Bio::Tools::Run::Phylo::Phast
PhastCons
Toolbar
Summary
Bio::Tools::Run::Phylo::Phast::PhastCons - Wrapper for footprinting using
phastCons
Package variables
No package variables defined.
Included modules
Bio::AlignIO
Bio::Annotation::SimpleValue
Bio::FeatureIO
Clone qw ( clone )
Cwd
File::Basename
Inherit
Synopsis
use Bio::Tools::Run::Phylo::Phast::PhastCons;
# Make a PhastCons factory
$factory = Bio::Tools::Run::Phylo::Phast::PhastCons->new();
# Pass the factory an alignment and the corresponding species tree
$align_filename = 't/data/apes.multi_fasta';
$species_tree_filename = 't/data/apes.newick';
@features = $factory->run($align_filename, $species_tree_filename);
# or get a Bio::Align::AlignI (SimpleAlign) object from somewhere, and
# generate the species tree automatically using a Bio::DB::Taxonomy database
$tdb = Bio::DB::Taxonomy->new(-source => 'entrez');
@features = $factory->run($aln_obj, $tdb);
# @features is an array of Bio::SeqFeature::Annotated, one feature per
# alignment sequence and prediction
Description
This is a wrapper for running the phastCons application by Adam Siepel. You
can get details here:
http://compgen.bscb.cornell.edu/~acs/software.html phastCons is used for phylogenetic footprinting/ shadowing.
Currently the interface is extremely simplified, allowing only one
analysis method. The focus here is on ease of use, allowing phastCons
to estimate as many parameters as possible and having it output just
the 'most conserved' blocks it detects. You can, however, try
supplying normal phastCons arguments to new(), or calling arg-named
methods (excluding initial hyphens and converting others to
underscores, eg. $factory->indels_only(1) to set the --indels-only
arg).
The particular analysis carried out here is to:
1. Use phyloFit to generate a tree model for initialization of the nonconserved
model from the supplied alignment (all data) and species tree
2. Run phastCons in 'training' mode for parameter estimation using all the
alignment data and the model from step 1
3. Run phastCons with the trees from step 2 to discover the most conserved
regions
See the 'HowTo' at
http://compgen.bscb.cornell.edu/~acs/phastCons-HOWTO.html for details on how to improve results.
WARNING: the API is likely to change in the future to allow for alternative
analysis types.
You will need to enable this phastCons wrapper to find the phast programs (at
least phastCons and phyloFit).
This can be done in (at least) three ways:
1. Make sure the phastCons and phyloFit executables are in your path.
2. Define an environmental variable PHASTDIR which is a
directory which contains the phastCons and phyloFit applications:
In bash:
export PHASTDIR=/home/username/phast/bin
In csh/tcsh:
setenv PHASTDIR /home/username/phast/bin
3. Include a definition of an environmental variable PHASTDIR in
every script that will use this PhastCons wrapper module, e.g.:
BEGIN { $ENV{PHASTDIR} = '/home/username/phast/bin' }
use Bio::Tools::Run::Phylo::Phast::PhastCons;
Methods
Methods description
Title : program_name Usage : $factory>program_name() Function: holds the program name Returns : string Args : None |
Title : program_dir Usage : $factory->program_dir(@params) Function: returns the program directory, obtained from ENV variable. Returns : string Args : None |
Title : new Usage : $factory = Bio::Tools::Run::Phylo::Phast::PhastCons->new(@params) Function: Creates a new PhastCons factory Returns : Bio::Tools::Run::Phylo::Phast::PhastCons Args : Optionally, provide any of the following (defaults are not to use, see the same-named methods for information on what each option does): { -target_coverage => number between 0 and 1 AND -expected_length => int } -rho => number between 0 and 1 -quiet => boolean (turn on or off program output to console)
Most other options understood by phastCons can be supplied as key =>
value pairs in this way. Options that don't normally take a value
should be given a value of 1. You can type the keys as you would on
the command line (eg. '--indels-only' => 1) or with only a single
hyphen to start and internal hyphens converted to underscores (eg.
-indels_only => 1) to avoid having to quote the key.
These options can NOT be used with this wrapper currently:
estimate_trees / T
estimate_rho / O
gc / G
msa_format / i
score / s
no_post_probs / n
seqname / N
idpref / P
help / h
alias / A
most_conserved / V / viterbi
refidx / r |
Title : target_coverage Usage : $factory->target_coverage(0.25); Function: Constrain transition parameters such that the expected fraction of sites in conserved elements is the supplied value. Returns : number (default undef) Args : None to get, number (between 0 and 1) to set |
Title : rho Usage : $factory->rho(0.3); Function: Set the *scale* (overall evolutionary rate) of the model for the conserved state to be the supplied number times that of the model for the non-conserved state (default 0.3). Returns : number (default undef) Args : None to get, number (between 0 and 1) to set |
Title : run Usage : $result = $factory->run($fasta_align_file, $newick_tree_file); -or- $result = $factory->run($align_object, $tree_object); -or- $result = $factory->run($align_object, $db_taxonomy_object); Function: Runs phastCons on an alignment to find the most conserved regions ('footprinting'). Returns : array of Bio::SeqFeature::Annotated (one feature per alignment sequence and prediction) Args : The first argument represents an alignment, the second argument a species tree. The alignment can be provided as a multi-fasta format alignment filename, or a Bio::Align::AlignI compliant object (eg. a Bio::SimpleAlign). The species tree can be provided as a newick format tree filename or a Bio::Tree::TreeI compliant object. Alternatively a Bio::DB::Taxonomy object can be supplied, in which case the species tree will be generated by using the alignment sequence names as species names and looking for those in the supplied database.
