base class for filtering centroided and profile data for peak patterns More...

#include <OpenMS/FEATUREFINDER/MultiplexFiltering.h>

Inheritance diagram for MultiplexFiltering:

Classes
struct	BlacklistEntry
	an entry of blacklist_ (spectrum index and peak index in exp_centroided_) and a value read from or written to it More...

Public Types
typedef std::vector< std::map< int, int > >	White2Original
	index mapping from a 'white' experiment to its original experiment

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

Public Member Functions
	MultiplexFiltering (const MSExperiment &exp_centroided, const std::vector< MultiplexIsotopicPeakPattern > &patterns, int isotopes_per_peptide_min, int isotopes_per_peptide_max, double intensity_cutoff, double rt_band, double mz_tolerance, bool mz_tolerance_unit, double peptide_similarity, double averagine_similarity, double averagine_similarity_scaling, std::string averagine_type="peptide")
	constructor

MSExperiment &	getCentroidedExperiment ()
	returns the intensity-filtered, centroided spectral data

MSExperiment	getBlacklist ()
	returns the blacklisted, centroided peaks

Public Member Functions inherited from ProgressLogger
	ProgressLogger ()
	Constructor.

virtual	~ProgressLogger ()
	Destructor.

	ProgressLogger (const ProgressLogger &other)
	Copy constructor.

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

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

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

void	setLogger (ProgressLoggerImpl *logger)
	Sets the logger to be used for progress logging.

void	startProgress (SignedSize begin, SignedSize end, const std::string &label) const
	Initializes the progress display.

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

void	endProgress (UInt64 bytes_processed=0) const

void	nextProgress () const
	increment progress by 1 (according to range begin-end)

Protected Types
typedef std::function< std::optional< MultiplexFilteredPeak >(SignedSize, std::vector< BlacklistEntry > *)>	PeakFilter
	decides on one peak of a 'white' spectrum

Protected Member Functions
void	updateWhiteMSExperiment_ ()
	construct an MS experiment from exp_centroided_ containing peaks which have not been previously blacklisted in blacklist_

int	checkForSignificantPeak_ (double mz, double mz_tolerance, MSExperiment::ConstIterator &it_rt, double intensity_first_peak) const
	check for significant peak

bool	filterPeakPositions_ (double mz, const MSExperiment::ConstIterator &it_rt_begin, const MSExperiment::ConstIterator &it_rt_band_begin, const MSExperiment::ConstIterator &it_rt_band_end, const MultiplexIsotopicPeakPattern &pattern, MultiplexFilteredPeak &peak, std::vector< BlacklistEntry > *blacklist_reads=nullptr) const
	check if there are enough peaks in the RT band to form the pattern

void	blacklistPeak_ (const MultiplexFilteredPeak &peak)
	blacklist this peak

void	blacklistPeak_ (const MultiplexFilteredPeak &peak, unsigned pattern_idx)
	blacklist this peak

void	blacklistEntries_ (const MultiplexFilteredPeak &peak, unsigned pattern_idx, std::vector< BlacklistEntry > &entries) const
	the entries of blacklist_ that blacklistPeak_() sets, with their new values, in the order it sets them

void	filterSpectrum_ (size_t idx_rt, unsigned pattern_idx, const PeakFilter &filter_peak, MultiplexFilteredMSExperiment &result)
	filter all peaks of one 'white' spectrum, add the ones that pass to `result` and blacklist them

bool	filterAveragineModel_ (const MultiplexIsotopicPeakPattern &pattern, const MultiplexFilteredPeak &peak) const
	check if the satellite peaks conform with the averagine model

bool	filterPeptideCorrelation_ (const MultiplexIsotopicPeakPattern &pattern, const MultiplexFilteredPeak &peak) const
	check if corresponding satellite peaks of different peptides show a good correlation

