#include <SequenceProfile.hh>

Inheritance diagram for core::sequence::SequenceProfile:

Public Member Functions
	SequenceProfile ()
	ctors More...

	SequenceProfile (FileName const &fn)

	SequenceProfile (FileName const &fn, bool use_occurrence_data)

	SequenceProfile (utility::vector1< utility::vector1< core::Real > > prof, std::string const &sequence, std::string const &id, Size start=1, bool negative_better=false, bool use_occurrence_data=false)

SequenceOP	clone () const override
	Returns an owning pointer to a new SequenceProfile object, with data that is a deep copy of the information in this object. More...

void	read_from_file (FileName const &fn) override
	Read an profile matrix from the given filename using the NCBI PSSM format for a position-specific scoring matrix. More...

virtual void	generate_from_sequence (Sequence const &seq, std::string matrix="BLOSUM62")
	Generate the profile matrix from a sequence and a given substitution matrix. More...

void	rescale (core::Real factor=1)
	Multiply all profile weights by factor. More...

void	convert_profile_to_probs (core::Real temp=1.0)
	Use boltzman scaling on a per-residue basis to convert the current profile values to probabilities ( range 0.0-1.0 ) More...

void	global_auto_rescale ()
	Use linear rescaling (with a fixed zero) to fit the values within the range -1.0 to 1.0. More...

void	read_from_checkpoint (FileName const &fn, bool negative_better=true)
	Read profile matrix from the given filename using the NNMAKE .checkpoint format. For compatability, negative_better defaults to true. Set manually if necessary. More...

void	read_from_binary_chk (FileName const &fn)
	Read profile matrix from the given filename in the legacy BLAST binary format. More...

utility::vector1 < utility::vector1< Real > > const &	profile () const
	Returns the 2D vector1 of Real values representing this profile. More...

utility::vector1 < utility::vector1< Real > > const &	occurrence_data () const
	Returns the 2D vector1 of Real values of the probabilties of each aa. More...

void	profile (utility::vector1< utility::vector1< Real > > const &new_profile)
	Sets the 2D vector1 of Real values representing this profile. More...

void	occurrence_data (utility::vector1< utility::vector1< Real > > const &new_occurrence_data)
	Sets the 2D vector1 of Real values of the probabilties of each aa. More...

void	insert_char (core::Size pos, char new_char) override
	Inserts a character at the given position. More...

void	delete_position (core::Size pos) override
	Deletes the given position from the Sequence and shifts everything else back by one. More...

std::string	type () const override

Size	width () const
	Returns the number of distinct values at each position in this profile. More...

utility::vector1< Real > const &	prof_row (Size pos) const
	Returns the vector1 of values at this position. More...

utility::vector1< Real > const &	probability_row (Size pos) const
	Returns the vector1 of values at this position. More...

void	prof_row (utility::vector1< Real > const &new_prof_row, core::Size pos)
	Sets the 1D vector1 of Real values representing this profile at pos X. More...

void	probabilty_row (utility::vector1< Real > const &new_prob_row, core::Size pos)

Size	size () const

bool	negative_better () const
	Returns true if negative values are better identities. More...

void	negative_better (bool negbet)
	Set whether negative identities are better. More...

core::Real	temp () const
	returns the temperature used in computing profile probabilities More...

utility::vector1< std::string >	alphabet () const
	Return the alphabet used by this sequence profile. This is an N-dimensional vector1 where N is the width of the profile, and the ith entry of any row in the profile represents the probability of ith character at that row in the sequence. More...

void	alphabet (utility::vector1< std::string > new_alphabet)

bool	operator== (SequenceProfile const &other) const

Public Member Functions inherited from core::sequence::Sequence
	Sequence ()
	ctor More...

	Sequence (std::string const &seq, std::string const &id, core::Size start=1)

	Sequence (core::pose::Pose const &pose)

	Sequence (Sequence const &src)
	copy constructor. More...

