSx headers, off (F) by default\ --save_regions {SSU,ITS1,5.8S,ITS2,LSU,all,none} : A comma separated list of regions to output separate FASTA files for, 'ITS1,ITS2' by default\ --anchor {integer or HMM} : Saves an additional number of bases before and after each extracted region. If set to 'HMM' all bases matching the corresponding HMM will be output, default = 0\ --only_full {T or F} : If true, output is limited to full-length regions, off (F) by default\ --partial {integer} : Saves additional FASTA-files for full and partial ITS sequences longer than the specified cutoff, default = 0 (off)\ --concat {T or F} : Saves a FASTA-file with concatenated ITS sequences (with 5.8S removed), off (F) by default\ --minlen {integer} : Minimum length the ITS regions must be to be outputted in the concatenated file (see above), default = 0\ --positions {T or F} : Table format output containing the positions ITS sequences were found in, on (T) by default\ --table {T or F} : Table format output of sequences containing probable ITS sequences, off (F) by default\ --not_found {T or F} : Saves a list of non-found entries, on (T) by default\ --detailed_results {T or F} : Saves a tab-separated list of all results, off (F) by default\ --truncate {T or F} : Truncates the FASTA output to only contain the actual ITS sequences found, on (T) by default\ --silent {T or F} : Supresses printing progress info to stderr, off (F) by default\ --graph_scale {value} : Sets the scale of the graph output, if value is zero, a percentage view is shown, default = 0\ --save_raw {T or F} : Saves all raw data for searches etc. instead of removing it on finish, off (F) by default\ -h : displays this help message\ --help : displays this help message\ --bugs : displays the bug fixes and known bugs in this version of ITSx\ --license : displays licensing information\ "; ## Print title message print STDERR "$app_title\nby $app_author\nVersion: $app_version\n$app_message"; print STDERR "-----------------------------------------------------------------\n"; ## Setup default variable values use List::Util qw(first max maxstr min minstr reduce shuffle sum); $bindir = $0; $bindir =~ s/_x//; $input = ""; $output = "ITSx_out"; $hmmscan = ""; $profileDB = "$bindir\_db/HMMs"; $type = "all"; $E = 1e-5; $S = 0; $N = 2; $priority = "sum"; $search_eval = 0.01; $search_score = ""; $allow_single_E = 1e-9; $allow_single_score = 0; #$allow_single_E = -1; # Turns off single-domain matching by E-value #$allow_single_score = 0; # Turns off single-domain matching by score $allow_reorder = 0; $complement = 1; $cpu = 1; $multi_thread = "unset"; $heuristics = 0; $out_sum = 1; $out_graph = 1; $out_fasta = 1; $out_preserve = 0; $out_ssu = 0; $out_its1 = 1; $out_its2 = 1; $out_58S = 0; $out_lsu = 0; $out_pos = 1; $out_table = 0; $out_not = 1; $out_date = 0; $out_joined = 0; $out_results = 0; $out_partial = 0; $out_concat = 0; $concat_minlen = 0; $truncate = 1; $anchor = 0; $only_full = 0; $graph_scale = 0; $debug = 0; $reset = 0; ## Read command-line options for ($i = 0; $i <= scalar(@ARGV); $i++) { # Goes through the list of arguments $arg = @ARGV[$i]; # Stores the current argument in $arg if ($arg eq "-i") { # Read input files from -i flag $i++; $input = @ARGV[$i]; } if ($arg eq "-o") { # Read output files from -o flag $i++; $output = @ARGV[$i]; } if ($arg eq "-p") { # Read profile database from -p flag $i++; $profileDB = @ARGV[$i]; } if ($arg eq "--hmmscan") { # Read pre-computed hmmscan output file from --hmmscan flag ('undocumented' feature) $i++; $hmmscan = @ARGV[$i]; } if ($arg eq "--date") { # Determine whether or not to add a date stamp based on the --date flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Ff0]/) { # Check if argument begins with "F", "f", or "0" $out_date = 0; } else { $out_date = 1; } } if ($arg eq "--reset") { # Reset HMM database? $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Ff0]/) { # Check if argument begins with "F", "f", or "0" $reset = 0; } else { $reset = 1; } } if ($arg eq "-t") { # Select what types of ITSs to look for using the -t flag $i++; $type = @ARGV[$i]; } if ($arg eq "-E") { # Set the E-value cutoff using the -E flag $i++; $E = @ARGV[$i]; } if ($arg eq "-S") { # Set the score cutoff using the -S flag $i++; $S = @ARGV[$i]; } if ($arg eq "-N") { # Set the number of found domains cutoff using the -N flag $i++; $N = @ARGV[$i]; } if ($arg eq "--selection_priority") { # Set how to order the ITS types using the --selection_priority flag $i++; $priority = @ARGV[$i]; } if ($arg eq "--search_eval") { # Set the E-value cutoff for the HMMER search using the --search_eval flag $i++; $search_eval = @ARGV[$i]; $search_score = ""; # Turns off score cutoff } if ($arg eq "--search_score") { # Set the score cutoff for the HMMER search using the --search_score flag $i++; $search_score = @ARGV[$i]; $search_eval = ""; # Turns off E-value cutoff } if ($arg eq "--allow_single_domain") { # Determine whether or not to allow single domain matches based on the --allow_single_domain flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Ff0]/) { # Check if argument begins with "F", "f", or "0" $allow_single_E = -1; # Turns off single-domain matching by E-value $allow_single_score = 0; # Turns off single-domain matching by score } else { ($allow_single_E,$allow_single_score) = split(',',@ARGV[$i]); # Turns on single-domain matching, assigning the first given value as the E-value cutoff, and the second as score cutoff } } if ($arg eq "--allow_reorder") { # Determine whether or not to allow the domains to be in the wrong order based on the --allow_reorder flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $allow_reorder = 1; } else { $allow_reorder = 0; } } if ($arg eq "--complement") { # Determine whether or not to scan the complementary strand of the input file based on the --complement flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $complement = 1; } else { $complement = 0; } } if ($arg eq "--cpu") { # Set the number of CPUs to use based on the --cpu flag $i++; $cpu = @ARGV[$i]; } if ($arg eq "--multi_thread") { # Determine whether or not to multi-thread the HMMER step based on the --multi_thread flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $multi_thread = 1; } else { $multi_thread = 0; } } if ($arg eq "--heuristics") { # Determine whether or not to use HMMER's heuristic filtering based on the --heuristics flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $heuristics = 1; } else { $heuristics = 0; } } if ($arg eq "--summary") { # Determine whether or not to output a summary based on the --summary flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_sum = 1; } else { $out_sum = 0; } } if ($arg eq "--graphical") { # Determine whether or not to output a graphical representation of matches based on the --graphical flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_graph = 1; } else { $out_graph = 0; } } if ($arg eq "--detailed_results") { # Determine whether or not to output a detailed results list, based on the --detailed_results flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_results = 1; } else { $out_results = 0; } } if ($arg eq "--partial") { # Set the full-and-partial cutoff $i++; $out_partial = @ARGV[$i]; } if ($arg eq "--anchor") { # Set the length of the sequence "anchors" $i++; $anchor = @ARGV[$i]; } if ($arg eq "--only_full") { # Output only full-length regions $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $only_full = 1; } else { $only_full = 0; } } if ($arg eq "--save_regions") { # Determine which regions to output FASTA files for based on the --save_regions flag $i++; @save_regions = split(',',uc(@ARGV[$i])); $out_ssu = 0; $out_its1 = 0; $out_its2 = 0; $out_58S = 0; $out_lsu = 0; foreach $save_region (@save_regions) { if ($save_region eq "SSU") { $out_ssu = 1; } if ($save_region eq "ITS1") { $out_its1 = 1; } if ($save_region eq "5.8S") { $out_58S = 1; } if ($save_region eq "ITS2") { $out_its2 = 1; } if ($save_region eq "LSU") { $out_lsu = 1; } if ($save_region eq "ALL") { $out_ssu = 1; $out_its1 = 1; $out_its2 = 1; $out_58S = 1; $out_lsu = 1; } if ($save_region eq "NONE") { $out_ssu = 0; $out_its1 = 0; $out_its2 = 0; $out_58S = 0; $out_lsu = 0; } } } if ($arg eq "--positions") { # Determine whether or not to output a positions file based on the --positions flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_pos = 1; } else { $out_pos = 0; } } if ($arg eq "--concat") { # Determine whether or not to output a concatednated ITS1 + ITS2 file $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_concat = 1; } else { $out_concat = 0; } } if ($arg eq "--minlen") { # Set the min length of the combined ITS1 and ITS2 sequences for concatenation $i++; $concat_minlen = @ARGV[$i]; } if ($arg eq "--fasta") { # Determine whether or not to output FASTA-files based on the --fasta flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_fasta = 1; } else { $out_fasta = 0; } } if ($arg eq "--preserve") { # Determine whether or not to preserve FASTA-headers based on the --preserve flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_preserve = 1; } else { $out_preserve = 0; } } if ($arg eq "--joined") { # Determine whether or not to output a FASTA-file containing ALL sorts of output sequences (for debugging) $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_joined = 1; } else { $out_joined = 0; } } if ($arg eq "--table") { # Determine whether or not to output tables of all potential matches based on the --table flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_table = 1; } else { $out_table = 0; } } if ($arg eq "--not_found") { # Determine whether or not to output a list of sequences that are not ITSs based on the --not_found flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $out_not = 1; } else { $out_not = 0; } } if ($arg eq "--silent") { # Determine whether or not to output anything to the screen based on the --silent flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $silent = 1; } else { $silent = 0; } } if ($arg eq "--graph_scale") { # Set the scale of the graphical output based on the --graph_scale flag $i++; $graph_scale = @ARGV[$i]; } if ($arg eq "--save_raw") { # Determine whether or not to save all the raw intermediate data based on the --save_raw flag $i++; if (substr(@ARGV[$i],0,1) =~ m/^[Tt1]/) { # Check if argument begins with "T", "t", or "1" $save_raw = 1; } else { $save_raw = 0; } } ## If "-h" or "--help" are among the options, output usage data and options if (($arg eq "-h") || ($arg eq "--help")) { print "Usage: ITSx -i <input file> -o <output file>\nOptions:$options"; print "-----------------------------------------------------------------\n"; exit; # Exit ITSx } ## If "--bugs" is among the options, output bugs and features information if ($arg eq "--bugs") { print "$bugs\n"; exit; # Exit ITSx } ## If "--license" is among the options, output license information if ($arg eq "--license") { print "$license\n"; exit; # Exit ITSx } if ($arg eq "--debug") { # Run ITSx in debug mode $debug = 1; } if ($arg eq "--pipeline") { # Run ITSx in pipeline mode $pipeline = 1; } } ## Setup some variables dependent on input if ($multi_thread eq "unset") { # If the multi-thread option is not set if ($cpu > 1) { # Then if the number of CPUs used > 1, then multi-thread HMMER searches $multi_thread = 1; } else { # Else, run HMMER searches sequentially on one CPU $multi_thread = 0; } } if ($hmmscan ne "") { # If a pre-computed hmmscan output is supplied $output = $hmmscan; # Then set the base of the output directory name to be the same as that hmmscan output file } ## Check for binaries chomp($path = `which hmmpress`); # Get the path for hmmpress if ($path eq "") { # If the path is empty, then show an error message and exit ITSx print STDERR "FATAL ERROR :: Could not locate HMMER binaries! It seems that hmmpress is not installed properly.\ Consult the manual for installation instructions. Note that HMMER3 is required. Previous HMMER-versions will not work.\ This error is fatal, and ITSx will now abort.\n"; print STDERR "-----------------------------------------------------------------\n"; exit; } chomp($path = `which hmmscan`); # Get the path for hmmscan if ($path eq "") { # If the path is empty, then show an error message and exit ITSx print STDERR "FATAL ERROR :: Could not locate HMMER binaries! It seems that hmmscan is not installed properly.\ Consult the manual for installation instructions. Note that HMMER3 is required. Previous HMMER-versions will not work.\ This error is fatal, and ITSx will now abort.\n"; print STDERR "-----------------------------------------------------------------\n"; exit; } ## Check for database chomp($errormsg = `ls $profileDB* 2>&1 1>/dev/null`); # Get the error msg when looking for the profile database if (substr($errormsg,0,4) eq "ls: ") { # If the error message begins with "ls: ", then show an error message and exit ITSx print STDERR "FATAL ERROR :: The specified profile database could not be found.\ Consult the manual for installation instructions.\ This error is fatal, and ITSx will now abort.