| 1 |
#!/usr/bin/env perl |
| 2 |
|
| 3 |
################################################################ |
| 4 |
# copyright (c) 2014,2015 by William R. Pearson and The Rector & |
| 5 |
# Visitors of the University of Virginia */ |
| 6 |
################################################################ |
| 7 |
# Licensed under the Apache License, Version 2.0 (the "License"); |
| 8 |
# you may not use this file except in compliance with the License. |
| 9 |
# You may obtain a copy of the License at |
| 10 |
# |
| 11 |
# http://www.apache.org/licenses/LICENSE-2.0 |
| 12 |
# |
| 13 |
# Unless required by applicable law or agreed to in writing, |
| 14 |
# software distributed under this License is distributed on an "AS |
| 15 |
# IS" BASIS, WITHOUT WRRANTIES OR CONDITIONS OF ANY KIND, either |
| 16 |
# express or implied. See the License for the specific language |
| 17 |
# governing permissions and limitations under the License. |
| 18 |
################################################################ |
| 19 |
|
| 20 |
## modified 29-Sept-2016 to use EBI/proteins JSON URL: |
| 21 |
## https://www.ebi.ac.uk/proteins/api/features/p12345 |
| 22 |
|
| 23 |
# ann_feats_up_www2.pl gets an annotation file from fasta36 -V with a line of the form: |
| 24 |
|
| 25 |
# SP:GSTM1_HUMAN P09488 218 |
| 26 |
# |
| 27 |
# it must: |
| 28 |
# (1) read in the line |
| 29 |
# (2) parse it to get the up_acc |
| 30 |
# (3) return the tab delimited features |
| 31 |
# |
| 32 |
|
| 33 |
# this version can read feature2 uniprot features (acc/pos/end/label/value), but returns sorted start/end domains |
| 34 |
|
| 35 |
use warnings; |
| 36 |
use strict; |
| 37 |
|
| 38 |
use Getopt::Long; |
| 39 |
use Pod::Usage; |
| 40 |
use LWP::Simple; |
| 41 |
use LWP::UserAgent; |
| 42 |
use JSON qw(decode_json); |
| 43 |
|
| 44 |
## use IO::String; |
| 45 |
|
| 46 |
my $ua = LWP::UserAgent->new(ssl_opts=>{verify_hostname => 0}); |
| 47 |
my $up_base = 'https://www.ebi.ac.uk/proteins/api/features'; |
| 48 |
my $uniprot_suff = ".json"; |
| 49 |
|
| 50 |
my %domains = (); |
| 51 |
my $domain_cnt = 0; |
| 52 |
|
| 53 |
my $hostname = `/bin/hostname`; |
| 54 |
|
| 55 |
my ($sstr, $lav, $neg_doms, $no_doms, $no_feats, $no_vars, $no_over, $data_file, $shelp, $help) = (0,0,0,0,0,0,0,0,0,0); |
| 56 |
my ($min_nodom) = (10); |
| 57 |
|
| 58 |
my ($show_color) = (1); |
| 59 |
my $color_sep_str = " :"; |
| 60 |
$color_sep_str = '~'; |
| 61 |
|
| 62 |
GetOptions( |
| 63 |
"lav" => \$lav, |
| 64 |
"no-over" => \$no_over, |
| 65 |
"no_doms" => \$no_doms, |
| 66 |
"no-doms" => \$no_doms, |
| 67 |
"nodoms" => \$no_doms, |
| 68 |
"no_vars" => \$no_vars, |
| 69 |
"no-vars" => \$no_vars, |
| 70 |
"novars" => \$no_vars, |
| 71 |
"neg" => \$neg_doms, |
| 72 |
"neg_doms" => \$neg_doms, |
