A map alignment algorithm based on spectrum similarity (dynamic programming). More...

#include <OpenMS/ANALYSIS/MAPMATCHING/MapAlignmentAlgorithmSpectrumAlignment.h>

Inheritance diagram for MapAlignmentAlgorithmSpectrumAlignment:

Classes
class	Compare
	inner class necessary for using the sort algorithm. More...

Public Member Functions
	MapAlignmentAlgorithmSpectrumAlignment ()
	Default constructor. More...

virtual	~MapAlignmentAlgorithmSpectrumAlignment ()
	Destructor. More...

virtual void	align (std::vector< PeakMap > &, std::vector< TransformationDescription > &)
	Align peak maps. More...

Public Member Functions inherited from DefaultParamHandler
	DefaultParamHandler (const String &name)
	Constructor with name that is displayed in error messages. More...

	DefaultParamHandler (const DefaultParamHandler &rhs)
	Copy constructor. More...

virtual	~DefaultParamHandler ()
	Destructor. More...

virtual DefaultParamHandler &	operator= (const DefaultParamHandler &rhs)
	Assignment operator. More...

virtual bool	operator== (const DefaultParamHandler &rhs) const
	Equality operator. More...

void	setParameters (const Param &param)
	Sets the parameters. More...

const Param &	getParameters () const
	Non-mutable access to the parameters. More...

const Param &	getDefaults () const
	Non-mutable access to the default parameters. More...

const String &	getName () const
	Non-mutable access to the name. More...

void	setName (const String &name)
	Mutable access to the name. More...

const std::vector< String > &	getSubsections () const
	Non-mutable access to the registered subsections. More...

Public Member Functions inherited from ProgressLogger
	ProgressLogger ()
	Constructor. More...

	~ProgressLogger ()
	Destructor. More...

	ProgressLogger (const ProgressLogger &other)
	Copy constructor. More...

ProgressLogger &	operator= (const ProgressLogger &other)
	Assignment Operator. More...

void	setLogType (LogType type) const
	Sets the progress log that should be used. The default type is NONE! More...

LogType	getLogType () const
	Returns the type of progress log being used. More...

void	startProgress (SignedSize begin, SignedSize end, const String &label) const
	Initializes the progress display. More...

void	setProgress (SignedSize value) const
	Sets the current progress. More...

void	endProgress () const
	Ends the progress display. More...

Private Member Functions
	MapAlignmentAlgorithmSpectrumAlignment (const MapAlignmentAlgorithmSpectrumAlignment &)
	Copy constructor is not implemented -> private. More...

MapAlignmentAlgorithmSpectrumAlignment &	operator= (const MapAlignmentAlgorithmSpectrumAlignment &)
	Assignment operator is not implemented -> private. More...

void	prepareAlign_ (const std::vector< MSSpectrum *> &pattern, PeakMap &aligned, std::vector< TransformationDescription > &transformation)
	A function to prepare the sequence for the alignment. It calls intern the main function for the alignment. More...

void	msFilter_ (PeakMap &peakmap, std::vector< MSSpectrum *> &spectrum_pointer_container)
	filtered the MSLevel to gain only MSLevel 1 More...

bool	insideBand_ (Size i, Size j, Size n, Size m, Int k_)
	function for the test if cell i,j of the grid is inside the band More...

Int	bestk_ (const std::vector< MSSpectrum > &pattern, std::vector< MSSpectrum > &aligned, std::map< Size, std::map< Size, float > > &buffer, bool column_row_orientation, Size xbegin, Size xend, Size ybegin, Size yend)
	calculate the size of the band for the alignment for two given Sequence More...

float	scoreCalculation_ (Size i, Size j, Size patternbegin, Size alignbegin, const std::vector< MSSpectrum > &pattern, std::vector< MSSpectrum > &aligned, std::map< Size, std::map< Size, float > > &buffer, bool column_row_orientation)
	calculate the score of two given MSSpectra calls intern scoring_ More...

float	scoring_ (const MSSpectrum &a, MSSpectrum &b)
	return the score of two given MSSpectra by calling the scorefunction More...