\n"; print STDERR "-----------------------------------------------------------------\n"; exit; } if ($pipeline == 0) { # If ITSx is not run in pipeline mode (i.e. from ITSx) if ($out_date == 1) { # If a date and time stamp should be supplied ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = localtime(time); # Get the date and time $year = $year + 1900; # Format the year $mon = $mon + 1; # Format the month if ($mon < 10) { # Add a zero to the month, if needed $mon = "0" . $mon; } if ($mday < 10) { # Add a zero to the day, if needed $mday = "0" . $mday; } if ($hour < 10) { # Add a zero to the hour, if needed $hour = "0" . $hour; } if ($min < 10) { # Add a zero to the minute, if needed $min = "0" . $min; } $outputDate = ".$year\-$mon\-$mday\_$hour.$min"; # Create a date and time stamp $outputDate =~ s./.-.g; # Remove any potential slashes in the output name (as this will confuse ITSx's file naming) $output = $output . $outputDate; # Add the date and time stamp top the output base name } } $tempDir = "ITSx_temp_directory__$output"; # Setup a temporary directory variable $tempDir =~ s./.-.g; # Remove any potential slashes in the output name (as this will confuse ITSx's file naming) if ($pipeline == 0) { # If not running in pipeline mode ## Create a summary file if ($out_sum == 1) { # If summary output is on $now = localtime; # Get the current time open (SUMMARY, ">$output.summary.txt"); # Create the summary file print SUMMARY "ITSx run started at $now.\n"; # Output the starting time for the analysis print SUMMARY "-----------------------------------------------------------------\n"; close (SUMMARY); # Close summary file } } ## Create a temporary directory for ITSx if ($pipeline == 0) { # If ITSx is not run in pipeline mode (i.e. from ITSx) `mkdir $tempDir 2> /dev/null`; # Create a temporary directory } ## Prepare profile database ## Get the current time and output info message $now = localtime; if ($silent == 0) { print STDERR "$now : Preparing HMM database (should be quick)...\n"; } ## Setup profile index %profileIndex = {}; # A = Alveolata # B = Bryophyta # C = Bacillariophyta # D = Amoebozoa # E = Euglenozoa # F = Fungi # G = Chlorophyta (green algae) # H = Rhodophyta (red algae) # I = Phaeophyceae (brown algae) # L = Marchantiophyta (liverworts) # M = Metazoa # N = Microsporidia # O = Oomycota # P = Haptophyceae (prymnesiophytes) # Q = Raphidophyceae # R = Rhizaria # S = Synurophyceae # T = Tracheophyta (higher plants) # U = Eustigmatophyceae # X = Apusozoa # Y = Parabasalia $profileIndex{"A"} = "alveolates"; $profileIndex{"B"} = "bryophyta"; $profileIndex{"C"} = "bacillariophyta"; $profileIndex{"D"} = "amoebozoa"; $profileIndex{"E"} = "euglenozoa"; $profileIndex{"F"} = "fungi"; $profileIndex{"G"} = "chlorophyta"; $profileIndex{"H"} = "rhodophyta"; $profileIndex{"I"} = "phaeophyceae"; $profileIndex{"J"} = "undefined"; $profileIndex{"K"} = "undefined"; $profileIndex{"L"} = "marchantiophyta"; $profileIndex{"M"} = "metazoa"; $profileIndex{"N"} = "microsporidia"; $profileIndex{"O"} = "oomycota"; $profileIndex{"P"} = "haptophyceae"; $profileIndex{"Q"} = "raphidophyceae"; $profileIndex{"R"} = "rhizaria"; $profileIndex{"S"} = "synurophyceae"; $profileIndex{"T"} = "tracheophyta"; $profileIndex{"U"} = "eustigmatophyceae"; $profileIndex{"V"} = "undefined"; $profileIndex{"W"} = "undefined"; $profileIndex{"X"} = "apusozoa"; $profileIndex{"Y"} = "parabasalia"; $profileIndex{"Z"} = "undefined"; @profileList = split(',',uc($type)); # Get the list of profile types foreach $entry (@profileList) { # Go through the entered types if (($entry eq "ALL") || ($entry eq ".")) { # If "all" among the entries push(@profileSet,"A"); # Add the alveolates profiles to the investigation set push(@profileSet,"B"); # Add the bryophytes profiles to the investigation set push(@profileSet,"C"); # Add the bacillariophyta profiles to the investigation set push(@profileSet,"D"); # Add the amoebozoa profiles to the investigation set push(@profileSet,"E"); # Add the euglenozoa profiles to the investigation set push(@profileSet,"F"); # Add the fungi profiles to the investigation set push(@profileSet,"G"); # Add the green algae profiles to the investigation set push(@profileSet,"H"); # Add the red algae profiles to the investigation set push(@profileSet,"I"); # Add the brown algae profiles to the investigation set push(@profileSet,"L"); # Add the liverworts profiles to the investigation set push(@profileSet,"M"); # Add the metazoa profiles to the investigation set #push(@profileSet,"N"); # Add the microsporidia profiles to the investigation set push(@profileSet,"O"); # Add the oomycetes profiles to the investigation set push(@profileSet,"P"); # Add the prymnesiophytes profiles to the investigation set push(@profileSet,"Q"); # Add the raphidophytes profiles to the investigation set push(@profileSet,"R"); # Add the rhizaria profiles to the investigation set push(@profileSet,"S"); # Add the synurophyceae profiles to the investigation set push(@profileSet,"T"); # Add the tracheophyta (higher plants) profiles to the investigation set push(@profileSet,"U"); # Add the eustigmatophytes profiles to the investigation set #push(@profileSet,"X"); # Add the apusozoa profiles to the investigation set #push(@profileSet,"Y"); # Add the parabasalia profiles to the investigation set } else { if (length($entry) == 1) { # If the name has only one character if ($entry =~ m/[ABCDERFHILMNOPQRSTUXY]/) { # If the selected set exists push(@profileSet,$entry); # Add the selected profiles to the investigation set } } else { if ($entry =~ m/ALVEOL/) { push(@profileSet,"A"); # Add the alveolates profiles to the investigation set } if ($entry =~ m/BRYO/) { push(@profileSet,"B"); # Add the bryophytes profiles to the investigation set } if ($entry =~ m/MOSS/) { push(@profileSet,"B"); # Add the bryophytes profiles to the investigation set } if ($entry =~ m/BACILL/) { push(@profileSet,"C"); # Add the bacillariophyta profiles to the investigation set } if ($entry =~ m/DIATOM/) { push(@profileSet,"C"); # Add the bacillariophyta profiles to the investigation set } if ($entry =~ m/AMOEB/) { push(@profileSet,"D"); # Add the amoebozoa profiles to the investigation set } if ($entry =~ m/EUGLE/) { push(@profileSet,"E"); # Add the euglenozoa profiles to the investigation set } if ($entry =~ m/FUNG/) { push(@profileSet,"F"); # Add the fungi profiles to the investigation set } if ($entry =~ m/GREEN/) { push(@profileSet,"G"); # Add the green algae profiles to the investigation set } if ($entry =~ m/CHLORO/) { push(@profileSet,"G"); # Add the green algae profiles to the investigation set } if ($entry =~ m/RED-AL/) { push(@profileSet,"H"); # Add the red algae profiles to the investigation set } if ($entry =~ m/RHODO/) { push(@profileSet,"H"); # Add the red algae profiles to the investigation set } if ($entry =~ m/BROWN/) { push(@profileSet,"I"); # Add the brown algae profiles to the investigation set } if ($entry =~ m/PHAEOP/) { push(@profileSet,"I"); # Add the brown algae profiles to the investigation set } if ($entry =~ m/LIVER/) { push(@profileSet,"L"); # Add the liverworts profiles to the investigation set } if ($entry =~ m/MARCH/) { push(@profileSet,"L"); # Add the liverworts profiles to the investigation set } if ($entry =~ m/METAZ/) { push(@profileSet,"M"); # Add the metazoa profiles to the investigation set } if ($entry =~ m/ANIMAL/) { push(@profileSet,"M"); # Add the metazoa profiles to the investigation set } if ($entry =~ m/MICROSPOR/) { push(@profileSet,"N"); # Add the microsporidia profiles to the investigation set } if ($entry =~ m/OOMYC/) { push(@profileSet,"O"); # Add the oomycetes profiles to the investigation set } if ($entry =~ m/PRYMN/) { push(@profileSet,"P"); # Add the prymnesiophytes profiles to the investigation set } if ($entry =~ m/HAPTO/) { push(@profileSet,"P"); # Add the prymnesiophytes profiles to the investigation set } if ($entry =~ m/RAPHID/) { push(@profileSet,"Q"); # Add the raphidophytes profiles to the investigation set } if ($entry =~ m/RHIZA/) { push(@profileSet,"R"); # Add the rhizaria profiles to the investigation set } if ($entry =~ m/SYNUR/) { push(@profileSet,"S"); # Add the synurophyceae profiles to the investigation set } if ($entry =~ m/TRACHE/) { push(@profileSet,"T"); # Add the tracheophyta profiles to the investigation set } if ($entry =~ m/PLANTS/) { push(@profileSet,"T"); # Add the tracheophyta profiles to the investigation set } if ($entry =~ m/EUSTIG/) { push(@profileSet,"U"); # Add the eustigmatophytes profiles to the investigation set } if ($entry =~ m/APUSO/) { push(@profileSet,"X"); # Add the apusozoa profiles to the investigation set } if ($entry =~ m/PARAB/) { push(@profileSet,"Y"); # Add the parabasalia profiles to the investigation set } } } } foreach $set (@profileSet) { # For each set of profiles in the the full profile set for investigation $hmmPath = $profileDB . "/" . $set . ".hmm"; # Determine the path to the HMM-file chomp($modelCount = `grep -c "//" $hmmPath`); # Count the number of models in the HMM-file push(@modelCount,$modelCount); # Add the number of models in this HMM-file to the list of model counts if ($reset == 1) { `rm -f $hmmPath.h3* 2> /dev/null`; # Delete old HMM-files `hmmpress $hmmPath 2> /dev/null`; # Prepare the HMM-file for searching ## Redirecting stderr is a quick and dirty solution to get rid of the messages... Could be made more elegant } } ## Clean-up input files and create complementary strand if needed ## Get the current time and output an info message $now = localtime; if ($silent == 0) { print STDERR "$now : Checking and handling input sequence data (should not take long)...\n"; } ## Open the summary file for writing if ($out_sum == 1) { # If summary output is on open (SUMMARY, ">>$output.summary.txt"); # Append to the summary file } if ($input ne "") { # If an input file is given ## Read from file open (SEQUENCES, $input); # Open the input file for reading open (MAIN, ">$tempDir/main.fasta"); # Create a temporary file for storing the cleaned sequences representing the main strand open (COMPLEMENT, ">$tempDir/complement.fasta"); # Create a temporary file for storing the cleaned sequences representing the complementary strand $inputSequenceCount = 0; # Reset input sequence counter while ($sequence = <SEQUENCES>) { # Repeat for every line in the input file chomp($sequence); # Truncate any potential line feeds $sequence =~ tr/\r\n//d; # Remove all carriage return and new line characters if (substr($sequence,0,1) eq ">") { # If a new FASTA entry is found in the input file $inputSequenceCount++; # Add one to the input sequence counter print MAIN $mainSeq . "\n"; # Write the previous main DNA sequence to the main sequence file print MAIN $sequence . " main\n"; # Write the definition line of the new sequence to the main sequence file $sequenceDB{"$sequenceID"} = $mainSeq; # Add sequence to sequence database $headers{"$sequenceID"} = $header; # Add the header to the header database push(@sequenceOrder,$sequenceID); # Add this sequence ID to the ordered list of sequences ($sequenceID) = split(" ",substr($sequence,1)); $header = $sequence; # Save the sequence header $mainSeq = ""; # Empty the main sequence entry if ($complement == 1) { # If the complementary file should be written $complementSeq = reverse($complementSeq); # Reverse the complementary DNA sequence print COMPLEMENT $complementSeq . "\n"; # Write the previous complementary DNA sequence to the complementary sequence file print COMPLEMENT $sequence . " complement\n"; # Write the definition line of the new sequence to the complement sequence file $complementSeq = ""; # Empty the complementary sequence entry } } else { # If this line is just a continuation of the current DNA sequence $mseq = $sequence; # Store this part of the DNA sequence in the intermediate varaible $mseq $mseq =~ s/[ .-]//g; # Remove any gap characters present in this sequence part (good if input was an alignment) $mseq =~ tr/[a-z]/[A-Z]/; # Make all letters uppercase $mseq =~ s/[^A-Z]//g; # Remove all non-alphabetic characters $mseq =~ tr/U/T/; # Exchanges U:s for T:s (Uracil to Thymine, good if input was RNA sequence) $mainSeq = $mainSeq . $mseq; # Add the intermediate DNA sequence to the end of the main DNA sequence entry if ($complement == 1) { # If the complementary file should be written $cseq = $sequence; # Store this part of the DNA sequence in the intermediate varaible $cseq $cseq =~ s/[ .