| 73 |
"neg-doms" => \$neg_doms, |
| 74 |
"negdoms" => \$neg_doms, |
| 75 |
"min_nodom=i" => \$min_nodom, |
| 76 |
"no_feats" => \$no_feats, |
| 77 |
"no-feats" => \$no_feats, |
| 78 |
"nofeats" => \$no_feats, |
| 79 |
"data:s" => \$data_file, |
| 80 |
"color!" => \$show_color, |
| 81 |
"sstr" => \$sstr, |
| 82 |
"h|?" => \$shelp, |
| 83 |
"help" => \$help, |
| 84 |
); |
| 85 |
|
| 86 |
pod2usage(1) if $shelp; |
| 87 |
pod2usage(exitstatus => 0, verbose => 2) if $help; |
| 88 |
pod2usage(1) unless @ARGV || $data_file || -p STDIN || -f STDIN; |
| 89 |
|
| 90 |
my @feat_keys = qw( ACT_SITE MOD_RES BINDING METAL SITE ); |
| 91 |
my @feat_vals = ( '=','*','#','^','@'); |
| 92 |
my @feat_names = ('Active site', 'Modified', 'Binding', 'Metal binding', 'Site'); |
| 93 |
|
| 94 |
unless ($no_vars) { |
| 95 |
push @feat_keys, qw(MUTAGEN VARIANT); |
| 96 |
push @feat_vals, ('V','V',); |
| 97 |
push @feat_names, ("","",); |
| 98 |
} |
| 99 |
|
| 100 |
my %feats_text = (); |
| 101 |
@feats_text{@feat_keys} = @feat_names; |
| 102 |
|
| 103 |
my %feats_label; |
| 104 |
@feats_label{@feat_keys} = @feat_names; |
| 105 |
|
| 106 |
my @dom_keys = qw( DOMAIN REPEAT ); |
| 107 |
my @dom_vals = ( [ '[', ']'],[ '[', ']']); |
| 108 |
|
| 109 |
my @ssr_keys = qw(STRAND HELIX); |
| 110 |
my @ssr_vals = ( [ '[', ']'], [ '[', ']']); |
| 111 |
|
| 112 |
my %annot_types = (); |
| 113 |
|
| 114 |
my $get_annot_sub = \&json_annots; |
| 115 |
if ($lav) { |
| 116 |
$no_feats = 1; |
| 117 |
} |
| 118 |
|
| 119 |
if ($sstr) { |
| 120 |
@annot_types{@ssr_keys} = @ssr_vals; |
| 121 |
} else { |
| 122 |
@annot_types{@feat_keys} = @feat_vals unless ($no_feats); |
| 123 |
@annot_types{@dom_keys} = @dom_vals unless ($no_doms); |
| 124 |
} |
| 125 |
|
| 126 |
if ($neg_doms) { |
| 127 |
$domains{'NODOM'}=0; |
| 128 |
} |
| 129 |
|
| 130 |
my ($tmp, $gi, $sdb, $acc, $id, $use_acc); |
| 131 |
|
| 132 |
unless ($no_feats || $sstr) { |
| 133 |
for my $i ( 0 .. $#feat_keys) { |
| 134 |
next unless $feats_label{$feat_keys[$i]}; |
| 135 |
print "=",$feat_vals[$i],":",$feats_label{$feat_keys[$i]},"\n"; |
| 136 |
} |
| 137 |
} |
| 138 |
|
| 139 |
# get the query |
| 140 |
my ($query, $seq_len) = @ARGV; |
| 141 |
$seq_len = 0 unless defined($seq_len); |
| 142 |
|
| 143 |
$query =~ s/^>// if ($query); |
| 144 |
|
| 145 |
my @annots = (); |
| 146 |
|
| 147 |
#if it's a file I can open, read and parse it |
| 148 |
|
| 149 |
unless ($data_file) { |
| 150 |
# unless ($query && ($query =~ m/[\|:]/ |
| 151 |
# || $query =~ m/^[OPQ][0-9][A-Z0-9]{3}[0-9]|[A-NR-Z][0-9]([A-Z][A-Z0-9]{2}[0-9]){1,2}\s/ |
| 152 |
# || $query =~ m/^(XN)(MP)_\d+/)) { |
| 153 |
if (! $query || -r $query) { |
| 154 |
while (my $a_line = <>) { |
| 155 |
$a_line =~ s/^>//; |
| 156 |
chomp $a_line; |
| 157 |