void	affineGapalign_ (Size xbegin, Size ybegin, Size xend, Size yend, const std::vector< MSSpectrum > &pattern, std::vector< MSSpectrum > &aligned, std::vector< int > &xcoordinate, std::vector< float > &ycoordinate, std::vector< int > &xcoordinatepattern)
	affine gap cost Alignment More...

void	bucketFilter_ (const std::vector< MSSpectrum > &pattern, std::vector< MSSpectrum > &aligned, std::vector< Int > &xcoordinate, std::vector< float > &ycoordinate, std::vector< Int > &xcoordinatepattern)
	preparation function of data points to construct later the spline function. More...

void	debugFileCreator_ (const std::vector< MSSpectrum > &pattern, std::vector< MSSpectrum > &aligned)
	Creates files for the debugging. More...

void	debugscoreDistributionCalculation_ (float score)
	Rounding the score of two spectra, only necessary for debugging. More...

void	updateMembers_ ()
	This method is used to update extra member variables at the end of the setParameters() method. More...

Private Attributes
float	gap_
	Represent the gap cost for opening or closing a gap in the alignment. More...

float	e_
	Extension cost after a gap is open. More...

PeakSpectrumCompareFunctor *	c1_
	Pointer holds the scoring function, which can be selected. More...

float	cutoffScore_
	This is the minimal score to be count as a mismatch(range 0.0 - 1.0) More...

Size	bucketsize_
	Defines the size of one bucket. More...

Size	anchorPoints_
	Defines the amount of anchor points which are selected within one bucket. More...

bool	debug_
	Debug mode flag default: False. More...

float	mismatchscore_
	Represent the cost of a mismatch in the alignment. More...

float	threshold_
	This is the minimum score for counting as a match(1-cutoffScore_) More...

std::vector< std::vector< float > >	debugmatrix_
	Container holding the score of the matchmatrix and also the insertmatrix. More...

std::vector< std::vector< float > >	debugscorematrix_
	Container holding the only the score of Spectra. More...

std::vector< std::pair< float, float > >	debugtraceback_
	Container holding the path of the traceback. More...

std::vector< float >	scoredistribution_
	Container holding the score of each cell(matchmatrix,insertmatrix, traceback) More...

Additional Inherited Members
Public Types inherited from ProgressLogger
enum	LogType { CMD, GUI, NONE }
	Possible log types. More...

Protected Member Functions inherited from DefaultParamHandler
void	defaultsToParam_ ()
	Updates the parameters after the defaults have been set in the constructor. More...

Static Protected Member Functions inherited from ProgressLogger
static String	logTypeToFactoryName_ (LogType type)
	Return the name of the factory product used for this log type. More...

Protected Attributes inherited from DefaultParamHandler
Param	param_
	Container for current parameters. More...

Param	defaults_
	Container for default parameters. This member should be filled in the constructor of derived classes! More...

std::vector< String >	subsections_
	Container for registered subsections. This member should be filled in the constructor of derived classes! More...

String	error_name_
	Name that is displayed in error messages during the parameter checking. More...

bool	check_defaults_
	If this member is set to false no checking if parameters in done;. More...

bool	warn_empty_defaults_
	If this member is set to false no warning is emitted when defaults are empty;. More...

Protected Attributes inherited from ProgressLogger
LogType	type_

time_t	last_invoke_

ProgressLoggerImpl *	current_logger_

Static Protected Attributes inherited from ProgressLogger
static int	recursion_depth_

Detailed Description

A map alignment algorithm based on spectrum similarity (dynamic programming).

Parameters of this class are:

Name	Type	Default	Restrictions	Description
gapcost	float	1	min: 0	This Parameter stands for the cost of opining a gap in the Alignment. A gap means that one spectrum can not be aligned directly to another spectrum in the Map. This happens, when the similarity of both spectra a too low or even not present. Imagine it as a insert or delete of the spectrum in the map (similar to sequence alignment). The gap is necessary for aligning, if we open a gap there is a possibility that an another spectrum can be correct aligned with a higher score as before without gap. But to open a gap is a negative event and needs to carry a punishment, so a gap should only be opened if the benefits outweigh the downsides. The Parameter is to giving as a positive number, the implementation convert it to a negative number.
affinegapcost	float	0.5	min: 0	This Parameter controls the cost of extension a already open gap. The idea behind the affine gapcost lies under the assumption, that it is better to get a long distance of connected gaps than to have a structure of gaps interspersed with matches (gap match gap match etc.). Therefor the punishment for the extension of a gap generally should be lower than the normal gapcost. If the result of the alignment shows high compression, it is a good idea to lower either the affine gapcost or gap opening cost.
cutoff_score	float	0.7	min: 0 max: 1	The Parameter defines the threshold which filtered spectra, these spectra are high potential candidate for deciding the interval of a sub-alignment. Only those pair of spectra are selected, which has a score higher or same of the threshold.
bucketsize	int	100	min: 1	Defines the numbers of buckets. It is a quantize of the interval of those points, which defines the main alignment (match points). These points have to filtered, to reduce the amount of points for the calculating a smoother spline curve.
anchorpoints	int	100	min: 1 max: 100	Defines the percent of numbers of match points which a selected from one bucket. The high score pairs are previously selected. The reduction of match points helps to get a smoother spline curve.
debug	string	false	true, false	Activate the debug mode, there a files written starting with debug prefix.
mismatchscore	float	-5	max: 0	Defines the score of two spectra if they have no similarity to each other.
scorefunction	string	SteinScottImproveScore	SteinScottImproveScore, ZhangSimilarityScore	The score function is the core of an alignment. The success of an alignment depends mostly of the elected score function. The score function return the similarity of two spectra. The score influence defines later the way of possible traceback. There are multiple spectra similarity scores available..

Note:

If a section name is documented, the documentation is displayed as tooltip.
Advanced parameter names are italic.

Experimental classes:: This algorithm is work in progress and might change.

Constructor & Destructor Documentation

◆ MapAlignmentAlgorithmSpectrumAlignment() [1/2]

MapAlignmentAlgorithmSpectrumAlignment ( )

Default constructor.

◆ ~MapAlignmentAlgorithmSpectrumAlignment()

virtual ~MapAlignmentAlgorithmSpectrumAlignment ( )

virtual

Destructor.

◆ MapAlignmentAlgorithmSpectrumAlignment() [2/2]

MapAlignmentAlgorithmSpectrumAlignment ( const MapAlignmentAlgorithmSpectrumAlignment & )

private

Copy constructor is not implemented -> private.

Member Function Documentation

◆ affineGapalign_()

void affineGapalign_	(	Size	xbegin,
		Size	ybegin,
		Size	xend,
		Size	yend,
		const std::vector< MSSpectrum *> &	pattern,
		std::vector< MSSpectrum *> &	aligned,
		std::vector< int > &	xcoordinate,
		std::vector< float > &	ycoordinate,
		std::vector< int > &	xcoordinatepattern
	)

private

affine gap cost Alignment

This Alignment is based on the Needleman Wunsch Algorithm. To improve the time complexity a banded version was implemented, known as k - alignment. To save some space, the alignment is going to be calculated by position xbegin to xend of one sequence and ybegin and yend by another given sequence. The result of the alignment is stored in the second argument. The first sequence is used as a template for the alignment.

Parameters

xbegin	coordinate for the beginning of the template sequence.
ybegin	coordinate for the beginning of the aligned sequence .
xend	coordinate for the end of the template sequence.
yend	coordinate for the end of the aligned sequence.
pattern	template map.
aligned	map to be aligned.
xcoordinate	save the position of anchor points
ycoordinate	save the retentiontimes of an anchor points
xcoordinatepattern	save the reference position of the anchor points from the pattern

Exceptions

Exception::OutOfRange if a out of bound appear pattern or aligned

◆ align()

virtual void align	(	std::vector< PeakMap > &	,
		std::vector< TransformationDescription > &
	)

virtual

Align peak maps.

◆ bestk_()

Int bestk_	(	const std::vector< MSSpectrum *> &	pattern,
		std::vector< MSSpectrum *> &	aligned,
		std::map< Size, std::map< Size, float > > &	buffer,
		bool	column_row_orientation,
		Size	xbegin,
		Size	xend,
		Size	ybegin,
		Size	yend
	)

private

calculate the size of the band for the alignment for two given Sequence

This function calculates the size of the band for the alignment. It takes three samples from the aligned sequence and tries to find the highscore pairs (matching against the template sequence). The highscore pair with the worst distance is to be chosen as the size of k.