	~Sequence () override
	dtor More...

virtual core::Size	length () const
	Returns the number of characters in this object. More...

virtual void	replace_char (core::Size pos, char new_char)

virtual std::string	to_string () const

void	sequence (std::string sequence)
	sets sequence to the given value. More...

void	id (std::string new_id)
	sets id to the given value. More...

void	start (core::Size new_start)
	sets starting index to the given value. More...

void	gap_char (char gap_char)
	sets gap_char to the given value. More...

void	spacer_positions (utility::vector1< Size > const &spacer_positions)
	sets spacer_positions to the given value. More...

core::Size	ungapped_length () const
	Returns the number of characters in this object, ignoring gaps. More...

core::Size	start () const
	Returns the start of this object. More...

std::string	id () const
	Returns the id of this object. More...

char	gap_char () const
	Returns the character used to represent a gap for this object. More...

utility::vector1< Size >	spacer_positions () const
	Returns the spacer positions (chain breaks, e.g., from spaces or commas in original sequence text) More...

std::string	ungapped_sequence () const
	Returns the string representing this sequence without gaps. More...

std::string	sequence () const
	Returns the full sequence, which may include gaps. More...

Sequence &	operator= (Sequence const &rhs)
	assignment operator. More...

bool	operator< (const Sequence &s) const
	Returns true if this Sequence object's id is lexicographically less than the given Sequence object's id, returns false otherwise. Uses C++ string < operator to compare this Sequence object's id() to the given Sequence object's id(). More...

bool	operator== (const Sequence &s) const
	Returns true if the given Sequence object is equal to this Sequence object. Tests for string equality of id(), start(), and sequence(), and returns false if any of these are not equal. More...

char	operator[] (core::Size pos) const
	Returns the character at the given sequence position. More...

char	at (core::Size pos) const

void	insert_gap (core::Size pos)
	Inserts a gap at the given position, where insert_gap( 0 ) inserts the character at the beginning of the sequence, and insert_gap( length() ) inserts the character at the end of the sequence. More...

void	append_char (char new_char)
	Append a character. More...

void	append_gap ()
	Append a gap. More...

bool	is_gap (core::Size pos) const
	Returns true if this position in the sequence represents a gap, returns false otherwise. More...

void	read_data (std::istream &in)
	Initializes the information in this sequence from the given std::istream. The istream should yield three pieces of information in the following order: More...

core::Size	resnum (core::Size idx) const
	Returns the index of the given sequence position, which is the position in the sequence minus any gaps that occur earlier in the sequence. For example, if the sequence is —AT, resnum(5) will return 2. Returns 0 for unaligned positions. More...

Private Types
typedef std::string	string

typedef utility::file::FileName	FileName

Private Member Functions
void	scores_to_probs_ (utility::vector1< core::Real > &scores, core::Real kT, bool negative_better=false) const
	converts a vector1 of arbitrary scores to values using Bolztmann averaging at a given kT. Scores should follow the convention that more positive -> better score. If not, set negative_better to true. More...

bool	check_internals_ () const
	Internal consistency check. Returns true if passed, causes a runtime_assertion failure if not. More...

Private Attributes
utility::vector1< std::string >	alphabet_

utility::vector1 < utility::vector1< Real > >	profile_

utility::vector1 < utility::vector1< Real > >	occurrence_data_

core::Real	temp_ = 1.0
	temp used to convert arbitrary scores to/from probabilities More...

bool	negative_better_ = false
	The orientation of the values. Are negative values better than zero/positive ones? More...

bool	use_occurrence_data_ = false
	Should only AAs for which we have occurrence data will be considered? More...

Friends
std::ostream &	operator<< (std::ostream &out, const SequenceProfile &p)
	Print this SequenceProfile object to the given std::ostream. More...

Member Typedef Documentation

typedef utility::file::FileName core::sequence::SequenceProfile::FileName

private

typedef std::string core::sequence::SequenceProfile::string

private

Constructor & Destructor Documentation

core::sequence::SequenceProfile::SequenceProfile ( )

ctors

References use_occurrence_data_.

Referenced by clone().

core::sequence::SequenceProfile::SequenceProfile ( FileName const & fn )

inline

References read_from_file().

core::sequence::SequenceProfile::SequenceProfile	(	FileName const &	fn,
		bool	use_occurrence_data
	)

inline

References read_from_file(), and use_occurrence_data_.

core::sequence::SequenceProfile::SequenceProfile	(	utility::vector1< utility::vector1< core::Real > >	prof,
		std::string const &	sequence,
		std::string const &	id,
		Size	start = `1`,
		bool	negative_better = `false`,
		bool	use_occurrence_data = `false`
	)

inline

References core::sequence::Sequence::length(), profile(), and size().