push @annots, lwp_annots($a_line, $get_annot_sub); |
| 158 |
} |
| 159 |
} else { |
| 160 |
push @annots, lwp_annots("$query\t$seq_len", $get_annot_sub); |
| 161 |
} |
| 162 |
} else { # just read the data from a file, give to $get_annot_sub(). |
| 163 |
my %annot_data = (seq_info => ">$data_file DATA"); |
| 164 |
|
| 165 |
open(DATA_IN, $data_file) || die "Cannot read $data_file"; |
| 166 |
|
| 167 |
my $lwp_data = ""; |
| 168 |
while (<DATA_IN>) { |
| 169 |
$lwp_data .= $_; |
| 170 |
} |
| 171 |
|
| 172 |
$annot_data{list} = $get_annot_sub->(\%annot_types, $lwp_data,0); |
| 173 |
|
| 174 |
push @annots, \%annot_data; |
| 175 |
} |
| 176 |
|
| 177 |
|
| 178 |
for my $seq_annot (@annots) { |
| 179 |
print ">",$seq_annot->{seq_info},"\n"; |
| 180 |
for my $annot (@{$seq_annot->{list}}) { |
| 181 |
if (!$lav && $show_color && defined($domains{$annot->[-1]})) { |
| 182 |
$annot->[-1] .= $color_sep_str.$domains{$annot->[-1]}; |
| 183 |
} |
| 184 |
print join("\t",@$annot),"\n"; |
| 185 |
} |
| 186 |
} |
| 187 |
|
| 188 |
exit(0); |
| 189 |
|
| 190 |
sub lwp_annots { |
| 191 |
my ($query_len, $get_annot_sub) = @_; |
| 192 |
|
| 193 |
my ($annot_line, $seq_len) = split(/\t/,$query_len); |
| 194 |
|
| 195 |
my %annot_data = (seq_info=>$annot_line); |
| 196 |
|
| 197 |
if ($annot_line =~ m/^gi\|/) { |
| 198 |
($tmp, $gi, $sdb, $acc, $id) = split(/\|/,$annot_line); |
| 199 |
} elsif ($annot_line =~ m/^(SP|TR):(\w+)\s(\w+)/) { |
| 200 |
($sdb, $id, $acc) = (lc($1), $2, $3); |
| 201 |
} elsif ($annot_line =~ m/^(SP|TR):(\w+)/) { |
| 202 |
($sdb, $id, $acc) = (lc($1), $2, ""); |
| 203 |
warn("*** $0 accession required: $annot_line\n"); |
| 204 |
} elsif ($annot_line =~ m/^(UR\d{3}:UniRef\d{2})_(\w+)/) { |
| 205 |
$sdb = lc($1); |
| 206 |
$id = $2; |
| 207 |
# $acc = $2; |
| 208 |
} elsif ($annot_line =~ m/\|/) { |
| 209 |
($sdb, $acc, $id) = split(/\|/,$annot_line); |
| 210 |
} |
| 211 |
else { |
| 212 |
($acc) = ($annot_line =~ m/^(\S+)/); |
| 213 |
} |
| 214 |
|
| 215 |
$acc =~ s/\.\d+// if ($acc); |
| 216 |
|
| 217 |
$annot_data{list} = []; |
| 218 |
my $lwp_features = ""; |
| 219 |
|
| 220 |
if ($acc && ($acc =~ m/^[A-Z][0-9][A-Z0-9]{3}[0-9]/)) { |
| 221 |
$lwp_features = get_https("$up_base/$acc.json"); |
| 222 |
} |
| 223 |
# elsif ($id && ($id =~ m/^\w+$/)) { |
| 224 |
# $lwp_features = get("$up_base/$id/$gff_post"); |
| 225 |
# } |
| 226 |
|
| 227 |
if ($lwp_features && ($lwp_features !~ /ERROR/)) { |
| 228 |
my $annot_json = decode_json($lwp_features); |
| 229 |
$annot_data{list} = $get_annot_sub->(\%annot_types, $annot_json, $seq_len); |
| 230 |
} |
| 231 |
|
| 232 |
return \%annot_data; |
| 233 |
} |
| 234 |
|
| 235 |
#### |
| 236 |
# parses www.ebi.ac.uk/uniprot/api json |
| 237 |
# |
| 238 |
sub json_annots { |
| 239 |