-]//g; # Remove any gap characters present in this sequence part (good if input was an alignment) $cseq =~ tr/[a-z]/[A-Z]/; # Make all letters uppercase $cseq =~ s/[^A-Z]//g; # Remove all non-alphabetic characters $cseq =~ tr/ACGTURYSWKMBDHVN/TGCAAYRWSMKVHDBN/; # Replace all characters with its complementary base $complementSeq = $complementSeq . $cseq; # Add the intermediate DNA sequence to the end of the complementary DNA sequence entry } } } ## When the input file's end is reached print MAIN $mainSeq . "\n"; # Write the last main DNA sequence to the main sequence file if ($complement == 1) { # If the complementary file should be written $complementSeq = reverse($complementSeq); # Reverse the complementary DNA sequence print COMPLEMENT $complementSeq . "\n"; # Write the last complementary DNA sequence to the complementary sequence file } $sequenceDB{"$sequenceID"} = $mainSeq; # Add sequence to sequence database $headers{"$sequenceID"} = $header; # Add the header to the header database push(@sequenceOrder,$sequenceID); # Add this sequence ID to the ordered list of sequences $mainSeq = ""; # Empty the main sequence entry $complementSeq = ""; # Empty the complementary sequence entry close (SEQUENCES); # Close the sequence input file close (COMPLEMENT); # Close the complementary output file close (MAIN); # Close the main output file } else { # If no input file is supplied, then read from stdin instead $input = "$tempDir/main.fasta"; # Set up a temporary input file path open (MAIN, ">$tempDir/main.fasta"); # Create a temporary file for storing the cleaned sequences representing the main strand open (COMPLEMENT, ">$tempDir/complement.fasta"); # Create a temporary file for storing the cleaned sequences representing the complementary strand $inputSequenceCount = 0; # Reset input sequence counter while ($sequence = <STDIN>) { # Repeat for every line in the standard input chomp($sequence); # Truncate any potential line feeds if (substr($sequence,0,1) eq ">") { # If a new FASTA entry is found in the input $inputSequenceCount++; # Add one to the input sequence counter print MAIN $mainSeq . "\n"; # Write the previous main DNA sequence to the main sequence file print MAIN $sequence . " main\n"; # Write the definition line of the new sequence to the main sequence file $sequenceDB{"$sequenceID"} = $mainSeq; # Add sequence to sequence database $headers{"$sequenceID"} = $header; # Add the header to the header database push(@sequenceOrder,$sequenceID); # Add this sequence ID to the ordered list of sequences $sequenceID = split(" ",substr($sequence,1)); $header = $sequence; # Save the sequence header $mainSeq = ""; # Empty the main sequence entry if ($complement == 1) { # If the complementary file should be written $complementSeq = reverse($complementSeq); # Reverse the complementary DNA sequence print COMPLEMENT $complementSeq . "\n"; # Write the previous complementary DNA sequence to the complementary sequence file print COMPLEMENT $sequence . " complement\n"; # Write the definition line of the new sequence to the complement sequence file $complementSeq = ""; # Empty the complementary sequence entry } } else { # If this line is just a continuation of the current DNA sequence $mseq = $sequence; # Store this part of the DNA sequence in the intermediate varaible $mseq $mseq =~ s/[ .-]//g; # Remove any gap characters present in this sequence part (good if input was an alignment) $mseq =~ tr/[a-z]/[A-Z]/; # Make all letters uppercase $mseq =~ s/[^A-Z]//g; # Remove all non-alphabetic characters $mseq =~ tr/U/T/; # Exchanges U:s for T:s (Uracil to Thymine, good if input was RNA sequence) $mainSeq = $mainSeq . $mseq; # Add the intermediate DNA sequence to the end of the main DNA sequence entry if ($complement == 1) { # If the complementary file should be written $cseq = $sequence; # Store this part of the DNA sequence in the intermediate varaible $cseq $cseq =~ s/[ .-]//g; # Remove any gap characters present in this sequence part (good if input was an alignment) $cseq =~ tr/[a-z]/[A-Z]/; # Make all letters uppercase $cseq =~ s/[^A-Z]//g; # Remove all non-alphabetic characters $cseq =~ tr/ACGTURYSWKMBDHVN/TGCAAYRWSMKVHDBN/; # Replace all characters with its complementary base $complementSeq = $complementSeq . $cseq; # Add the intermediate DNA sequence to the end of the complementary DNA sequence entry } } } ## When the input file's end is reached print MAIN $mainSeq . "\n"; # Write the last main DNA sequence to the main sequence file if ($complement == 1) { # If the complementary file should be written $complementSeq = reverse($complementSeq); # Reverse the complementary DNA sequence print COMPLEMENT $complementSeq . "\n"; # Write the last complementary DNA sequence to the complementary sequence file } $sequenceDB{"$sequenceID"} = $mainSeq; # Add sequence to sequence database $headers{"$sequenceID"} = $header; # Add the header to the header database push(@sequenceOrder,$sequenceID); # Add this sequence ID to the ordered list of sequences $mainSeq = ""; # Empty the main sequence entry $complementSeq = ""; # Empty the complementary sequence entry close (COMPLEMENT); # Close the complementary output file close (MAIN); # Close the main output file } if ($out_sum == 1) { # If summary output should be written print SUMMARY "Number of sequences in input file: \t$inputSequenceCount\n"; # Write info on the number of input sequences to the summary file } ## Perform HMM-scan if ($hmmscan eq "") { # If a pre-computed hmmscan output file is not supplied if ($heuristics == 0) { # If HMMER's heuristic filtering should not be used $heurMax = "--max"; # Set the heurMax to "--max" (indicating that HMMER should turn off filtering) } else { $heurMax = ""; # Set the heurMax to empty (indicating that HMMER should turn on filtering) } if ($multi_thread == 0) { # If multi-threading is off ## Get the current time and output info message $now = localtime; if ($silent == 0) { print STDERR "$now : Comparing sequences to HMM database (this may take a long while)...\n"; } foreach $set (@profileSet) { # Go sequentially through all profile sets to search for $hmmPath = $profileDB . "/" . $set . ".hmm"; # Set the path to the HMM-file of the current set if ($search_eval ne "") { # If E-value cutoff is use for the search hmmerSearch("hmmscan --cpu $cpu $heurMax -E $search_eval $hmmPath $tempDir/main.fasta","$tempDir/main.$set.hmmscan","M",$set); # Call HMMER with E-value cutoff } else { # If score cutoff is use for the search hmmerSearch("hmmscan --cpu $cpu $heurMax -T $search_score $hmmPath $tempDir/main.fasta","$tempDir/main.$set.hmmscan","M",$set); # Call HMMER with score cutoff } if ($complement == 1) { # If the complementary file should be scanned if ($search_eval ne "") { # If E-value cutoff is use for the search hmmerSearch("hmmscan --cpu $cpu $heurMax -E $search_eval $hmmPath $tempDir/complement.fasta","$tempDir/complement.$set.hmmscan","C",$set); # Call HMMER with E-value cutoff } else { # If score cutoff is use for the search hmmerSearch("hmmscan --cpu $cpu $heurMax -T $search_score $hmmPath $tempDir/complement.fasta","$tempDir/complement.$set.hmmscan","C",$set); # Call HMMER with score cutoff } } } } else { # If multi-threading is on ## Get the current time and output info message $now = localtime; if ($silent == 0) { print STDERR "$now : Doing paralellised comparison to HMM database (this may take a long while)...\n"; } ## Determining number of cpus per thread if ($complement == 1) { # If the complementary file should be scanned $hmmcpu = int(0.5 * $cpu / scalar(@profileSet)); # Assign X CPUs to each thread, X = 0.5 * (TOTAL_CPUs_USED) / (TOTAL_NUMBER_OF_PROFILE_SETS) } else { # If the complementary file should not be scanned $hmmcpu = int($cpu / scalar(@profileSet)); # Assign X CPUs to each thread, X = (TOTAL_CPUs_USED) / (TOTAL_NUMBER_OF_PROFILE_SETS) } if ($hmmcpu < 1) { # If the number of CPUs per thread is smaller than 1 $hmmcpu = 1; # Give each thread at least one CPU to work on } ## Main strand searches... $cpuCount = 0; foreach $set (@profileSet) { # Go through each profile set to investigate if ($cpuCount < $cpu) { $cpuCount++; $pid = fork(); # Fork off a copy of this process for this set } else { $deceasedPID = wait(); # Wait until a PID is finished, and gather its number $pid = fork(); # Fork off a copy of this process for this set } if ($pid != 0) { # If this is the parent process push(@pids,$pid); # Add the new process ID to the list of active process IDs } else { # If this is the new child process $hmmPath = $profileDB . "/" . $set . ".hmm"; # Set the path to the HMM-file of the current set if ($search_eval ne "") { # If E-value cutoff is use for the search hmmerSearch("hmmscan --cpu $hmmcpu $heurMax -E $search_eval $hmmPath $tempDir/main.fasta","$tempDir/main.$set.hmmscan","M",$set); # Call HMMER with E-value cutoff } else { # If score cutoff is use for the search hmmerSearch("hmmscan --cpu $hmmcpu $heurMax -T $search_score $hmmPath $tempDir/main.fasta","$tempDir/main.$set.hmmscan","M",$set); # Call HMMER with score cutoff } ## Stop child process... exit; # Exits the child process } } ## Revese strand searches... if ($complement == 1) { # If the complementary file should be scanned foreach $set (@profileSet) { # Go through each profile set to investigate if ($cpuCount < $cpu) { $cpuCount++; $pid = fork(); # Fork off a copy of this process for this set } else { $deceasedPID = wait(); # Wait until a PID is finished, and gather its number $pid = fork(); # Fork off a copy of this process for this set } if ($pid != 0) { # If this is the parent process push(@pids,$pid); # Add the new process ID to the list of active process IDs } else { # If this is the new child process $hmmPath = $profileDB . "/" . $set . ".hmm"; # Set the path to the HMM-file of the current set if ($search_eval ne "") { # If E-value cutoff is use for the search hmmerSearch("hmmscan --cpu $hmmcpu $heurMax -E $search_eval $hmmPath $tempDir/complement.fasta","$tempDir/complement.$set.hmmscan","C",$set); # Call HMMER with E-value cutoff } else { # If score cutoff is use for the search hmmerSearch("hmmscan --cpu $hmmcpu $heurMax -T $search_score $hmmPath $tempDir/complement.fasta","$tempDir/complement.$set.hmmscan","C",$set); # Call HMMER with score cutoff } ## Stop child process... exit; # Exits the child process } } } ## Get the current time and output the active process IDs $now = localtime; #print STDERR " $now : Active PIDs: "; #foreach $p (@pids) { # Go through the list of PIDs # print STDERR "$p "; # Print the PID #} #print STDERR "\n"; # Print a new line do { # Loop until all child PIDs have finished. $deceasedPID = wait(); # Wait until a PID is finished, and gather its number $now = localtime; # Get the current time if ($deceasedPID > -1) { # If the PID that finished wasn't the last active one # print STDERR " $now : PID $deceasedPID finished.\n"; # Print finished PID } else { # If PID that finished was the last print STDERR " $now : All processes finished.\n"; # Print that all PIDs have finished } } until (wait() == -1); # Do this loop until all PIDs have finished $now = localtime; # Get current time print STDERR "$now : Parallel HMM-scan finished.\n"; # Print informative finishing message } } else { # If a pre-computed hmmscan file is supplied then ## Get the current time and output that the hmmscan step is skipped $now = localtime; if ($silent == 0) { print STDERR "$now : Skipping hmmscan! Using $hmmscan as input for the analysis instead.\n"; } } ## Analyse HMM-scan output ## Get the current time and output info $now = localtime; if ($silent == 0) { print STDERR "$now : Analysing results of HMM-scan (this might take quite some time)...\n"; } ## Set up output files if ($out_table == 1) { # If table output is on open (TABLE, ">$output.hmmer.table"); # Create a table output file } if ($out_graph == 1) { # If graphical output is on open (GRAPH, ">$output.graph"); # Create a graph output file } if ($out_not == 1) { # If not-found output is on open (NOTFOUND, ">$tempDir/$output\_hmmer_no_detections.txt"); # Create a HMMER not-found output file } $setI = 0; # Set the profile set indicator to zero foreach $set (@profileSet) { # Go through all the profile sets to be investigated for ($co = 0; $co <= 1; $co++) { # Do main (and complementary) strand analysis in order if ($co > 0) { # If main strand analysis is finished if ($complement == 1) { # If complementary strand should be analysed open (HMMOUTPUT, "$tempDir/complement.$set.hmmscan"); # Open hmmscan output for reading open (SEQUENCES, "$tempDir/complement.fasta"); # Open complementary sequence file for reading if ($out_table == 1) { # If table output is on, write a header for this set print TABLE "***********************************************************\n"; print TABLE "$set matches on complementary strand:\n"; } if ($out_graph == 1) { # If graphical output is on, write a header for this set print GRAPH "***********************************************************\n"; print GRAPH "$set matches on complementary strand:\n"; } } else { # If complementary strand should not be analysed last; # Exit this loop } } if ($co == 0) { # If main strand analysis is not finished open (HMMOUTPUT, "$tempDir/main.