Member Function Documentation

utility::vector1< std::string > core::sequence::SequenceProfile::alphabet ( ) const

inline

Return the alphabet used by this sequence profile. This is an N-dimensional vector1 where N is the width of the profile, and the ith entry of any row in the profile represents the probability of ith character at that row in the sequence.

References alphabet_.

Referenced by protocols::indexed_structure_store::movers::SegmentSequenceProfileMover::to_sequence_profile(), and protocols::indexed_structure_store::movers::SegmentSequenceProfileMover::write_profile().

void core::sequence::SequenceProfile::alphabet ( utility::vector1< std::string > new_alphabet )

inline

References alphabet_.

bool core::sequence::SequenceProfile::check_internals_ ( ) const

private

Internal consistency check. Returns true if passed, causes a runtime_assertion failure if not.

References alphabet_, core::sequence::Sequence::length(), profile_, and size().

Referenced by width().

SequenceOP core::sequence::SequenceProfile::clone ( ) const

inlineoverridevirtual

Returns an owning pointer to a new SequenceProfile object, with data that is a deep copy of the information in this object.

Reimplemented from core::sequence::Sequence.

References SequenceProfile().

void core::sequence::SequenceProfile::convert_profile_to_probs ( core::Real temp = 1.0 )

Use boltzman scaling on a per-residue basis to convert the current profile values to probabilities ( range 0.0-1.0 )

References negative_better_, prof_row(), profile(), scores_to_probs_(), temp(), and temp_.

void core::sequence::SequenceProfile::delete_position ( core::Size pos )

overridevirtual

Deletes the given position from the Sequence and shifts everything else back by one.

Reimplemented from core::sequence::Sequence.

References core::sequence::Sequence::delete_position(), core::sequence::Sequence::length(), profile(), and core::sequence::Sequence::start().

Referenced by core::sequence::SequenceCoupling::delete_position().

void core::sequence::SequenceProfile::generate_from_sequence	(	Sequence const &	seq,
		std::string	matrix = `"BLOSUM62"`
	)

virtual

Generate the profile matrix from a sequence and a given substitution matrix.

References core::sequence::end, core::sequence::Sequence::length(), negative_better_, core::chemical::num_canonical_aas, prof_row(), profile(), core::sequence::MatrixScoringScheme::read_from_database(), core::sequence::Sequence::sequence(), and core::sequence::MatrixScoringScheme::values_for_aa().

void core::sequence::SequenceProfile::global_auto_rescale ( )

Use linear rescaling (with a fixed zero) to fit the values within the range -1.0 to 1.0.

References profile_, and rescale().

void core::sequence::SequenceProfile::insert_char	(	core::Size	pos,
		char	new_char
	)

overridevirtual

Inserts a character at the given position.

Reimplemented from core::sequence::Sequence.

References core::sequence::Sequence::insert_char(), core::sequence::Sequence::length(), profile(), and width().

Referenced by core::sequence::SequenceCoupling::insert_char().

bool core::sequence::SequenceProfile::negative_better ( ) const

inline

Returns true if negative values are better identities.

The "default" use case is for storing log likelihood values where positive is better. If you're using this class to store energy-like values, set negative_better to true.

References negative_better_.

Referenced by read_from_checkpoint().

void core::sequence::SequenceProfile::negative_better ( bool negbet )

inline

Set whether negative identities are better.

References negative_better_.

utility::vector1< utility::vector1< Real > > const & core::sequence::SequenceProfile::occurrence_data ( ) const

Returns the 2D vector1 of Real values of the probabilties of each aa.

References occurrence_data_.

void core::sequence::SequenceProfile::occurrence_data ( utility::vector1< utility::vector1< Real > > const & new_occurrence_data )

Sets the 2D vector1 of Real values of the probabilties of each aa.

bool core::sequence::SequenceProfile::operator== ( SequenceProfile const & other ) const

References alphabet_, negative_better_, occurrence_data_, and profile_.

utility::vector1< Real > const & core::sequence::SequenceProfile::probability_row ( Size pos ) const

Returns the vector1 of values at this position.

References occurrence_data_.

Referenced by read_from_file().

void core::sequence::SequenceProfile::probabilty_row	(	utility::vector1< Real > const &	new_prob_row,
		core::Size	pos
	)

References occurrence_data_.

utility::vector1< Real > const & core::sequence::SequenceProfile::prof_row ( Size pos ) const

Returns the vector1 of values at this position.