my ($annot_types, $json_ref, $seq_len) = @_; |
| 240 |
|
| 241 |
my ($acc, $pos, $end, $label, $value, $comment, $len); |
| 242 |
|
| 243 |
$seq_len = 0; |
| 244 |
|
| 245 |
my @feats2 = (); # features with start/stop, for checking overlap, adding negative |
| 246 |
my @sites = (); # sites with one position |
| 247 |
|
| 248 |
my ($seq_str, $seq_acc, $seq_id) = @{$json_ref}{qw(sequence accession entryName)}; |
| 249 |
$seq_len = length($seq_str); |
| 250 |
|
| 251 |
for my $feat ( @{$json_ref->{features}} ) { |
| 252 |
if ($annot_types->{$feat->{type}}) { |
| 253 |
|
| 254 |
my ($label, $pos, $end, $value) = @{$feat}{qw(type begin end description)}; |
| 255 |
|
| 256 |
$pos =~ s/[<>]//g; |
| 257 |
$end =~ s/[<>]//g; |
| 258 |
|
| 259 |
if ($label =~ m/DOMAIN/ || $label =~ m/REPEAT/) { |
| 260 |
$value = domain_name($label,$value); |
| 261 |
push @feats2, [$pos, "-", $end, $value]; |
| 262 |
} elsif ($label =~ m/HELIX/) { |
| 263 |
push @feats2, [$pos, "-", $end, $label]; |
| 264 |
} elsif ($label =~ m/STRAND/) { |
| 265 |
push @feats2, [$pos, "-", $end, $label]; |
| 266 |
} elsif ($label =~ m/VARIANT/ || $label =~ m/MUTAGEN/) { |
| 267 |
my $alt = $feat->{alternativeSequence}; |
| 268 |
if ($alt eq '') { $alt = 'X';} |
| 269 |
elsif (length($alt) > 1) { $alt=substr($alt,0,1);} |
| 270 |
push @sites, [$pos, $annot_types->{$label}, $alt, $value]; |
| 271 |
} |
| 272 |
else { |
| 273 |
next unless ($pos == $end); |
| 274 |
if ($feats_text{$label}) { |
| 275 |
my $info = $feats_text{$label}; |
| 276 |
if ($value) { |
| 277 |
$info .= ": $value"; |
| 278 |
} |
| 279 |
push @sites, [$pos, $annot_types->{$label}, "-", $info]; |
| 280 |
} else { |
| 281 |
push @sites, [$pos, $annot_types->{$label}, "-", $value]; |
| 282 |
} |
| 283 |
} |
| 284 |
} |
| 285 |
} |
| 286 |
|
| 287 |
@feats2 = sort { $a->[0] <=> $b->[0] } @feats2; |
| 288 |
|
| 289 |
if ($no_over) { |
| 290 |
# check for containment |
| 291 |
my $have_contained = 0; |
| 292 |
my $last_container = 0; |
| 293 |
for (my $i=1; $i < scalar(@feats2); $i++) { |
| 294 |
if ($feats2[$i]->[0] >= $feats2[$last_container]->[0] && $feats2[$i]->[2] <= $feats2[$last_container]->[2]) { |
| 295 |
$feats2[$i]->[1] = 'Delete'; |
| 296 |
$have_contained = 1; |
| 297 |
} else { |
| 298 |
$last_container=$i; |
| 299 |
} |
| 300 |
} |
| 301 |
|
| 302 |
if ($have_contained) { |
| 303 |
@feats2 = grep { $_->[1] !~ /Delete/ } @feats2; |
| 304 |
} |
| 305 |
|
| 306 |
# ensure that domains do not overlap |
| 307 |
for (my $i=1; $i < scalar(@feats2); $i++) { |
| 308 |
my $diff = $feats2[$i-1]->[2] - $feats2[$i]->[0]; |
| 309 |
if ($diff >= 0) { |
| 310 |
$feats2[$i-1]->[2] = $feats2[$i]->[0]+ int($diff/2); |
| 311 |
$feats2[$i]->[0] = $feats2[$i-1]->[2] + 1; |
| 312 |
} |
| 313 |
} |
| 314 |
} |
| 315 |
|