$set.hmmscan"); # Open hmmscan output for reading open (SEQUENCES, "$tempDir/main.fasta"); # Open main sequence file for reading if ($out_table == 1) { # If table output is on, write a header for this set print TABLE "***********************************************************\n"; print TABLE "$set matches on main strand:\n"; } if ($out_graph == 1) { # If graphical output is on, write a header for this set print GRAPH "***********************************************************\n"; print GRAPH "$set matches on main strand:\n"; } } ## Read and analyse hmmscan output file while ($line = <HMMOUTPUT>) { # Read in the hmmscan output file, line by line chomp($line); # Remove any potential line feeds if (substr($line,0,13) eq "## New query:") { # If this line begin with "## New query:", then this is a new entry undef %hits; # Empty the hits hash undef %evals; # Empty the e-value hash undef %scores; # Empty the score hash $querytemp = substr($line,14); # Extract everything from this line, except for the start ("## New query:") ($query,$length) = split('\t',$querytemp); # Split query name, length and DNA sequence if ($co == 0) { # If main strand analysis $DNA = $sequenceDB{"$query"}; } else { # If complementary strand $cseq = $sequenceDB{"$query"}; # Store the DNA sequence in the intermediate varaible $cseq $cseq =~ tr/[a-z]/[A-Z]/; # Make all letters uppercase $cseq =~ s/[^A-Z]//g; # Remove all non-alphabetic characters $cseq =~ tr/ACGTURYSWKMBDHVN/TGCAAYRWSMKVHDBN/; # Replace all characters with its complementary base $DNA = reverse($cseq); } } else { if ($line ne "//") { ## Find domain annotations... ($query,$matchProfile,$length,$domNo,$sign,$score,$bias,$cE,$iE,$hmmFrom,$hmmTo,$hmmends,$queryFrom,$queryTo,$queryends,$envFrom,$envTo,$envends,$acc) = split('\t',$line); # Split the line into a collection of stat variables $useQueryFrom = $queryFrom; if ($hmmFrom > 1) { # If the HMM-profile is not matched from the beginning $hmmDiff = $hmmFrom - 1; if ($useQueryFrom > $hmmDiff) { $useQueryFrom = $queryFrom - $hmmDiff; } else { $useQueryFrom = 1; } } if (uc($anchor) eq "HMM") { $anchorLen = $hmmTo - $hmmFrom + 1; } else { $anchorLen = $anchor; } $query_profile_match = $query . ":" . $matchProfile; $profileExists = 0; # Assume that the newly found match profile ($matchProfile) is not already found for this sequence if (exists($hits{$query_profile_match})) { # If a profile from the list is the same as the match profile ($hitFrom,$hitTo,$hitProfile,$hitScore,$hitE) = split('\t',$hits{$query_profile_match}); # Split the entry in list into stat variables if ($iE < $hitE) { # If the new match profile has a smaller E-value than the one from the list $hits{$query_profile_match} = "$useQueryFrom\t$envTo\t$matchProfile\t$score\t$iE\t$anchorLen"; # Replace the data in the hit list with the data for the newly found match profile $evals{$query_profile_match} = $iE; # Replace the E-value in the hit list with the E-value for the newly found match profile $scores{$query_profile_match} = $score; # Replace the score in the hit list with the score for the newly found match profile } $profileExists = 1; # Indicate that this match profile was found in the hit list } if ($profileExists == 0) { # If the match profile was not found in the hit list if (($iE <= $E) && ($score >= $S)) { # If this hits lives up to the minimal score and E-value cutoffs $hits{$query_profile_match} = "$useQueryFrom\t$envTo\t$matchProfile\t$score\t$iE\t$anchorLen"; # Add the data for the newly found match profile $evals{$query_profile_match} = $iE; # Add the E-value for the newly found match profile $scores{$query_profile_match} = $score; # Add the score for the newly found match profile } } } else { # If the line only contains "//", the end of this sequence's hmmscan entry is reached ## Save analysis results @sortedKeys = sort {$hits{$a} <=> $hits{$b}} keys(%hits); # Sort the the list of hits numerically ascending (smallest first) undef @sortedHits; undef @scores; undef @evals; foreach $key (@sortedKeys) { push(@sortedHits, $hits{$key}); # Add the hit to the list of hits numerically ascending (smallest first) push(@scores, $scores{$key}); # Add the score to the scores array push(@evals, $evals{$key}); # Add the E-value to the evals array } ## If the number of hits > N, the min eval < E and the max score > S then include query sequence ## OR if a single domain satisfies the thresholds and this is allowed, include it! if ( ((scalar(@sortedHits) >= $N) && (min(@evals) <= $E) && (max(@scores) >= $S)) || ((scalar(@sortedHits) > 0) && ($allow_single_E >= 0) && (min(@evals) <= $allow_single_E) && (max(@scores) >= $allow_single_score)) ) { if ($debug == 1) { # If debugging mode is on print STDERR $query . " :\t" . scalar(@sortedHits) . "\t" . min(@evals) . "\t" . max(@scores) . "\n"; # Print some top hit statistics } ## Save some total stats to be able to determine origin of ITS sequence if (scalar(@evals) > 0) { # If there are any E-values stored $averageE = sum(@evals) / scalar(@evals); # Calculate the average E-value for this profile set $averageScore = sum(@scores) / scalar(@scores); # Calculate the average score for this profile set $numberOfDomains = scalar(@sortedHits); # Calculate the number of domains matched on this sequence #$scoreSum = sum(@scores) / @modelCount[$setI]; # Calculate score sum as: sum / (no. of profiles of this given type) $scoreSum = sum(@scores) / 4; # Calculate score sum as: sum / (no. of profiles of this given type) $saveThis = "$query\t$set\t$co\t$numberOfDomains\t$averageE\t$averageScore\t$scoreSum\t$DNA\t"; # Collect the variables to save for this sequence and this profile set foreach $hit (@sortedHits) { # Go through the list of hits and add specific information to save from each hit ($hitFrom,$hitTo,$hitProfile,$hitScore,$hitE,$hitanchorlen) = split('\t',$hit); # Extract information from this hit $saveThis = $saveThis . "$hitFrom;$hitTo;$hitProfile;$hitScore;$hitE;$hitanchorlen\t"; # Add information to the list of variables to save } #push(@allHits, $saveThis); # Add this information to the collection of all hits for this sequence, across all profile sets if (exists($allHits{$query})) { $allHits{$query} = $allHits{$query} . "\n" . $saveThis; } else { $allHits{$query} = $saveThis; } } if ($out_table == 1) { # If table output is on print TABLE $query . "\t" . $length . "\t"; # Print query and length information to table foreach $hit (@sortedHits) { # Go through each hit in the hit list ($hitFrom,$hitTo,$hitProfile,$hitScore,$hitE,$hitanchorlen) = split('\t',$hit); # Extract data corresponding to this hit print TABLE "$hitFrom - $hitTo: $hitProfile ($hitScore, $hitE)\t"; # Print hit information to table } print TABLE "\n"; # Print new line } if ($out_graph == 1) { # If graphical output is on print GRAPH ">> " . $query . "\t" . $length . " bp\n"; # Print a sequence header $insertPoint = 0; # Set the domain insert point to beginning of line $hi = 0; # Set hit number to zero foreach $hit (@sortedHits) { # Go through the hit list ($hitFrom,$hitTo,$hitProfile,$hitScore,$hitE,$anchorLen) = split('\t',$hit); # Split the hit into stat variables if ($graph_scale == 0) { # If the graph scale is scaled individually to 100% for each sequence $pFrom = $hitFrom / $length * 100; # Set the profile start on graph relative to its position in the sequence $pTo = $hitTo / $length * 100; # Set the profile end on graph relative to its position in the sequence $pEnd = 100; # Set the end of the sequence graph to be at 100 } else { # If the scale is the same for all sequences $pFrom = $hitFrom * $graph_scale; # Set the profile start on graph scaled to the parameter given $pTo = $hitTo * $graph_scale; # Set the profile end on graph scaled to the parameter given $pEnd = $length * $graph_scale; # Set the end of the sequence graph to be at the end of the sequence scaled to the parameter given } for ($insertPoint = $insertPoint; $insertPoint <= $pFrom; $insertPoint++) { # Go forward through the sequence, moving the insert point one step at a time until the beginning of the next profile is reached print GRAPH "-"; # Print a "-" } print GRAPH substr($hitProfile,2,3); # When the profile is reached, print its name $insertPoint = $insertPoint + 3; # Move the insert point three steps forward, to account for the inserted name ($nextHitStart,$nextHitEnd,$nextProfile) = split('\t',@sortedHits[$hi + 1]); # Check where the next hit in the list is located if (($nextHitStart <= $hitTo) && ($nextHitStart > 0)) { # If the next hit in the list overlaps with this profile if ($graph_scale == 0) { # If the scale is relative $pTo = $nextHitStart / $length * 100 - 1; # Change the profile end on the graph to be where this next profile starts } else { # If the scale is the same for all sequences $pTo = $nextHitStart * $graph_scale - 1; # Change the profile end on the graph to be where this next profile starts } } for ($insertPoint = $insertPoint; $insertPoint <= $pTo; $insertPoint++) { # Go forward through the sequence, moving the insert point one step at a time until the end of the current profile is reached print GRAPH "="; # Print a "=" } if (($nextHitStart <= $hitTo) && ($nextHitStart > 0)) { # If the next hit in the list overlaps with this profile print GRAPH ">"; # Print a ">" to indicate the profile overlap $insertPoint++; # Move the insert point one additional step forward to account for the ">" inserted } $hi++; # Increase the hit number by one } for ($insertPoint = $insertPoint; $insertPoint <= $pEnd; $insertPoint++) { # If there is no more profile matches to sequence, go forward through the sequence, moving the insert point one step at a time until the end of the sequence is reached print GRAPH "-"; # Print a "-" } print GRAPH "\n"; # Print a new line, indicating the end of this sequence entry } } else { # If this sequence didn't find any good-enough profile matches if ($out_not == 1) { # If not-found output is on print NOTFOUND $query . "\n"; # Print the name of this query to the not-found list } } } } } close (SEQUENCES); # Close the input sequence file close (HMMOUTPUT); # Close the hmmscan output file } $setI++; # Add one to the profile set indicator } ## Close output files if ($out_table == 1) { # If table output is on, close the table file close (TABLE); } if ($out_graph == 1) { # If graphical output is on, close the graph file close (GRAPH); } if ($out_not == 1) { # If not-found output is on, close the not-found file close (NOTFOUND); #$profileCount = scalar(@profileSet); # Count the number of profile sets #if ($complement == 1) { # If complementary strand was scanned # $profileCount = $profileCount * 2; # Double the number of profile sets that was investigated (and thus the number of not-founds that could at max be found) #} #`sort $tempDir/hmmer_no_detections.txt | uniq -c | grep " *$profileCount " | sed "s/ *$profileCount //" > $output\_no_detections.txt`; # Sort the not-found list, count the number of profile sets having no matches for each query. Save those that have only non-matches to the hmmer-not-found file } ## Create total collected output and FASTA output if ($out_results == 1) { open (RESULTS, ">$output.extraction.results"); # Create a results file } open (RAWOUT, ">$tempDir/ITSx_output.raw"); # Create a raw output file for ALL data open (PROBLEM, ">$output.problematic.txt"); # Create a file for problematic entries $foundProblem = 0; if ($out_pos == 1) { open (POS, ">$output.positions.txt"); # Create a positions file } if ($out_fasta == 1) { # If FASTA output should be written open (FASTA, ">$output.full.fasta"); # Create a FASTA output file for found sequences if ($allow_reorder == 0) { # If reordering of domains is not allowed open (CHIMERA, ">$output.chimeric.fasta"); # Create a FASTA file for potential chimera sequences with profile matches in the wrong order $foundChimera = 0; } if ($out_partial > 0) { open (FULLPARTIAL, ">$output.full_and_partial.fasta"); } } if ($out_joined == 1) { # If SSU FASTA output should be written open (JOINED, ">$output.joined.fasta"); } if ($out_ssu == 1) { # If SSU FASTA output should be written open (SSU, ">$output.SSU.fasta"); } if ($out_lsu == 1) { # If LSU FASTA output should be written open (LSU, ">$output.LSU.fasta"); } if ($out_58S == 1) { # If 5.8S FASTA output should be written open (MID, ">$output.5_8S.fasta"); } if ($out_its1 == 1) { # If ITS1 FASTA output should be written open (ITS1, ">$output.ITS1.fasta"); if ($out_partial > 0) { open (ITS1PARTIAL, ">$output.ITS1.full_and_partial.fasta"); } } if ($out_its2 == 1) { # If ITS2 FASTA output should be written open (ITS2, ">$output.ITS2.fasta"); if ($out_partial > 0) { open (ITS2PARTIAL, ">$output.ITS2.full_and_partial.fasta"); } } if ($out_concat == 1) { # If concatenated output should be written open (CONCAT, ">$output.concat.fasta"); } undef @sortedHits; # Empty the array of sorted hits # @sortedHits = sort @allHits; # Sort the full list of hits in alphabetical order (to be able to analyse all sequences with same ID at once) if ($out_not == 1) { # If not-found output is on open (NOTFOUND, ">$output\_no_detections.txt"); # Create a not-found output file $noDetect = 0; } foreach $sequenceID (@sequenceOrder) { # Sort the full list of hits in their original order (to be able to analyse all sequences with same ID at once) if ($sequenceID ne "") { $countsInList = 0; if (exists($allHits{$sequenceID})) { @allHits = split('\n',$allHits{$sequenceID}); } else { undef @allHits; } foreach $line (@allHits) { @item = split('\t',$line); # Split the line into an array if (@item[0] eq $sequenceID) { # If this item corresponds to the current sequence ID push(@sortedHits,$line); # Add it to the sorted list of hits $countsInList++; } } if ($countsInList == 0) { # If no matches were found if ($out_not == 1) { # If not-found output is on print NOTFOUND "$sequenceID\n"; # Output the sequence ID $noDetect++; } } } } if ($out_not == 1) { # If not-found output is on close (NOTFOUND); if ($noDetect == 0) { `rm $output\_no_detections.txt 2> /dev/null`; } } ## Set all counts for different ITS types to zero undef @itsCounts; $itsChimeric = 0; $itsMain = 0; $itsCompl = 0; push(@sortedHits,"--END--"); # Add a last