Protected Attributes
MSExperiment	exp_centroided_
	centroided experimental data

std::vector< std::vector< int > >	blacklist_
	auxiliary structs for blacklisting

std::vector< std::vector< BlacklistEntry > >	blacklist_reads_
	for each peak of the spectrum being filtered, the entries of blacklist_ its decision read (see filterSpectrum_())

std::vector< std::vector< BlacklistEntry > >	blacklist_writes_
	for each peak of the spectrum being filtered that passes, the entries of blacklist_ it sets (see filterSpectrum_())

MSExperiment	exp_centroided_white_
	"white" centroided experimental data

White2Original	exp_centroided_mapping_
	mapping of peak indices from a 'white' experiment exp_centroided_white_ to its original experiment exp_centroided_

std::vector< MultiplexIsotopicPeakPattern >	patterns_
	list of peak patterns

size_t	isotopes_per_peptide_min_
	minimum number of isotopic peaks per peptide

size_t	isotopes_per_peptide_max_
	maximum number of isotopic peaks per peptide

double	intensity_cutoff_
	intensity cutoff

double	rt_band_
	RT range used for filtering.

double	mz_tolerance_
	m/z shift tolerance

bool	mz_tolerance_unit_in_ppm_
	unit for m/z shift tolerance (ppm - true, Da - false)

double	peptide_similarity_
	peptide similarity

double	averagine_similarity_
	averagine similarity

double	averagine_similarity_scaling_
	averagine similarity scaling

std::string	averagine_type_
	type of averagine to use

Protected Attributes inherited from ProgressLogger
LogType	type_

time_t	last_invoke_

ProgressLoggerImpl *	current_logger_

Additional Inherited Members
Static Protected Attributes inherited from ProgressLogger
static std::atomic< int >	recursion_depth_

Detailed Description

base class for filtering centroided and profile data for peak patterns

The algorithm searches for patterns of multiple peptides in the data. The peptides appear as characteristic patterns of isotopic peaks in MS1 spectra. We first search the centroided data, and optionally in a second step the spline interpolated profile data. For each peak pattern the algorithm generates a filter result.

The algorithm differs slightly for centroided and profile input data. This base class comprises code common to both. The two child classes MultiplexFilteringCentroided and MultiplexFilteringProfile contain specific functions and the primary filter() method.

See also: MultiplexIsotopicPeakPattern; MultiplexFilteredMSExperiment; MultiplexFilteringCentroided; MultiplexFilteringProfile

Class Documentation

◆ OpenMS::MultiplexFiltering::BlacklistEntry

struct OpenMS::MultiplexFiltering::BlacklistEntry

an entry of blacklist_ (spectrum index and peak index in exp_centroided_) and a value read from or written to it

Class Members
size_t	mz_idx
size_t	rt_idx
int	value

Member Typedef Documentation

◆ PeakFilter

typedef std::function<std::optional<MultiplexFilteredPeak>(SignedSize, std::vector<BlacklistEntry>*)> PeakFilter

protected

decides on one peak of a 'white' spectrum

Called with the peak index in the white spectrum and, if not null, the list to which every entry of blacklist_ read is to be appended (see filterPeakPositions_()). Returns the peak with its satellites if it passes all filters.

◆ White2Original

typedef std::vector<std::map<int, int> > White2Original

index mapping from a 'white' experiment to its original experiment

An MSExperiment contains a set of spectra each containing a number of peaks. In the course of the filtering, some peaks are blacklisted since they are identified to belong to a certain pattern i.e. peptide. An experiment in which blacklisted peaks are removed is called 'white'. White spectra contain fewer peaks than their corresponding primary spectra. Consequently, their indices are shifted. The type maps a peak index in a 'white' spectrum back to its original spectrum.