| 316 |
my @n_feats2 = (); |
| 317 |
|
| 318 |
if ($neg_doms) { |
| 319 |
my $last_end = 0; |
| 320 |
for my $feat ( @feats2 ) { |
| 321 |
if ($feat->[0] - $last_end > $min_nodom) { |
| 322 |
push @n_feats2, [$last_end+1, "-", $feat->[0]-1, "NODOM"]; |
| 323 |
} |
| 324 |
$last_end = $feat->[2]; |
| 325 |
} |
| 326 |
if ($seq_len - $last_end > $min_nodom) { |
| 327 |
push @n_feats2, [$last_end+1, "-", $seq_len, "NODOM"]; |
| 328 |
} |
| 329 |
} |
| 330 |
|
| 331 |
my @feats = (); |
| 332 |
for my $feat (@feats2, @n_feats2) { |
| 333 |
if (!$lav) { |
| 334 |
push @feats, [$feat->[0], '-', $feat->[2], $feat->[-1] ]; |
| 335 |
# push @feats, [$feat->[2], ']', '-', ""]; |
| 336 |
} |
| 337 |
else { |
| 338 |
push @feats, [$feat->[0], $feat->[2], $feat->[-1]]; |
| 339 |
} |
| 340 |
} |
| 341 |
|
| 342 |
@feats = sort { $a->[0] <=> $b->[0] } (@sites, @feats); |
| 343 |
|
| 344 |
return \@feats; |
| 345 |
} |
| 346 |
|
| 347 |
# domain name takes a uniprot domain label, removes comments ( ; |
| 348 |
# truncated) and numbers and returns a canonical form. Thus: |
| 349 |
# Cortactin 6. |
| 350 |
# Cortactin 7; truncated. |
| 351 |
# becomes "Cortactin" |
| 352 |
# |
| 353 |
|
| 354 |
sub domain_name { |
| 355 |
|
| 356 |
my ($label, $value) = @_; |
| 357 |
|
| 358 |
$value = 'UnDef' unless $value; |
| 359 |
|
| 360 |
$value =~ s/ /_/g; |
| 361 |
|
| 362 |
if ($label =~ /Domain|Repeat/i) { |
| 363 |
$value =~ s/;.*$//; |
| 364 |
$value =~ s/\.\s*$//; |
| 365 |
$value =~ s/\s+\d+$//; |
| 366 |
if (!defined($domains{$value})) { |
| 367 |
$domain_cnt++; |
| 368 |
$domains{$value} = $domain_cnt; |
| 369 |
} |
| 370 |
return $value; |
| 371 |
} |
| 372 |
else { |
| 373 |
return $value; |
| 374 |
} |
| 375 |
} |
| 376 |
|
| 377 |
sub get_https { |
| 378 |
my ($url) = @_; |
| 379 |
|
| 380 |
my $result = ""; |
| 381 |
my $response = $ua->get($url); |
| 382 |
|
| 383 |
if ($response->is_success) { |
| 384 |
$result = $response->decoded_content; |
| 385 |
} else { |
| 386 |
$result = ''; |
| 387 |
} |
| 388 |
return $result; |
| 389 |
} |
| 390 |
|
| 391 |
|
| 392 |
__END__ |
| 393 |
|
| 394 |
=pod |
| 395 |
|
| 396 |
=head1 NAME |
| 397 |
|
| 398 |
ann_feats_up_www2.pl |
| 399 |
|
| 400 |
=head1 SYNOPSIS |
| 401 |
|
| 402 |
ann_feats_up_www2.pl --no_doms --no_feats --lav 'sp|P09488|GSTM1_NUMAN' | accession.file |
| 403 |
|
| 404 |
=head1 OPTIONS |
| 405 |
|
| 406 |
-h short help |
| 407 |
--help include description |
| 408 |
--no-doms do not show domain boundaries (domains are always shown with --lav) |
| 409 |
--no-feats do not show feature (variants, active sites, phospho-sites) |
| 410 |
--lav produce lav2plt.pl annotation format, only show domains/repeats |
| 411 |
|
| 412 |
--neg-doms, -- report domains between annotated domains as NODOM |