Parameters

pattern	vector of pointers of the template sequence
aligned	vector of pointers of the aligned sequence
buffer	holds the calculated score of index i,j.
column_row_orientation	indicate the order of the matrix
xbegin	indicate the beginning of the template sequence
xend	indicate the end of the template sequence
ybegin	indicate the beginning of the aligned sequence
yend	indicate the end of the aligned sequence

◆ bucketFilter_()

void bucketFilter_	(	const std::vector< MSSpectrum *> &	pattern,
		std::vector< MSSpectrum *> &	aligned,
		std::vector< Int > &	xcoordinate,
		std::vector< float > &	ycoordinate,
		std::vector< Int > &	xcoordinatepattern
	)

private

preparation function of data points to construct later the spline function.

This function reduced the amount of data values for the next step. The reduction is done by using a number of buckets, where the data points a selected. Within the buckets, only defined number a selected, to be written back as a data point. The selection within the buckets is done by scoring.

Parameters

pattern	template map.
aligned	map to be aligned.
xcoordinate	save the position of anchor points
ycoordinate	save the retention times of an anchor points
xcoordinatepattern	save the reference position of the anchor points from the pattern

◆ debugFileCreator_()

void debugFileCreator_	(	const std::vector< MSSpectrum *> &	pattern,
		std::vector< MSSpectrum *> &	aligned
	)

private

Creates files for the debugging.

This function is only active if the debug_ flag is true. The debugfileCreator creates following files:

debugtraceback.txt(gnuplotScript),
debugscoreheatmap.r and
debugRscript.

Debugscoreheatmap.r contains the scores of the Spectra to each other from the alignment and also the traceback. DebugRscript is the R script which reads those data. So both files are only working under R. Start R and type main(location of debugscoreheatmap.r). The output will be a heatmap of each sub-alignment. Debugtraceback.txt shows the way of the Traceback by using gnuplot.

Parameters

pattern	template map.
aligned	map to be aligned.

◆ debugscoreDistributionCalculation_()

void debugscoreDistributionCalculation_ ( float score )

private

Rounding the score of two spectra, only necessary for debugging.

This function rounded the score of two spectra. This is necessary for some function in the Debug-Mode

◆ insideBand_()

bool insideBand_	(	Size	i,
		Size	j,
		Size	n,
		Size	m,
		Int	k_
	)

private

function for the test if cell i,j of the grid is inside the band

The function returns true if the cell underlie these conditions: -k<=i-j<=k+n-m else return false.

Parameters

i	coordinate i
j	coordinate j
n	size of column
m	size of row
k_	size of k_

◆ msFilter_()

void msFilter_	(	PeakMap &	peakmap,
		std::vector< MSSpectrum *> &	spectrum_pointer_container
	)

private

filtered the MSLevel to gain only MSLevel 1

The alignment works only on MSLevel 1 data, so a filter has to be run.

Parameters

peakmap	map which has to be filtered
spectrum_pointer_container	output container, where pointers of the MSSpectrum are saved (only with MS level 1)

Exceptions

Exception::IllegalArgument is thrown if no spectra are contained in peakmap

◆ operator=()

MapAlignmentAlgorithmSpectrumAlignment& operator= ( const MapAlignmentAlgorithmSpectrumAlignment & )

private

Assignment operator is not implemented -> private.

◆ prepareAlign_()

void prepareAlign_	(	const std::vector< MSSpectrum *> &	pattern,
		PeakMap &	aligned,
		std::vector< TransformationDescription > &	transformation
	)

private

A function to prepare the sequence for the alignment. It calls intern the main function for the alignment.

This function takes two arguments. These argument types are two MSExperiments. The first argument should have been filtered, so that only the type of MSLevel 1 exists in the Sequence. The second argument doesn't have to fulfill this restriction. It's going to be filtered automatically. With these two arguments a pre-calculation is done to find some corresponding data points(maximum 4) for building alignment blocks. After the alignment a re-transformation is done, the new Retention Times appear in the original data.