Constructor & Destructor Documentation

◆ MultiplexFiltering()

MultiplexFiltering	(	const MSExperiment &	exp_centroided,
		const std::vector< MultiplexIsotopicPeakPattern > &	patterns,
		int	isotopes_per_peptide_min,
		int	isotopes_per_peptide_max,
		double	intensity_cutoff,
		double	rt_band,
		double	mz_tolerance,
		bool	mz_tolerance_unit,
		double	peptide_similarity,
		double	averagine_similarity,
		double	averagine_similarity_scaling,
		std::string	averagine_type = `"peptide"`
	)

constructor

Parameters

[in]	exp_centroided	experimental data in centroid mode
[in]	patterns	patterns of isotopic peaks to be searched for
[in]	isotopes_per_peptide_min	minimum number of isotopic peaks in peptides
[in]	isotopes_per_peptide_max	maximum number of isotopic peaks in peptides
[in]	intensity_cutoff	intensity cutoff
[in]	rt_band	RT range used for filtering
[in]	mz_tolerance	error margin in m/z for matching expected patterns to experimental data
[in]	mz_tolerance_unit	unit for mz_tolerance, ppm (true), Da (false)
[in]	peptide_similarity	similarity score for two peptides in the same multiplet
[in]	averagine_similarity	similarity score for peptide isotope pattern and averagine model
[in]	averagine_similarity_scaling	scaling factor x for the averagine similarity parameter p when detecting peptide singlets. With p' = p + x(1-p).
[in]	averagine_type	Averagine model to use: 'peptide', 'RNA', 'DNA'

Member Function Documentation

◆ blacklistEntries_()

void blacklistEntries_	(	const MultiplexFilteredPeak &	peak,
		unsigned	pattern_idx,
		std::vector< BlacklistEntry > &	entries
	)		const

protected

the entries of blacklist_ that blacklistPeak_() sets, with their new values, in the order it sets them

They do not depend on the blacklist.

Parameters

[in]	peak	peak to be blacklisted
[in]	pattern_idx	index of the pattern in patterns_
[out]	entries	the entries are appended to it

◆ blacklistPeak_() [1/2]

void blacklistPeak_ ( const MultiplexFilteredPeak & peak )

protected

blacklist this peak

Blacklist all satellites associated with this peak.

Parameters

[in] peak peak to be blacklisted

◆ blacklistPeak_() [2/2]

void blacklistPeak_	(	const MultiplexFilteredPeak &	peak,
		unsigned	pattern_idx
	)

protected

blacklist this peak

Each of the satellites is associated with a specific mass trace. We blacklist all peaks in these mass traces (even if they are not a satellite) extending them by a margin rt_band_.

Parameters

[in]	peak	peak to be blacklisted
[in]	pattern_idx	index of the pattern in patterns_

◆ checkForSignificantPeak_()

int checkForSignificantPeak_	(	double	mz,
		double	mz_tolerance,
		MSExperiment::ConstIterator &	it_rt,
		double	intensity_first_peak
	)		const

protected

check for significant peak

Parameters

[in]	mz	position where the peak is expected
[in]	mz_tolerance	m/z tolerance within the peak may lie
[in]	it_rt	pointer to the spectrum
[in]	intensity_first_peak	intensity to compare to

Returns: -1 (if there is no significant peak), or peak index mz_idx (if there is a significant peak)

◆ filterAveragineModel_()

bool filterAveragineModel_	(	const MultiplexIsotopicPeakPattern &	pattern,
		const MultiplexFilteredPeak &	peak
	)		const

protected

check if the satellite peaks conform with the averagine model

Check if the intensities of the satellite peaks correlate with the peak intensities of the averagine model. We check both Pearson and Spearman rank correlation.

Parameters

[in]	pattern	m/z pattern to search for
[in]	peak	peak with set of satellite peaks

Returns: boolean if this filter was passed i.e. the correlation coefficient is greater than averagine_similarity_

◆ filterPeakPositions_()

bool filterPeakPositions_	(	double	mz,
		const MSExperiment::ConstIterator &	it_rt_begin,
		const MSExperiment::ConstIterator &	it_rt_band_begin,
		const MSExperiment::ConstIterator &	it_rt_band_end,
		const MultiplexIsotopicPeakPattern &	pattern,
		MultiplexFilteredPeak &	peak,
		std::vector< BlacklistEntry > *	blacklist_reads = `nullptr`
	)		const

protected

check if there are enough peaks in the RT band to form the pattern

Checks if there are peaks at m/z positions corresponding to the pattern and that the primary peak position is not blacklisted.