item to the sorted list, so that all items are securely saved $lc = 1; # Set the line count to one foreach $line (@sortedHits) { # Go through the list of found hits print RAWOUT "$line\n"; # Write the raw data associated with this hit to the raw data output file @item = split('\t',$line); # Split the line into an array ## If this sequence ID is the same as the saved ones, then add it, else empty the array sequence ID and save if ((@seqID[0] ne @item[0]) && (@item[0] ne "") || ($lc > scalar(@sortedHits))) { # If this sequence ID is not the same as the last one and is non-empty, or if the end of the list has been reached ## Save profile-type which is most likely... if ($priority eq "sum") { # If the sum-of-scores algorithm should be used to determine the most likely profile-type ## Reset variables to unrealisticly high or low values $best = 0; $bestCount = 0; $bestEval = 1000; $bestScore = -1000; $bestSum = -1000; for ($i = 0; $i < scalar(@seqScoreSum); $i++) { # Go through all sum-of-scores entries if (@seqScoreSum[$i] > $bestSum) { # If the current value is larger than the previous top value $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; $bestSum = @seqScoreSum[$i]; } if (@seqScoreSum[$i] == $bestSum) { # If the current value is equal to the previous top value if (@seqDomCounts[$i] > $bestCount) { # If the current domain count is larger than the previous top domain count $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; $bestSum = @seqScoreSum[$i]; } if (@seqDomCounts[$i] == $bestCount) { # If the current domain count is equal to the previous top domain count if (@seqAvgE[$i] < $bestEval) { # If the current E-value is smaller than the previous top E-value $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; $bestSum = @seqScoreSum[$i]; } } } } } if ($priority eq "domains") { # If the number of found domains should be used to determine the most likely profile-type ## Reset variables to unrealisticly high or low values $best = 0; $bestCount = 0; $bestEval = 1000; $bestScore = -1000; for ($i = 0; $i < scalar(@seqDomCounts); $i++) { # Go through all domain count entries if (@seqDomCounts[$i] > $bestCount) { # If the current domain count is larger than the previous top domain count $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } if (@seqDomCounts[$i] == $bestCount) { # If the current domain count is equal to the previous top domain count if (@seqAvgE[$i] < $bestEval) { # If the current E-value is smaller than the previous top E-value $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } if (@seqAvgE[$i] == $bestEval) { # If the current E-value is equal to the previous top E-value if (@seqAvgScore[$i] > $bestScore) { # If the current average score is larger than the previous top average score $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } } } } } if ($priority eq "eval") { # If the average E-value should be used to determine the most likely profile-type ## Reset variables to unrealisticly high or low values $best = 0; $bestCount = 0; $bestEval = 1000; $bestScore = -1000; for ($i = 0; $i < scalar(@seqDomCounts); $i++) { # Go through all domain counts entries if (@seqAvgE[$i] < $bestEval) { # If the current E-value is smaller than the previous top E-value $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } if (@seqAvgE[$i] == $bestEval) { # If the current E-value is equal to the previous top E-value if (@seqAvgScore[$i] > $bestScore) { # If the current average score is larger than the previous top average score $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } if (@seqAvgScore[$i] == $bestScore) { # If the current average score is equal to the previous top average score if (@seqDomCounts[$i] > $bestCount) { # If the current number of domains is larger than the previous top number of domains $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } } } } } if ($priority eq "score") { # If the average score should be used to determine the most likely profile-type ## Reset variables to unrealisticly high or low values $best = 0; $bestCount = 0; $bestEval = 1000; $bestScore = -1000; for ($i = 0; $i < scalar(@seqDomCounts); $i++) { # Go through all domain counts entries if (@seqAvgScore[$i] > $bestScore) { # If the current average score is larger than the previous top average score $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } if (@seqAvgScore[$i] == $bestScore) { # If the current average score is equal to the previous top average score if (@seqAvgE[$i] < $bestEval) { # If the current E-value is smaller than the previous top E-value $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } if (@seqAvgE[$i] == $bestEval) { # If the current E-value is equal to the previous top E-value if (@seqDomCounts[$i] > $bestCount) { # If the current number of domains is larger than the previous top number of domains $best = $i; # Set the best value to be the current value ## Set all other best variables to those corresponding to the current value $bestCount = @seqDomCounts[$i]; $bestEval = @seqAvgE[$i]; $bestScore = @seqAvgScore[$i]; } } } } } if (@seqID[$best] ne "") { # If the sequence ID of the most likely profile is not empty $allanchorLens = @anchorLens[$best]; # Get the best anchor lengths @allanchorLens = split(',', $allanchorLens); # Split the anchor lens into an array $chimeric = 0; # Assume the sequence is not chimeric if ($allow_reorder == 0) { # If re-order of domain is not allowed $domain_order = @allSeqDomains[$best]; # Gather the order the domains are found in @domain_order = split(' ',$domain_order); # Split the list into an array @sorted_domain_order = sort {$a cmp $b} @domain_order; # Sort the array alphabetically for ($di = 0; $di <= scalar(@domain_order); $di++) { # Go through the sorted array if ((@domain_order[$di] ne @sorted_domain_order[$di]) || (@problemCode[$best] =~ m/C/)) { # Check if the order of the arrays differ at any poiny $chimeric = 1; # If they do differ, mark the sequence as chimeric } } } $seqDNALength = length(@seqDNA[$best]); # Get the length of the DNA sequence ## Print sequence and match data... ## Order of columns in the output file: ## ID Length Type Main/Compl Domains Avg.Eval Avg.Score Start End Start_domain End_domain Chimeric if ($out_results == 1) { print RESULTS @seqID[$best] . "\t" . $seqDNALength . "\t" . @seqITSType[$best] . "\t" . @seqCompl[$best] . "\t" . @seqDomCounts[$best] . "\t" . @seqAvgE[$best] . "\t" . @seqAvgScore[$best] . "\t" . @seqScoreSum[$best] . "\t" . @dnaStart[$best] . "\t" . @dnaEnd[$best] . "\t" . @startDomain[$best] . "\t" . @endDomain[$best] . "\t"; # Print sequence and match data to the results file if ($chimeric == 1) { # If the sequence was regarded chimeric print RESULTS "Chimeric\t"; # Add a chimeric tag to the entry } else { # If not chimeric print RESULTS "\t"; # Add an empty column } $allDomains = @allSeqDomains[$best]; # Get the domain order of the entry $allDomains =~ tr/ /,/; # Replace spaces with commas in the domain order string $allDomains = substr($allDomains,0,length($allDomains) - 1); # Remove the last character (a comma) print RESULTS $allDomains; # Write the domain order to the results file print RESULTS "\t"; # Write a tab to the results file } if ($out_pos == 1) { $out_all_pos = 1; if ($out_all_pos == 1) { # Output positions of all domains $seqPartLen = @dnaEnd[$best] - @dnaStart[$best] + 1; if ($seqPartLen < 0) { $seqPartLen = $seqPartLen * -1; } ## Print the positions of all identified domains to the position file print POS @seqID[$best] . "\t" . $seqDNALength . " bp." . "\t"; if (@problemCode[$best] !~ m/S/) { print POS "SSU: " . @ssuStart[$best] . "-" . @ssuEnd[$best] . "\t"; } else { print POS "SSU: Not found\t"; } if ((@problemCode[$best] =~ m/X/) || ((@problemCode[$best] =~ m/[15]/) && (@problemCode[$best] =~ m/S/))) { print POS "ITS1: Not found\t"; } else { if (@problemCode[$best] =~ m/O/) { print POS "ITS1: " . (@ssuEnd[$best] + 1) . "-" . (@ssuEnd[$best] + 1) . "\t"; } else { print POS "ITS1: " . @its1Start[$best] . "-" . @its1End[$best] . "\t"; } } if (@problemCode[$best] !~ m/[125OP]/) { print POS "5.8S: " . @midStart[$best] . "-" . @midEnd[$best] . "\t"; } else { if (@problemCode[$best] =~ m/5/) { print POS "5.8S: Not found\t"; } else { if (@problemCode[$best] =~ m/1/) { print POS "5.8S: No start\t"; } else { if (@problemCode[$best] =~ m/O/) { print POS "5.8S: Overlap SSU\t"; } else { if (@problemCode[$best] =~ m/P/) { print POS "5.8S: Overlap LSU\t"; } else { print POS "5.8S: No end\t"; } } } } } if ((@problemCode[$best] =~ m/Y/) ||((@problemCode[$best] =~ m/[25]/) && (@problemCode[$best] =~ m/L/))) { print POS "ITS2: Not found\t"; } else { if (@problemCode[$best] =~ m/O/) { print POS "ITS2: " . (@lsuStart[$best] - 1) . "-" . (@lsuStart[$best] - 1) . "\t"; } else { print POS "ITS2: " . @its2Start[$best] . "-" . @its2End[$best] . "\t"; } } if (@problemCode[$best] !~ m/L/) { print POS "LSU: " . @lsuStart[$best] . "-" . @lsuEnd[$best] . "\t"; } else { print POS "LSU: Not found" . "\t"; } if (@problemCode[$best] =~ m/5/) { print POS "Broken or partial sequence, no 5.8S! "; } else { if (@problemCode[$best] =~ m/[12]/) { print POS "Broken or partial sequence, only partial 5.8S! "; } } if (@problemCode[$best] =~ m/B/) { print POS "ITS region too long! "; } if (@problemCode[$best] =~ m/O/) { print POS "5.8S seem to overlap with SSU! "; } if (@problemCode[$best] =~ m/P/) { print POS "5.8S seem to overlap with LSU! "; } if (@problemCode[$best] =~ m/C/) { print POS "Chimeric! "; } print POS "\n"; } else { # Output only ITS positions $seqPartLen = @dnaEnd[$best] - @dnaStart[$best] + 1; print POS @seqID[$best] . "\t" . $seqDNALength . " bp." . "\t" . "ITS1: " . @its1Start[$best] . "-" . @its1End[$best] . "\t" . "ITS2: " . @its2Start[$best] . "-" . @its2End[$best] . "\n"; # Print the positions of the ITS sequences to the position file } } if (@problem[$best] ne "") { $foundProblem++; print PROBLEM @seqID[$best] . "\t" . @problem[$best] . "\n"; } ## Set extended type string (the string going into the definition line of the FASTA file) $extendedType = $profileIndex{@seqITSType[$best]} . " ITS sequence"; @itsCounts[ord(@seqITSType[$best])]++; # Add one ITS to the appropriate counter if (@seqCompl[$best] == 1) { # If domains were found on complementary strand $extendedStrand = "complementary strand"; # Set the strand string to complementary $itsCompl++; # Add one to the complementary strand counter } else { # If domains were found on the main strand $extendedStrand = "main strand"; # Set the strand string to main $itsMain++; # Add one to the main strand counter } ## Print (extracted) ITS sequence... if ($out_fasta == 1) { # If FASTA-output is on if ($truncate == 0) { # If the whole sequence should be kept in output file if ($chimeric == 0) { # If the sequence is not chimeric if (@problemCode[$best] !~ m/[SL]/) { if ($out_preserve == 1) { # If sequence headers should be preserved print FASTA $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print FASTA ">" . @seqID[$best] . "|" . @seqITSType[$best] ." " . $extendedType . " (" . $seqDNALength . " bp) on " . $extendedStrand . "\n"; # Write FASTA definition line } print FASTA @seqDNA[$best] . "\n"; # Write DNA sequence } if ($out_partial > 0) { if (@problemCode[$best] !~ m/[SL125]/) { if ($out_preserve == 1) { # If sequence headers should be preserved print FULLPARTIAL $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print FULLPARTIAL ">" . @seqID[$best] . "|" . @seqITSType[$best] ." " . $extendedType . " (" . $seqDNALength . " bp) Full ITS region on " . $extendedStrand . "\n"; # Write FASTA defline } print FULLPARTIAL @seqDNA[$best] . "\n"; # Write DNA sequence } else { $its1PartLen = @its1End[$best] - @its1Start[$best] + 1; $its2PartLen = @its2End[$best] - @its2Start[$best] + 1; if (($out_partial < $its1PartLen) && ($out_partial < $its2PartLen)) { if ($out_preserve == 1) { # If sequence headers should be preserved print FULLPARTIAL $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print FULLPARTIAL ">" . @seqID[$best] . "|" . @seqITSType[$best] ." " . $extendedType . " (" . $seqDNALength . " bp) Partial ITS region on " . $extendedStrand . "\n"; # Write FASTA defline } print FULLPARTIAL @seqDNA[$best] . "\n"; # Write DNA sequence } } } } else { # If sequence is regarded chimeric $itsChimeric++; # Add one to the chimeric counter if ($out_preserve == 1) { # If sequence headers should be preserved print CHIMERA $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print CHIMERA ">" . @seqID[$best] . "|" . @seqITSType[$best] ." Chimeric " . $extendedType . " (" . length(@seqDNA[$best]) . " bp) on " . $extendedStrand . "\n"; # Write FASTA definition line } print CHIMERA @seqDNA[$best] . "\n"; # Write DNA sequence $foundChimera++; } } else { # If only the ITS part of the sequence should be saved to output file $fastaStartPoint = @dnaStart[$best] - 1; # Start extraction at the start of the first domain $fastaEndPoint = @dnaEnd[$best]+10; # End extraction 10 bp after the last domain if (@seqDomCounts[$best] > 1) { # If more than one domain was found if (substr(@startDomain[$best],0,5) eq "1_SSU") { # If the first domain was SSU $fastaStartPoint = @ssuEnd[$best]; # Set the start point of the extraction to the end of the SSU domain } if (substr(@startDomain[$best],0,5) eq "4_LSU") { # If the first domain was LSU $fastaStartPoint = @lsuEnd[$best]; # Set the start point of the extraction to the end of the LSU domain } if (substr(@endDomain[$best],0,5) eq "1_SSU") { # If the last domain was SSU $fastaEndPoint = @ssuStart[$best] - 1; # Set the end point of the extraction to the start of the SSU domain } if (substr(@endDomain[$best],0,5) eq "4_LSU") { # If the last domain was LSU $fastaEndPoint = @lsuStart[$best] - 1; # Set the end point of the extraction to the start of the LSU domain } } if ($fastaStartPoint < 0) { # If the start point is smaller than zero, set the start point to zero $fastaStartPoint = 0; } if ($fastaEndPoint > length(@seqDNA[$best])) { # If the end point is larger than the length of the sequence, set the end point to the sequence end $fastaEndPoint = length(@seqDNA[$best]); } $fastaLength = $fastaEndPoint - $fastaStartPoint + 1; if ($chimeric == 0) { # If the sequence is not chimeric if (@problemCode[$best] !