In all cases, the alignment sequence names must correspond to node
ids in the species tree. Multi-word species names should be joined
with underscores to form the sequence names, eg. Homo_sapiens |
Title : _setparams Usage : Internal function, not to be called directly Function: Creates a string of params to be used in the command string Returns : string of params Args : alignment file name for result production, AND filename of phyloFit generated init.mod file to estimate trees |
Methods code
sub program_name
{ return $PROGRAM_NAME; } |
sub program_dir
{ return $PROGRAM_DIR; } |
sub new
{ my ($class, @args) = @_;
my $self = $class->SUPER::new(@args);
$self->_set_from_args(\@args, -methods => {(map { $_ => $PARAMS{$_} } keys %PARAMS),
(map { $_ => $SWITCHES{$_} } keys %SWITCHES)},
-create => 1);
return $self;} |
sub target_coverage
{ my ($self, $num) = @_;
if (defined ($num)) {
($num > 0 && $num < 1) || $self->throw("target_coverage value must be between 0 and 1, exclusive");
$self->{coverage} = $num;
}
return $self->{coverage} || return;} |
sub rho
{ my ($self, $num) = @_;
if (defined ($num)) {
($num > 0 && $num < 1) || $self->throw("rho value must be between 0 and 1, exclusive");
$self->{rho} = $num;
}
return $self->{rho} || return;} |
sub run
{ my ($self, $aln, $tree) = @_;
($aln && $tree) || $self->throw("alignment and tree must be supplied");
my $aln_obj = $self->_alignment($aln);
$tree = $self->_tree($tree);
if (-e $aln) {
my $aln_id = basename($aln);
($aln_id) = $aln_id =~ /^([^\.]+)/;
$aln_obj->id($aln_id);
}
return $self->_run;} |
sub _run
{ my $self = shift;
my $exe = $self->executable || return;
my $pf = Bio::Tools::Run::Phylo::Phast::PhyloFit->new(-verbose => $self->verbose, -quiet => $self->quiet);
my $init_mod = $pf->run($self->_alignment, $self->_tree) || $self->throw("phyloFit failed to work as expected, is it installed?");
my $temp_dir = $self->tempdir;
my $cwd = Cwd->cwd();
chdir($temp_dir) || $self->throw("Couldn't change to temp dir '$temp_dir'");
my $aln_file = $self->_write_alignment;
my $command = $exe.$self->_setparams($aln_file, $init_mod);
$self->debug("phastCons training command = $command\n");
system($command) && $self->throw("phastCons training call ($command) crashed: $?");
$command = $exe.$self->_setparams($aln_file);
$self->debug("phastCons command = $command\n");
system($command) && $self->throw("phastCons call ($command) crashed: $?");
my $bedin = Bio::FeatureIO->new(-format => 'bed', -file => 'most_cons.bed');
chdir($cwd) || $self->throw("Couldn't change back to working directory '$cwd'");
my @feats = ();
my $aln = $self->_alignment;
while (my $feat = $bedin->next_feature) {
$feat->source_tag('phastCons');
my $sv = Bio::Annotation::SimpleValue->new(-tagname => 'predicted', -value => 1);
$feat->annotation->add_Annotation($sv);
foreach my $seq ($aln->each_seq) {
my $clone = clone($feat);
$clone->seq_id($seq->id);
my $sloc = $seq->location_from_column($feat->start + 1) || next;
my $eloc = $seq->location_from_column($feat->end + 1) || next;
$clone->start($sloc->start - 1);
$clone->end($eloc->end - 1);
push(@feats, $clone);
}
}
return @feats;} |
sub _setparams
{ my ($self, $aln_file, $init_mod) = @_;
my $param_string = $self->SUPER::_setparams(-params => [keys %PARAMS],
-switches => [keys %SWITCHES],
-double_dash => 1,
-underscore_to_dash => 1);
$param_string .= ' --no-post-probs';
my $aln_id = $self->_alignment->id;
$param_string .= " --seqname $aln_id --idpref $aln_id" if $aln_id;
$param_string .= ' --refidx 0';
my $input = ' --msa-format FASTA '.$aln_file;
if ($init_mod) {
$param_string .= ' --estimate-trees mytrees '.$input.' '.$init_mod;
}
else {
$param_string .= $input.' --most-conserved most_cons.bed --score mytrees.cons.mod,mytrees.noncons.mod';
}
return $param_string;
}
1;} |
General documentation
User feedback is an integral part of the evolution of this and other
Bioperl modules. Send your comments and suggestions preferably to
the Bioperl mailing list. Your participation is much appreciated.
bioperl-l@bioperl.org - General discussion
http://bioperl.org/wiki/Mailing_lists - About the mailing lists
Please direct usage questions or support issues to the mailing list:
bioperl-l@bioperl.org
rather than to the module maintainer directly. Many experienced and
reponsive experts will be able look at the problem and quickly
address it. Please include a thorough description of the problem
with code and data examples if at all possible.
The rest of the documentation details each of the object methods.
Internal methods are usually preceded with a _
Title : expected_length
Usage : $factory->expected_length(5);
Function: Set transition probabilities such that the expected length of a
conserved element is the supplied value. target_coverage() must also
be set.
Returns : int (default undef)
Args : None to get, int to set