References profile_.

Referenced by convert_profile_to_probs(), generate_from_sequence(), core::sequence::operator<<(), read_from_binary_chk(), read_from_checkpoint(), core::sequence::ChemicalShiftSequence::read_from_file(), read_from_file(), and protocols::indexed_structure_store::movers::SegmentSequenceProfileMover::write_profile().

void core::sequence::SequenceProfile::prof_row	(	utility::vector1< Real > const &	new_prof_row,
		core::Size	pos
	)

Sets the 1D vector1 of Real values representing this profile at pos X.

References profile_.

utility::vector1< utility::vector1< Real > > const & core::sequence::SequenceProfile::profile ( ) const

Returns the 2D vector1 of Real values representing this profile.

Returns the 2D vector1 of Real-values representing this profile.

References profile_.

Referenced by core::sequence::ChemicalShiftSequence::check_internals_(), core::sequence::ChemicalShiftSequence::ChemicalShiftSequence(), convert_profile_to_probs(), delete_position(), core::sequence::SequenceCoupling::edgePotBetween(), generate_from_sequence(), insert_char(), read_from_checkpoint(), core::sequence::ChemicalShiftSequence::read_from_file(), rescale(), SequenceProfile(), and protocols::indexed_structure_store::movers::SegmentSequenceProfileMover::to_sequence_profile().

void core::sequence::SequenceProfile::profile ( utility::vector1< utility::vector1< Real > > const & new_profile )

Sets the 2D vector1 of Real values representing this profile.

void core::sequence::SequenceProfile::read_from_binary_chk ( FileName const & fn )

Read profile matrix from the given filename in the legacy BLAST binary format.

References core::chemical::aa_from_oneletter_code(), core::conformation::membrane::in, protocols::forge::methods::order(), prof_row(), profile_, and core::sequence::tr().

void core::sequence::SequenceProfile::read_from_checkpoint	(	FileName const &	fn,
		bool	negative_better = `true`
	)

Read profile matrix from the given filename using the NNMAKE .checkpoint format. For compatability, negative_better defaults to true. Set manually if necessary.

References protocols::cluster::calibur::aa, core::chemical::aa_from_oneletter_code(), alphabet_, negative_better(), negative_better_, core::chemical::oneletter_code_from_aa(), protocols::forge::methods::order(), prof_row(), profile(), core::sequence::Sequence::sequence(), size(), core::id::to_string(), and core::sequence::tr().

void core::sequence::SequenceProfile::read_from_file ( FileName const & fn )

overridevirtual

Read an profile matrix from the given filename using the NCBI PSSM format for a position-specific scoring matrix.

Reimplemented from core::sequence::Sequence.

References protocols::cluster::calibur::aa, core::chemical::aa_from_oneletter_code(), alphabet_, core::sequence::end, core::sequence::Sequence::id(), negative_better_, occurrence_data_, core::chemical::oneletter_code_from_aa(), protocols::forge::methods::order(), probability_row(), prof_row(), profile_, protocols::hybridization::score, core::sequence::Sequence::sequence(), core::id::to_string(), core::sequence::tr(), and use_occurrence_data_.

Referenced by SequenceProfile().

void core::sequence::SequenceProfile::rescale ( core::Real factor = 1 )

Multiply all profile weights by factor.

References negative_better_, profile(), profile_, and core::sequence::tr().

Referenced by global_auto_rescale().

void core::sequence::SequenceProfile::scores_to_probs_	(	utility::vector1< core::Real > &	scores,
		core::Real	kT,
		bool	negative_better = `false`
	)		const

private

converts a vector1 of arbitrary scores to values using Bolztmann averaging at a given kT. Scores should follow the convention that more positive -> better score. If not, set negative_better to true.

References protocols::hybridization::score.

Referenced by convert_profile_to_probs().

Size core::sequence::SequenceProfile::size ( ) const

inline

References profile_.

Referenced by core::sequence::ChemicalShiftSequence::check_internals_(), check_internals_(), core::sequence::ChemicalShiftSequence::ChemicalShiftSequence(), core::sequence::SequenceCoupling::edgePotBetween(), read_from_checkpoint(), and SequenceProfile().