The parameters are MSExperiments.

Parameters

pattern	template map.
aligned	map which has to be aligned.
transformation	container for rebuilding the alignment only by specific data-points

◆ scoreCalculation_()

float scoreCalculation_	(	Size	i,
		Size	j,
		Size	patternbegin,
		Size	alignbegin,
		const std::vector< MSSpectrum *> &	pattern,
		std::vector< MSSpectrum *> &	aligned,
		std::map< Size, std::map< Size, float > > &	buffer,
		bool	column_row_orientation
	)

private

calculate the score of two given MSSpectra calls intern scoring_

This function calculates the score from two MSSpectra. These two MSSpectra are chosen by the coordinates i,j. The two coordinates i,j indicate the index in the matrix. To find the right index on the sequence, each beginning is also given to the function. A flag indicates the labeling of the axes. The buffermatrix stores the result of the scoring. If the band expands only a lookup of known scores is done.

Parameters

i	is a index from the matrix.
j	is a index from the matrix.
patternbegin	indicate the beginning of the template sequence
alignbegin	indicate the beginning of the aligned sequence
pattern	vector of pointers of the template sequence
aligned	vector of pointers of the aligned sequence
buffer	holds the calculated score of index i,j.
column_row_orientation	indicate the order of the matrix

◆ scoring_()

float scoring_	(	const MSSpectrum &	a,
		MSSpectrum &	b
	)

private

return the score of two given MSSpectra by calling the scorefunction

◆ updateMembers_()

void updateMembers_ ( )

privatevirtual

This method is used to update extra member variables at the end of the setParameters() method.

Also call it at the end of the derived classes' copy constructor and assignment operator.

The default implementation is empty.

Reimplemented from DefaultParamHandler.

Member Data Documentation

◆ anchorPoints_

Size anchorPoints_

private

Defines the amount of anchor points which are selected within one bucket.

◆ bucketsize_

Size bucketsize_

private

Defines the size of one bucket.

◆ c1_

PeakSpectrumCompareFunctor* c1_

private

Pointer holds the scoring function, which can be selected.

◆ cutoffScore_

float cutoffScore_

private

This is the minimal score to be count as a mismatch(range 0.0 - 1.0)

◆ debug_

bool debug_

private

Debug mode flag default: False.

◆ debugmatrix_

std::vector<std::vector<float> > debugmatrix_

private

Container holding the score of the matchmatrix and also the insertmatrix.

◆ debugscorematrix_

std::vector<std::vector<float> > debugscorematrix_

private

Container holding the only the score of Spectra.

◆ debugtraceback_

std::vector<std::pair<float, float> > debugtraceback_

private

Container holding the path of the traceback.

◆ e_

float e_

private

Extension cost after a gap is open.

◆ gap_

float gap_

private

Represent the gap cost for opening or closing a gap in the alignment.

◆ mismatchscore_

float mismatchscore_

private

Represent the cost of a mismatch in the alignment.

◆ scoredistribution_

std::vector<float> scoredistribution_

private

Container holding the score of each cell(matchmatrix,insertmatrix, traceback)

◆ threshold_

float threshold_

private

This is the minimum score for counting as a match(1-cutoffScore_)

Classes

Public Member Functions

Private Member Functions

Private Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ MapAlignmentAlgorithmSpectrumAlignment() [1/2]

◆ ~MapAlignmentAlgorithmSpectrumAlignment()

◆ MapAlignmentAlgorithmSpectrumAlignment() [2/2]

Member Function Documentation

◆ affineGapalign_()

◆ align()

◆ bestk_()

◆ bucketFilter_()

◆ debugFileCreator_()

◆ debugscoreDistributionCalculation_()

◆ insideBand_()

◆ msFilter_()

◆ operator=()

◆ prepareAlign_()

◆ scoreCalculation_()

◆ scoring_()

◆ updateMembers_()

Member Data Documentation

◆ anchorPoints_

◆ bucketsize_

◆ c1_

◆ cutoffScore_

◆ debug_

◆ debugmatrix_

◆ debugscorematrix_

◆ debugtraceback_

◆ e_

◆ gap_

◆ mismatchscore_

◆ scoredistribution_

◆ threshold_