~ m/[SL]/) { if ($out_preserve == 1) { # If sequence headers should be preserved print FASTA $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print FASTA ">" . @seqID[$best] . "|" . @seqITSType[$best] . " " . $extendedType . " (" . $fastaLength . " bp) on " . $extendedStrand . "\n"; # Write FASTA definition line } if ($anchorLen > 0) { if ($fastaStartPoint - @allanchorLens[0] > 0) { print FASTA substr(@seqDNA[$best],$fastaStartPoint - @allanchorLens[0],$fastaLength + @allanchorLens[0] + @allanchorLens[3]) . "\n"; # Write DNA sequence } else { print FASTA substr(@seqDNA[$best],0,$fastaLength + $fastaStartPoint + @allanchorLens[3]) . "\n"; # Write DNA sequence } } else { print FASTA substr(@seqDNA[$best],$fastaStartPoint,$fastaLength) . "\n"; # Write DNA sequence } } if ($out_partial > 0) { if (@problemCode[$best] !~ m/[SL125]/) { if ($out_preserve == 1) { # If sequence headers should be preserved print FULLPARTIAL $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print FULLPARTIAL ">" . @seqID[$best] . "|" . @seqITSType[$best] ." " . $extendedType . " (" . $seqDNALength . " bp) Full ITS region on " . $extendedStrand . "\n"; # Write FASTA defline } print FULLPARTIAL substr(@seqDNA[$best],$fastaStartPoint,$fastaLength) . "\n"; # Write DNA sequence } else { $its1PartLen = @its1End[$best] - @its1Start[$best] + 1; $its2PartLen = @its2End[$best] - @its2Start[$best] + 1; if (($out_partial < $its1PartLen) && ($out_partial < $its2PartLen)) { if ($out_preserve == 1) { # If sequence headers should be preserved print FULLPARTIAL $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print FULLPARTIAL ">" . @seqID[$best] . "|" . @seqITSType[$best] ." " . $extendedType . " (" . $seqDNALength . " bp) Partial ITS region on " . $extendedStrand . "\n"; # Write FASTA defline } print FULLPARTIAL substr(@seqDNA[$best],$fastaStartPoint,$fastaLength) . "\n"; # Write DNA sequence } } } } else { # If sequence is regarded chimeric $itsChimeric++; # Add one to the chimeric counter if ($out_preserve == 1) { # If sequence headers should be preserved print CHIMERA $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print CHIMERA ">" . @seqID[$best] . "|" . @seqITSType[$best] . " Chimeric " . $extendedType . " (" . $fastaLength . " bp) on " . $extendedStrand . "\n"; # Write FASTA definition line } print CHIMERA @seqDNA[$best] . "\n"; # Write DNA sequence $foundChimera++; } } } if ($out_joined == 1) { if ($chimeric == 0) { # If the sequence is not chimeric if ($out_preserve == 1) { # If sequence headers should be preserved print JOINED $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print JOINED ">" . @seqID[$best] . "|" . @seqITSType[$best] . " " . $extendedType . " (" . $fastaLength . " bp) From domain " . @startDomain[$best] . " to " . @endDomain[$best] . " on " . $extendedStrand . " Found domains: "; # Write FASTA definition line print JOINED substr(@allSeqDomains[$best],0,length(@allSeqDomains[$best]) - 1) . "\n"; # Write domain order } print JOINED substr(@seqDNA[$best],$fastaStartPoint,$fastaLength) . "\n"; # Write DNA sequence } else { # If sequence is regarded chimeric if ($out_preserve == 1) { # If sequence headers should be preserved print JOINED $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print JOINED ">" . @seqID[$best] . "|" . @seqITSType[$best] . " Chimeric " . $extendedType . " (" . $fastaLength . " bp) From domain " . @startDomain[$best] . " to " . @endDomain[$best] . " on " . $extendedStrand . " Found domains: "; # Write FASTA definition line print JOINED substr(@allSeqDomains[$best],0,length(@allSeqDomains[$best]) - 1) . "\n"; # Write domain order } print JOINED substr(@seqDNA[$best],$fastaStartPoint,$fastaLength) . "\n"; # Write DNA sequence } } ## Write SSU sequence to file if ($out_ssu == 1) { # If SSU output is on if (@problemCode[$best] !~ m/S/) { if ($only_full == 0) { $seqPartLen = @ssuEnd[$best] - @ssuStart[$best] + 1; if ($out_preserve == 1) { # If sequence headers should be preserved print SSU $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print SSU ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|SSU " . "Extracted SSU sequence (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@ssuStart[$best] - 1 - $anchorLen > 0) { print SSU substr(@seqDNA[$best], @ssuStart[$best] - 1 - $anchorLen, $anchorLen); } else { print SSU substr(@seqDNA[$best], 0, @ssuStart[$best] - 1); } print SSU substr(@seqDNA[$best], @ssuStart[$best] - 1, @ssuEnd[$best] - @ssuStart[$best] + 1); if (@ssuEnd[$best] - @ssuStart[$best] + 1 - $anchorLen > 0) { print SSU substr(@seqDNA[$best], @ssuEnd[$best], $anchorLen); } else { print SSU substr(@seqDNA[$best], @ssuEnd[$best]); } } else { print SSU substr(@seqDNA[$best], @ssuStart[$best] - 1, @ssuEnd[$best] - @ssuStart[$best] + 1); } print SSU "\n"; # Write DNA sequence } } } ## Write LSU sequence to file if ($out_lsu == 1) { # If LSU output is on if ($only_full == 0) { if (@problemCode[$best] !~ m/L/) { $seqPartLen = @lsuEnd[$best] - @lsuStart[$best] + 1; if ($out_preserve == 1) { # If sequence headers should be preserved print LSU $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print LSU ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|LSU " . "Extracted LSU sequence (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@lsuStart[$best] - 1 - $anchorLen > 0) { print LSU substr(@seqDNA[$best], @lsuStart[$best] - 1 - $anchorLen, $anchorLen); } else { print LSU substr(@seqDNA[$best], 0, @lsuStart[$best] - 1); } print LSU substr(@seqDNA[$best], @lsuStart[$best] - 1, @lsuEnd[$best] - @lsuStart[$best] + 1); #if (@lsuEnd[$best] - @lsuStart[$best] + 1 - $anchorLen > 0) { # print LSU substr(@seqDNA[$best], @lsuEnd[$best] - @lsuStart[$best] + 1, $anchorLen); #} else { # print LSU substr(@seqDNA[$best], @lsuEnd[$best] - @lsuStart[$best] + 1); #} } else { print LSU substr(@seqDNA[$best], @lsuStart[$best] - 1, @lsuEnd[$best] - @lsuStart[$best] + 1); } print LSU "\n"; # Write DNA sequence } } } ## Write 5.8S sequence to file if ($out_58S == 1) { # If 5.8S output is on if (@problemCode[$best] !~ m/[125]/) { $seqPartLen = @midEnd[$best] - @midStart[$best] + 1; if ($out_preserve == 1) { # If sequence headers should be preserved print MID $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print MID ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|5.8S " . "Extracted 5.8S sequence (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@midStart[$best] - 1 - $anchorLen > 0) { print MID substr(@seqDNA[$best], @midStart[$best] - 1 - $anchorLen, $anchorLen); } else { print MID substr(@seqDNA[$best], 0, @midStart[$best] - 1); } print MID substr(@seqDNA[$best], @midStart[$best] - 1, @midEnd[$best] - @midStart[$best] + 1); if (@midEnd[$best] - @midStart[$best] + 1 - $anchorLen > 0) { print MID substr(@seqDNA[$best], @midEnd[$best], $anchorLen); } else { print MID substr(@seqDNA[$best], @midEnd[$best]); } } else { print MID substr(@seqDNA[$best], @midStart[$best] - 1, @midEnd[$best] - @midStart[$best] + 1); } print MID "\n"; # Write DNA sequence } } ## Write ITS1 sequence to file if ($out_its1 == 1) { # If ITS1 output is on $seqPartLen = @its1End[$best] - @its1Start[$best] + 1; if ($seqPartLen > 1) { if ( (($only_full == 0) && (@problemCode[$best] !~ m/[15]/)) || (($only_full == 1) && (@problemCode[$best] !~ m/[S15]/))) { if ($out_preserve == 1) { # If sequence headers should be preserved print ITS1 $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print ITS1 ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS1 " . "Extracted ITS1 sequence " . @its1Start[$best] . "-" . @its1End[$best] . " (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@its1Start[$best] - 1 - $anchorLen > 0) { print ITS1 substr(@seqDNA[$best], @its1Start[$best] - 1 - $anchorLen, $anchorLen); } else { print ITS1 substr(@seqDNA[$best], 0, @its1Start[$best] - 1); } print ITS1 substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1); if (length(@seqDNA[$best]) - @its1End[$best] - $anchorLen > 0) { print ITS1 substr(@seqDNA[$best], @its1End[$best], $anchorLen); } else { print ITS1 substr(@seqDNA[$best], @its1End[$best]); } } else { if (@its1Start[$best] - 1 - @allanchorLens[0] > 0) { print ITS1 substr(@seqDNA[$best], @its1Start[$best] - 1 - @allanchorLens[0], @allanchorLens[0]); } else { print ITS1 substr(@seqDNA[$best], 0, @its1Start[$best] - 1); } print ITS1 substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1); if (length(@seqDNA[$best]) - @its1End[$best] - @allanchorLens[1] > 0) { print ITS1 substr(@seqDNA[$best], @its1End[$best], @allanchorLens[1]); } else { print ITS1 substr(@seqDNA[$best], @its1End[$best]); } } print ITS1 "\n"; # Write DNA sequence } } if ($out_partial > 0) { if (@problemCode[$best] !~ m/[S15]/) { if ($out_preserve == 1) { # If sequence headers should be preserved print ITS1PARTIAL $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print ITS1PARTIAL ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS1 " . "Extracted Full ITS1 sequence " . @its1Start[$best] . "-" . @its1End[$best] . " (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@its1Start[$best] - 1 - $anchorLen > 0) { print ITS1PARTIAL substr(@seqDNA[$best], @its1Start[$best] - 1 - $anchorLen, $anchorLen); } else { print ITS1PARTIAL substr(@seqDNA[$best], 0, @its1Start[$best] - 1); } print ITS1PARTIAL substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1); if (length(@seqDNA[$best]) - @its1End[$best] - $anchorLen > 0) { print ITS1PARTIAL substr(@seqDNA[$best], @its1End[$best], $anchorLen); } else { print ITS1PARTIAL substr(@seqDNA[$best], @its1End[$best]); } } else { if (@its1Start[$best] - 1 - @allanchorLens[0] > 0) { print ITS1PARTIAL substr(@seqDNA[$best], @its1Start[$best] - 1 - @allanchorLens[0], @allanchorLens[0]); } else { print ITS1PARTIAL substr(@seqDNA[$best], 0, @its1Start[$best] - 1); } print ITS1PARTIAL substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1); if (length(@seqDNA[$best]) - @its1End[$best] - @allanchorLens[1] > 0) { print ITS1PARTIAL substr(@seqDNA[$best], @its1End[$best], @allanchorLens[1]); } else { print ITS1PARTIAL substr(@seqDNA[$best], @its1End[$best]); } } print ITS1PARTIAL "\n"; # Write DNA sequence } else { if ($out_partial < $seqPartLen) { if ($out_preserve == 1) { # If sequence headers should be preserved print ITS1PARTIAL $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print ITS1PARTIAL ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS1 " . "Extracted Partial ITS1 sequence " . @its1Start[$best] . "-" . @its1End[$best] . " (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@its1Start[$best] - 1 - $anchorLen > 0) { print ITS1PARTIAL substr(@seqDNA[$best], @its1Start[$best] - 1 - $anchorLen, $anchorLen); } else { print ITS1PARTIAL substr(@seqDNA[$best], 0, @its1Start[$best] - 1); } print ITS1PARTIAL substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1); if (length(@seqDNA[$best]) - @its1End[$best] - $anchorLen > 0) { print ITS1PARTIAL substr(@seqDNA[$best], @its1End[$best], $anchorLen); } else { print ITS1PARTIAL substr(@seqDNA[$best], @its1End[$best]); } } else { if (@its1Start[$best] - 1 - @allanchorLens[0] > 0) { print ITS1PARTIAL substr(@seqDNA[$best], @its1Start[$best] - 1 - @allanchorLens[0], @allanchorLens[0]); } else { print ITS1PARTIAL substr(@seqDNA[$best], 0, @its1Start[$best] - 1); } print ITS1PARTIAL substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1); if (length(@seqDNA[$best]) - @its1End[$best] - @allanchorLens[1] > 0) { print ITS1PARTIAL substr(@seqDNA[$best], @its1End[$best], @allanchorLens[1]); } else { print ITS1PARTIAL substr(@seqDNA[$best], @its1End[$best]); } } print ITS1PARTIAL "\n"; # Write DNA sequence } } } } ## Write ITS2 sequence to file if ($out_its2 == 1) { # If ITS2 output is on $seqPartLen = @its2End[$best] - @its2Start[$best] + 1; if ($seqPartLen > 1) { if ( (($only_full == 0) && (@problemCode[$best] !~ m/[25]/)) || (($only_full == 1) && (@problemCode[$best] !