| 413 |
(also --neg, --neg_doms) |
| 414 |
--min_nodom=10 -- minimum length between domains for NODOM |
| 415 |
|
| 416 |
--host, --user, --password, --port --db -- info for mysql database |
| 417 |
|
| 418 |
=head1 DESCRIPTION |
| 419 |
|
| 420 |
C<ann_feats_up_www2.pl> extracts feature, domain, and repeat |
| 421 |
information from the Uniprot DAS server through an XSLT transation |
| 422 |
provided by https://www.ebi.ac.uk/Tools/dbfetch/dbfetch/uniprotkb. |
| 423 |
This server provides GFF descriptions of Uniprot entries, with most of |
| 424 |
the information provided in UniProt feature tables. |
| 425 |
|
| 426 |
C<ann_feats_up_www2.pl> is an alternative to C<ann_pfam.pl> and |
| 427 |
C<ann_pfam.pl> that does not require a local MySQL copy of Pfam. |
| 428 |
|
| 429 |
Given a command line argument that contains a sequence accession |
| 430 |
(P09488), the program looks up the features available for that |
| 431 |
sequence and returns them in a tab-delimited format: |
| 432 |
|
| 433 |
>sp|P09488|GSTM1_HUMAN |
| 434 |
2 [ - GST N-terminal :1 |
| 435 |
7 V F Mutagen: Reduces catalytic activity 100- fold. |
| 436 |
23 * - MOD_RES: Phosphotyrosine (By similarity). |
| 437 |
33 * - MOD_RES: Phosphotyrosine (By similarity). |
| 438 |
34 * - MOD_RES: Phosphothreonine (By similarity). |
| 439 |
88 ] - |
| 440 |
90 [ - GST C-terminal :2 |
| 441 |
108 V Q Mutagen: Reduces catalytic activity by half. |
| 442 |
108 V S Mutagen: Changes the properties of the enzyme toward some substrates. |
| 443 |
109 V I Mutagen: Reduces catalytic activity by half. |
| 444 |
116 # - BINDING: Substrate. |
| 445 |
116 V A Mutagen: Reduces catalytic activity 10-fold. |
| 446 |
116 V F Mutagen: Slight increase of catalytic activity. |
| 447 |
173 V N in allele GSTM1B; dbSNP:rs1065411. |
| 448 |
208 ] - |
| 449 |
210 V T in dbSNP:rs449856. |
| 450 |
|
| 451 |
If features are provided, then a legend of feature symbols is provided |
| 452 |
as well: |
| 453 |
|
| 454 |
=*:phosphorylation |
| 455 |
==:active site |
| 456 |
=@:site |
| 457 |
=^:binding |
| 458 |
=!:metal binding |
| 459 |
|
| 460 |
If the C<--lav> option is specified, domain and repeat features are |
| 461 |
presented in a different format for the C<lav2plt.pl> program: |
| 462 |
|
| 463 |
>sp|P09488|GSTM1_HUMAN |
| 464 |
2 88 GST N-terminal. |
| 465 |
90 208 GST C-terminal. |
| 466 |
|
| 467 |
C<ann_feats_up_www2.pl> is designed to be used by the B<FASTA> programs with |
| 468 |
the C<-V \!ann_feats_up_www2.pl> option. It can also be used with the lav2plt.pl |
| 469 |
program with the C<--xA "\!ann_feats_up_www2.pl --lav"> or C<--yA "\!ann_feats_up_www2.pl --lav"> options. |
| 470 |
|
| 471 |
=head1 AUTHOR |
| 472 |
|
| 473 |
William R. Pearson, [email protected] |
| 474 |
|
| 475 |
=cut |