~ m/[L25]/))) { if ($out_preserve == 1) { # If sequence headers should be preserved print ITS2 $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print ITS2 ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS2 " . "Extracted ITS2 sequence " . @its2Start[$best] . "-" . @its2End[$best] . " (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@its2Start[$best] - 1 - $anchorLen > 0) { print ITS2 substr(@seqDNA[$best], @its2Start[$best] - 1 - $anchorLen, $anchorLen); } else { print ITS2 substr(@seqDNA[$best], 0, @its2Start[$best] - 1); } print ITS2 substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1); if (length(@seqDNA[$best]) - @its2End[$best] - $anchorLen > 0) { print ITS2 substr(@seqDNA[$best], @its2End[$best], $anchorLen); } else { print ITS2 substr(@seqDNA[$best], @its2End[$best]); } } else { if (@its2Start[$best] - 1 - @allanchorLens[2] > 0) { print ITS2 substr(@seqDNA[$best], @its2Start[$best] - 1 - @allanchorLens[2], @allanchorLens[2]); } else { print ITS2 substr(@seqDNA[$best], 0, @its2Start[$best] - 1); } print ITS2 substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1); if (length(@seqDNA[$best]) - @its2End[$best] - @allanchorLens[3] > 0) { print ITS2 substr(@seqDNA[$best], @its2End[$best], @allanchorLens[3]); } else { print ITS2 substr(@seqDNA[$best], @its2End[$best]); } } print ITS2 "\n"; # Write DNA sequence } if ($out_partial > 0) { if (@problemCode[$best] !~ m/[L25]/) { if ($out_preserve == 1) { # If sequence headers should be preserved print ITS2PARTIAL $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print ITS2PARTIAL ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS2 " . "Extracted Full ITS2 sequence " . @its2Start[$best] . "-" . @its2End[$best] . " (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@its2Start[$best] - 1 - $anchorLen > 0) { print ITS2PARTIAL substr(@seqDNA[$best], @its2Start[$best] - 1 - $anchorLen, $anchorLen); } else { print ITS2PARTIAL substr(@seqDNA[$best], 0, @its2Start[$best] - 1); } print ITS2PARTIAL substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1); if (length(@seqDNA[$best]) - @its2End[$best] - $anchorLen > 0) { print ITS2PARTIAL substr(@seqDNA[$best], @its2End[$best], $anchorLen); } else { print ITS2PARTIAL substr(@seqDNA[$best], @its2End[$best]); } } else { if (@its2Start[$best] - 1 - @allanchorLens[2] > 0) { print ITS2PARTIAL substr(@seqDNA[$best], @its2Start[$best] - 1 - @allanchorLens[2], @allanchorLens[2]); } else { print ITS2PARTIAL substr(@seqDNA[$best], 0, @its2Start[$best] - 1); } print ITS2PARTIAL substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1); if (length(@seqDNA[$best]) - @its2End[$best] - @allanchorLens[3] > 0) { print ITS2PARTIAL substr(@seqDNA[$best], @its2End[$best], @allanchorLens[3]); } else { print ITS2PARTIAL substr(@seqDNA[$best], @its2End[$best]); } } print ITS2PARTIAL "\n"; # Write DNA sequence } else { if ($out_partial < $seqPartLen) { if ($out_preserve == 1) { # If sequence headers should be preserved print ITS2PARTIAL $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print ITS2PARTIAL ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS2 " . "Extracted Partial ITS2 sequence " . @its2Start[$best] . "-" . @its2End[$best] . " (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } if (uc($anchor) ne "HMM") { if (@its2Start[$best] - 1 - $anchorLen > 0) { print ITS2PARTIAL substr(@seqDNA[$best], @its2Start[$best] - 1 - $anchorLen, $anchorLen); } else { print ITS2PARTIAL substr(@seqDNA[$best], 0, @its2Start[$best] - 1); } print ITS2PARTIAL substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1); if (length(@seqDNA[$best]) - @its2End[$best] - $anchorLen > 0) { print ITS2PARTIAL substr(@seqDNA[$best], @its2End[$best], $anchorLen); } else { print ITS2PARTIAL substr(@seqDNA[$best], @its2End[$best]); } } else { if (@its2Start[$best] - 1 - @allanchorLens[2] > 0) { print ITS2PARTIAL substr(@seqDNA[$best], @its2Start[$best] - 1 - @allanchorLens[2], @allanchorLens[2]); } else { print ITS2PARTIAL substr(@seqDNA[$best], 0, @its2Start[$best] - 1); } print ITS2PARTIAL substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1); if (length(@seqDNA[$best]) - @its2End[$best] - @allanchorLens[3] > 0) { print ITS2PARTIAL substr(@seqDNA[$best], @its2End[$best], @allanchorLens[3]); } else { print ITS2PARTIAL substr(@seqDNA[$best], @its2End[$best]); } } print ITS2PARTIAL "\n"; # Write DNA sequence } } } } } ## Output concatenated ITS1 + ITS2 sequences if ($out_concat == 1) { $seqPartLen1 = @its1End[$best] - @its1Start[$best] + 1; $seqPartLen2 = @its2End[$best] - @its2Start[$best] + 1; if (($seqPartLen1 >= $concat_minlen) && ($seqPartLen2 >= $concat_minlen)) { if ($out_preserve == 1) { # If sequence headers should be preserved print CONCAT $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print CONCAT ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS1+2 " . "Concatenated ITS1 and ITS2 sequences (" . ($seqPartLen1 + $seqPartLen2) . " bp)\n"; # Write FASTA definition line } print CONCAT substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1); # Write ITS1 DNA sequence print CONCAT "-----"; # Write spacer print CONCAT substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1); # Write ITS2 DNA sequence print CONCAT "\n"; # Write newline } else { if ($seqPartLen1 >= $concat_minlen) { if ($out_preserve == 1) { # If sequence headers should be preserved print CONCAT $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print CONCAT ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS1 " . "ITS1 sequence (ITS2 too short) (" . $seqPartLen1 . " bp)\n"; # Write FASTA definition line } print CONCAT substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1); # Write ITS1 DNA sequence print CONCAT "-----"; # Write spacer print CONCAT "\n"; # Write newline } if ($seqPartLen2 >= $concat_minlen) { if ($out_preserve == 1) { # If sequence headers should be preserved print CONCAT $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print CONCAT ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS2 " . "ITS2 sequence (ITS1 too short) (" . $seqPartLen2 . " bp)\n"; # Write FASTA definition line } print CONCAT "-----"; # Write spacer print CONCAT substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1); # Write ITS2 DNA sequence print CONCAT "\n"; # Write newline } } } ## Print all sequences to the joined file for debugging if ($out_joined == 1) { $seqPartLen = @ssuEnd[$best] - @ssuStart[$best] + 1; if ($out_preserve == 1) { # If sequence headers should be preserved print JOINED $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print JOINED ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|SSU " . "Extracted SSU sequence (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } print JOINED substr(@seqDNA[$best], @ssuStart[$best] - 1, @ssuEnd[$best] - @ssuStart[$best] + 1) . "\n"; # Write DNA sequence $seqPartLen = @its1End[$best] - @its1Start[$best] + 1; if ($out_preserve == 1) { # If sequence headers should be preserved print JOINED $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print JOINED ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS1 " . "Extracted ITS1 sequence " . @its1Start[$best] . "-" . @its1End[$best] . " (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } print JOINED substr(@seqDNA[$best], @its1Start[$best] - 1, @its1End[$best] - @its1Start[$best] + 1) . "\n"; # Write DNA sequence $seqPartLen = @midEnd[$best] - @midStart[$best] + 1; if ($out_preserve == 1) { # If sequence headers should be preserved print JOINED $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print JOINED ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|5.8S " . "Extracted 5.8S sequence (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } print JOINED substr(@seqDNA[$best], @midStart[$best] - 1, @midEnd[$best] - @midStart[$best] + 1) . "\n"; # Write DNA sequence $seqPartLen = @its2End[$best] - @its2Start[$best] + 1; if ($out_preserve == 1) { # If sequence headers should be preserved print JOINED $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print JOINED ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|ITS2 " . "Extracted ITS2 sequence (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } print JOINED substr(@seqDNA[$best], @its2Start[$best] - 1, @its2End[$best] - @its2Start[$best] + 1) . "\n"; # Write DNA sequence $seqPartLen = @lsuEnd[$best] - @lsuStart[$best] + 1; if ($out_preserve == 1) { # If sequence headers should be preserved print JOINED $headers{@seqID[$best]} . "\n"; # Write FASTA definition line } else { print JOINED ">" . @seqID[$best] . "|" . @seqITSType[$best] . "|LSU " . "Extracted LSU sequence (" . $seqPartLen . " bp)\n"; # Write FASTA definition line } print JOINED substr(@seqDNA[$best], @lsuStart[$best] - 1, @lsuEnd[$best] - @lsuStart[$best] + 1) . "\n"; # Write DNA sequence } if ($out_results == 1) { ## Print info on all matches, also not top ones to the results file... for ($i = 0; $i < scalar(@seqITSType); $i++) { # Go through all the possible ITS types... print RESULTS @seqITSType[$i] . ": " . @seqDomCounts[$i] . " " . @seqAvgE[$i] . " " . @seqAvgScore[$i]; # Write some info on this type (Type, Domain count, Average E-value, Average score) if ($i < scalar(@seqITSType) - 1) { # If this is no the last domain type print RESULTS ", "; # Write a comma } } print RESULTS "\n"; # Write end of line } } ## Undefine all used arrays for the next round... undef @seqID; undef @seqITSType; undef @seqCompl; undef @seqDomCounts; undef @seqAvgE; undef @seqAvgScore; undef @seqScoreSum; undef @seqDNA; undef @allSeqDomains; undef @dnaStart; undef @dnaEnd; undef @ssuStart; undef @ssuEnd; undef @lsuStart; undef @lsuEnd; undef @midStart; undef @midEnd; undef @its1Start; undef @its1End; undef @its2Start; undef @its2End; undef @startDomain; undef @endDomain; undef @domain_order; undef @sorted_domain_order; undef @problem; undef @problemCode; undef @anchorLens; } if (@item[0] ne "") { # Add this entry to the set (regardless if the entry has the same ID as entries already in the set), as long as it is non-empty push(@seqID, @item[0]); # Add sequence ID push(@seqITSType, @item[1]); # Add ITS type push(@seqCompl, @item[2]); # Add main/complementary strand info push(@seqDomCounts, @item[3]); # Add domain count push(@seqAvgE, @item[4]); # Add average E-value push(@seqAvgScore, @item[5]); # Add average score push(@seqScoreSum, @item[6]); # Add sum-of-scores push(@seqDNA, @item[7]); # Add DNA sequence ## Determine first and last domains, and their positions ## Set variables to unrealistic values $dnaEnd = length(@item[7]); $startDomain = "***"; $endDomain = "***"; $dnaStart = 1; $allDomains = ""; $ssuStart = 1; $ssuEnd = 0; $midStart = 1; $midEnd = 0; $lsuStart = 1; $lsuEnd = 0; $its1Start = 1; $its1End = length(@item[7]); $its2Start = 1; $its2End = length(@item[7]); $problem = ""; $problemCode = ""; $ssuFound = 0; $lsuFound = 0; $midSFound = 0; $midEFound = 0; $order = ""; undef @hitanchorlens; for ($i = 8; $i < scalar(@item); $i++) { # Go through the list of found domains in this sequence ($hitStart,$hitEnd,$hitProfile,$hitScore,$hitEval,$hitanchorlen) = split(';',@item[$i]); # Separate the hit stats into variables $allDomains = $allDomains . $hitProfile . " "; # Add found domain to the list of all domains # if ($hitStart < $dnaStart) { # If this domain is the first one so far # $dnaStart = $hitStart; # Set the start of the ITS sequence to this domain's start # $startDomain = $hitProfile; # Set this domain as the starting domain # } # if ($hitEnd > $dnaEnd) { # If this domain is the last one so far # $dnaEnd = $hitEnd; # Set the end of the ITS sequence to this domain's end # $endDomain = $hitProfile; # Set this domain as the ending domain # } if (substr($hitProfile,0,5) eq "1_SSU" ) { # If this domain is the SSU's end $dnaStart = $hitStart; # Set the start of the ITS sequence to this domain's start $startDomain = $hitProfile; # Set this domain as the starting domain $ssuStart = 1; # Set the start of the SSU sequence to the start of the sequence $ssuEnd = $hitEnd; # Set the end of the SSU sequence to this domain's end $its1Start = $hitEnd + 1; # Set the start of the ITS1 sequence to right after this domain's end if ($midEFound == 0) { $its2Start = $hitEnd + 1; # Set the end of the ITS2 sequence to right after this domain's end } $ssuFound = 1; $order = $order . "1"; @hitanchorlens[1] = $hitanchorlen; } if (substr($hitProfile,0,5) eq "2_5.8" ) { # If this domain is the 5.8S's start if ($startDomain eq "***") { $dnaStart = 1; # Set the start of the ITS sequence to the start of the sequence $startDomain = $hitProfile; # Set this domain as the starting domain } $its1End = $hitStart - 1; # Set the end of the ITS1 sequence to right before this domain's start if ($midEFound == 0) { $its2Start = $hitEnd + 1; # Set the start of the ITS2 sequence to right after this domain's end } $midStart = $hitStart; # Set the start of the 5.8S sequence to this domain's start $midSFound = 1; $order = $order . "2"; @hitanchorlens[2] = $hitanchorlen; } if (substr($hitProfile,0,5) eq "3_End" ) { # If this domain is the 5.8S's end if ($startDomain eq "***") { $dnaStart = 1; # Set the start of the ITS sequence to the start of the sequence $startDomain = $hitProfile; # Set this domain as the starting domain } if ($midSFound == 0) { $its1End = $hitStart - 1; # Set the end of the ITS1 sequence to right before this domain's start } $its2Start = $hitEnd + 1; # Set the end of the ITS2 sequence to right after this domain's end $midEnd = $hitEnd; # Set the end of the 5.8S sequence to this domain's end $midEFound = 1; $order = $order . "3"; @hitanchorlens[3] = $hitanchorlen; } if (substr($hitProfile,0,5) eq "4_LSU" ) { # If this domain is the LSU's start if ($startDomain eq "***") { $dnaStart = 1; # Set the start of the ITS sequence to the start of the sequence $startDomain = $hitProfile; # Set this domain as the starting domain } $dnaEnd = $hitEnd; # Set the end of the ITS sequence to this domain's end $endDomain = $hitProfile; # Set this domain as the ending domain $lsuStart = $hitStart; # Set the start of the LSU sequence to the start of the this domain $lsuEnd = length(@item[7]); # Set the end of the LSU sequence to the end of the sequence $its2End = $hitStart - 1; # Set the end of the ITS2 sequence to right before this domain's start if ($midSFound == 0) { $its1End = $hitStart - 1; # Set the end of the ITS1 sequence to right before this domain's start } $lsuFound = 1; $order = $order . "4"; @hitanchorlens[4] = $hitanchorlen; } } if ($ssuFound == 0) { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "End of SSU sequence not found"; $problemCode = $problemCode . "S"; $order = "1" . $order; } if ($lsuFound == 0) { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "Start of LSU sequence not found"; $problemCode = $problemCode . "L"; $order = $order . "4"; } if (($midSFound == 0) && ($midEFound == 0)) { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "The 5.8S sequence was not found at all"; $problemCode = $problemCode . "5"; } else { if ($midSFound == 0) { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "Start of 5.8S sequence not found"; $problemCode = $problemCode . "1"; } if ($midEFound == 0) { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "End of 5.8S sequence not found"; $problemCode = $problemCode . "2"; } } if ($dnaEnd - $dnaStart > 1500) { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "ITS region is suspiciously long (> 1500 bp)"; $problemCode = $problemCode . "B"; } if (length($order) == 4) { if ($order ne "1234") { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "Domains found in wrong order, sequence may be chimeric"; $problemCode = $problemCode . "C"; } } else { if ((substr($order,0,1) ne "1") || (substr($order,-1) ne "4")) { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "Domains found in wrong order, sequence may be chimeric"; $problemCode = $problemCode . "C"; } else { if (($midSFound == 1) && ($midEFound == 1)) { if ($order !~ m/23/) { if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "Domains found in wrong order, sequence may be chimeric"; $problemCode = $problemCode . "C"; } } } } if ($its1End - $its1Start < 0) { # 5.8S overlaps SSU if (($midSFound == 1) && ($ssuFound == 1)) { $problemCode = $problemCode . "OC"; $its1Start = 0; $its1End = 0; if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "SSU seems to overlap 5.8S, sequence may be chimeric"; } else { $problemCode = $problemCode . "X"; $its1Start = 0; $its1End = 0; if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "No ITS1 sequence"; } } if ($its2End - $its2Start < 0) { # 5.8S overlaps LSU if (($midEFound == 1) && ($lsuFound == 1)) { $problemCode = $problemCode . "PC"; $its2Start = 0; $its2End = 0; if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "LSU seems to overlap 5.8S, sequence may be chimeric"; } else { $problemCode = $problemCode . "Y"; $its2Start = 0; $its2End = 0; if ($problem ne "") { $problem = $problem . "; "; } $problem = $problem . "No ITS2 sequence"; } } if (($problemCode =~ m/[15]/) && ($problemCode =~ m/[S]/)) { # Sequence lack all indications of ITS1 $its1Start = 0; $its1End = 0; } if (($problemCode =~ m/[25]/) && ($problemCode =~ m/[L]/)) { # Sequence lack all indications of ITS2 $its2Start = 0; $its2End = 0; } $allhitanchorlens = @hitanchorlens[1] . "," . @hitanchorlens[2] . "," . @hitanchorlens[3] . "," . @hitanchorlens[4]; push(@allSeqDomains,$allDomains); # Add list of all domains push(@dnaStart, $dnaStart); # Add start of the ITS sequence push(@startDomain, $startDomain); # Add start domain push(@dnaEnd, $dnaEnd); # Add end of ITS sequence push(@endDomain, $endDomain); # Add end domain push(@ssuStart, $ssuStart); # Add start of the SSU sequence push(@ssuEnd, $ssuEnd); # Add end of the SSU sequence push(@midStart, $midStart); # Add start of the 5.8S sequence push(@midEnd, $midEnd); # Add end of the 5.8S sequence push(@lsuStart, $lsuStart); # Add start of the LSU sequence push(@lsuEnd, $lsuEnd); # Add end of the LSU sequence push(@its1Start, $its1Start); # Add start of the ITS1 sequence push(@its1End, $its1End); # Add end of the ITS1 sequence push(@its2Start, $its2Start); # Add start of the ITS2 sequence push(@its2End, $its2End); # Add end of the ITS2 sequence push(@problem, $problem); # Add potential problem info push(@problemCode, $problemCode); # Add potential problem info in code form push(@anchorLens, $allhitanchorlens); # Add the anchor lengths } $lc++; # Increase the line count by one } ## Save results to the summary file if ($out_sum == 1) { # If summary file should be written $itsTotal = 0; # Reset the total ITS sum foreach $typeCount (@itsCounts) { # Add ITSs from all different origins $itsTotal += $typeCount; # Add the number of ITSs for this origin } ## Write info on the found ITS sequences to the summary file print SUMMARY "Sequences detected as ITS by ITSx:\t$itsTotal\n"; print SUMMARY " On main strand: \t$itsMain\n"; print SUMMARY " On complementary strand:\t$itsCompl\n"; if ($allow_reorder == 0) { # If re-ordering of domains is not allowed print SUMMARY "Sequences detected as chimeric by ITSx:\t$itsChimeric\n"; # Write the number of reported chimeric sequences to the summary file } ## Write info on the found ITS sequence types to the summary file print SUMMARY "ITS sequences by preliminary origin:\n"; print SUMMARY " Alveolates: \t" . int(@itsCounts[ord("A")]) . "\n"; print SUMMARY " Amoebozoa: \t" . int(@itsCounts[ord("D")]) . "\n"; print SUMMARY " Bacillariophyta: \t" . int(@itsCounts[ord("C")]) . "\n"; print SUMMARY " Brown algae: \t" . int(@itsCounts[ord("I")]) . "\n"; print SUMMARY " Bryophytes: \t" . int(@itsCounts[ord("B")]) . "\n"; print SUMMARY " Euglenozoa: \t" . int(@itsCounts[ord("E")]) . "\n"; print SUMMARY " Eustigmatophytes:\t" . int(@itsCounts[ord("U")]) . "\n"; print SUMMARY " Fungi: \t" . int(@itsCounts[ord("F")]) . "\n"; print SUMMARY " Green algae: \t" . int(@itsCounts[ord("G")]) . "\n"; print SUMMARY " Liverworts: \t" . int(@itsCounts[ord("L")]) . "\n"; print SUMMARY " Metazoa: \t" . int(@itsCounts[ord("M")]) . "\n"; print SUMMARY " Microsporidia: \t" . int(@itsCounts[ord("N")]) . "\n"; print SUMMARY " Oomycetes: \t" . int(@itsCounts[ord("O")]) . "\n"; print SUMMARY " Prymnesiophytes: \t" . int(@itsCounts[ord("P")]) . "\n"; print SUMMARY " Raphidophytes: \t" . int(@itsCounts[ord("Q")]) . "\n"; print SUMMARY " Red algae: \t" . int(@itsCounts[ord("H")]) . "\n"; print SUMMARY " Rhizaria: \t" . int(@itsCounts[ord("R")]) . "\n"; print SUMMARY " Synurophyceae: \t" . int(@itsCounts[ord("S")]) . "\n"; print SUMMARY " Tracheophyta: \t" . int(@itsCounts[ord("T")]) . "\n"; print SUMMARY "-----------------------------------------------------------------\n"; close (SUMMARY); # Close the summary file } if ($out_results == 1) { close (RESULTS); # Close the results file } close (RAWOUT); # Close the raw output file close (PROBLEM); # Close the file for problematic entries if ($foundProblem == 0) { `rm $output.problematic.txt 2> /dev/null`; } if ($out_pos == 1) { close (POS); # Close the positions file } if ($out_fasta == 1) { # If FASTA output is on close (FASTA); # Close the FASTA output file if ($allow_reorder == 0) { # If re-ordering of domains is not allowed close (CHIMERA); # Close the chimera file if ($foundChimera == 0) { `rm $output.chimeric.fasta 2> /dev/null`; } } if ($out_partial > 0) { close FULLPARTIAL; } } if ($out_ssu == 1) { close (SSU); # Close the SSU file } if ($out_lsu == 1) { close (LSU); # Close the LSU file } if ($out_58S == 1) { close (MID); # Close the 58S file } if ($out_its1 == 1) { close (ITS1); # Close the ITS1 file if ($out_partial > 0) { close ITS1PARTIAL; } } if ($out_its2 == 1) { close (ITS2); # Close the ITS2 file if ($out_partial > 0) { close ITS2PARTIAL; } } if ($out_concat == 1) { close CONCAT; } if ($out_not == 1) { # If not-found output is on open (NOTFOUND, "$output\_no_detections.txt"); # Open the not-found output file while ($line = <NOTFOUND>) { # Read all entries from file chomp($line); # Remove newline char push(@nodetectionlist,$line); # Add to non-detection list } close NOTFOUND; open (NOTFOUND, ">$output\_no_detections.fasta"); # Create a not-found FASTA output file foreach $seqID (@nodetectionlist) { # For all non-detections $seq = $sequenceDB{"$seqID"}; # Get sequence from sequence database print NOTFOUND ">$seqID\n"; # Print not found sequence ID print NOTFOUND $seq . "\n"; # Print not found sequence } close NOTFOUND; } ## Clean up and finish if ($pipeline == 0) { # If ITSx is not called from the pipeline mode (i.e. from ITSx) if ($save_raw == 1) { # If raw data should be saved `mv $tempDir $output\_ITSx_raw_output`; # Change the name of the temporary directory to ..._ITSx_raw_output } else { # Else, discard the raw data `rm -rf $tempDir`; # Remove the temporary directory } } ## Get the current time and output a finished message $now = localtime; if ($silent == 0) { print STDERR "$now : Extraction finished!\n"; print STDERR "-----------------------------------------------------------------\n"; print STDERR "Thank you for using ITSx!\n"; print STDERR "Please report bugs or unsupported lineages to itsx\@microbiology.se\n"; print STDERR "\n"; } ## Write end time a summary file if ($pipeline == 0) { # If not running in pipeline mode if ($out_sum == 1) { # If summary output is on open (SUMMARY, ">>$output.summary.txt"); # Append to the summary file print SUMMARY "ITSx run finished at $now.\n"; # Write ending time for the analysis close (SUMMARY); # Close summary file } } sub hmmerSearch { $hmmerCommand = $_[0]; $outputFile = $_[1]; $strand = $_[2]; $profileSet = $_[3]; open (HMMEROUTPUT, ">$outputFile"); open (HMMER, "$hmmerCommand |"); $totalHitCount = 0; $hitCount = 0; $SSUCount = 0; $LSUCount = 0; $startCount = 0; $endCount = 0; $maxCount = 1; while (chomp($line = <HMMER>)) { if (substr($line,0,6) eq "Query:") { $hitCount = 0; $SSUCount = 0; $LSUCount = 0; $startCount = 0; $endCount = 0; undef @bestScore; undef @bestEntry; $query = substr($line,7); $queryLength = $query; $queryLength =~ s/.* *//; $queryLength =~ s/[^0-9]//g; $query =~ s/ *//; $query =~ s/ *.*//; print HMMEROUTPUT "## New query:\t$query\t$queryLength\n"; } if (substr($line,0,12) eq "Description:") { $desc = $line; } if (substr($line,0,3) eq ">> ") { ($tempshit,$hmmerSubjectName) = split(' ',$line); } if ($line =~ m/[0-9] ! /) { $stats = $line; $stats =~ s/ */\t/g; ($empty,$no,$excl,$score) = split('\t',$stats); $hitCount++; $totalHitCount++; if ($maxCount == 0) { print HMMEROUTPUT "$query\t$hmmerSubjectName\t$queryLength$stats\n"; } else { if (substr($hmmerSubjectName,0,5) eq "1_SSU") { $SSUCount++; if ($SSUCount <= $maxCount) { @bestScore[1] = $score; @bestEntry[1] = "$query\t$hmmerSubjectName\t$queryLength$stats\n"; } else { if ($score > @bestScore[1]) { @bestScore[1] = $score; @bestEntry[1] = "$query\t$hmmerSubjectName\t$queryLength$stats\n"; } } } if (substr($hmmerSubjectName,0,5) eq "4_LSU") { $LSUCount++; if ($LSUCount <= $maxCount) { @bestScore[4] = $score; @bestEntry[4] = "$query\t$hmmerSubjectName\t$queryLength$stats\n"; } else { if ($score > @bestScore[4]) { @bestScore[4] = $score; @bestEntry[4] = "$query\t$hmmerSubjectName\t$queryLength$stats\n"; } } } if (substr($hmmerSubjectName,0,5) eq "2_5.8") { $startCount++; if ($startCount <= $maxCount) { @bestScore[2] = $score; @bestEntry[2] = "$query\t$hmmerSubjectName\t$queryLength$stats\n"; } else { if ($score > @bestScore[2]) { @bestScore[2] = $score; @bestEntry[2] = "$query\t$hmmerSubjectName\t$queryLength$stats\n"; } } } if (substr($hmmerSubjectName,0,5) eq "3_End") { $endCount++; if ($endCount <= $maxCount) { @bestScore[3] = $score; @bestEntry[3] = "$query\t$hmmerSubjectName\t$queryLength$stats\n"; } else { if ($score > @bestScore[3]) { @bestScore[3] = $score; @bestEntry[3] = "$query\t$hmmerSubjectName\t$queryLength$stats\n" } } } } } if (substr($line,0,2) eq "//") { if ($maxCount > 0) { if ($SSUCount > 0) { print HMMEROUTPUT @bestEntry[1]; } if ($startCount > 0) { print HMMEROUTPUT @bestEntry[2]; } if ($endCount > 0) { print HMMEROUTPUT @bestEntry[3]; } if ($LSUCount > 0) { print HMMEROUTPUT @bestEntry[4]; } } print HMMEROUTPUT "//\n"; } } close(HMMEROUTPUT); $now = localtime; if ($silent == 0) { if ($strand eq "M") { print STDERR " $now : " . ucfirst($profileIndex{$profileSet}) . " analysis of main strand finished.\n"; # Print finished type } else { print STDERR " $now : " . ucfirst($profileIndex{$profileSet}) . " analysis of complementary strand finished.\n"; # Print finished type } } } ## Please send beers, pizzas, cakes, fruit pies, job positions and other types of feedback to: ## johan.bengtsson [at] microbiology.se ## Looking forward to hearing from you.... visit my website: www.microbiology.se for info on my research ## //Johan Bengtsson, 2012-2014