Parameters

[in]	mz	m/z of the primary peak
[in]	it_rt_begin	RT iterator of the very first spectrum of the experiment (needed to determine indices)
[in]	it_rt_band_begin	RT iterator of the first spectrum in the RT band
[in]	it_rt_band_end	RT iterator of the spectrum after the last spectrum in the RT band
[in]	pattern	m/z pattern to search for
[out]	peak	filter result output
[out]	blacklist_reads	if not null, every entry of blacklist_ read is appended to it, with the value read

Returns: boolean if this filter was passed i.e. there are isotopes_per_peptide_min_ or more mass traces which form the pattern.

◆ filterPeptideCorrelation_()

bool filterPeptideCorrelation_	(	const MultiplexIsotopicPeakPattern &	pattern,
		const MultiplexFilteredPeak &	peak
	)		const

protected

check if corresponding satellite peaks of different peptides show a good correlation

Different peptides in the same multiplet have the same amino acid sequence and should therefore exhibit very similar isotope distributions. The filter checks if satellite peaks corresponding to different isotopes in different peptide features show a strong correlation. The filter is of course ignored for singlet feature detection.

Parameters

[in]	pattern	m/z pattern to search for
[in]	peak	peak with set of satellite peaks

Returns: boolean if this filter was passed i.e. the correlation coefficient is greater than peptide_similarity_

◆ filterSpectrum_()

void filterSpectrum_	(	size_t	idx_rt,
		unsigned	pattern_idx,
		const PeakFilter &	filter_peak,
		MultiplexFilteredMSExperiment &	result
	)

protected

filter all peaks of one 'white' spectrum, add the ones that pass to result and blacklist them

Whether a peak passes depends on the blacklist, which every peak that passes extends. The result is that of filtering the peaks one after another in order of m/z (as a single thread does), whatever the number of threads: The peaks are filtered in parallel against the blacklist as it is when the spectrum is reached; every decision records the blacklist entries it read, and every peak that passes the entries it would blacklist. Then the peaks are accepted in order of m/z. A decision stands unless a peak accepted before it in this spectrum has changed one of the entries it read; then the peak is filtered again against the current blacklist. With one thread, the peaks are simply filtered in order of m/z.

Parameters

[in]	idx_rt	index of the spectrum in exp_centroided_white_
[in]	pattern_idx	index of the pattern in patterns_
[in]	filter_peak	decides on one peak of the spectrum
[out]	result	filter result the peaks that pass are added to

◆ getBlacklist()

MSExperiment getBlacklist ( )

returns the blacklisted, centroided peaks

◆ getCentroidedExperiment()

MSExperiment & getCentroidedExperiment ( )

returns the intensity-filtered, centroided spectral data

◆ updateWhiteMSExperiment_()

void updateWhiteMSExperiment_ ( )

protected

construct an MS experiment from exp_centroided_ containing peaks which have not been previously blacklisted in blacklist_

In addition, construct an index mapping of 'white' peak positions to their position in the corresponding, original spectrum.

Member Data Documentation

◆ averagine_similarity_

double averagine_similarity_

protected

averagine similarity

◆ averagine_similarity_scaling_

double averagine_similarity_scaling_

protected

averagine similarity scaling

◆ averagine_type_

std::string averagine_type_

protected

type of averagine to use

◆ blacklist_

std::vector<std::vector<int> > blacklist_

protected

auxiliary structs for blacklisting

◆ blacklist_reads_

std::vector<std::vector<BlacklistEntry> > blacklist_reads_

protected

for each peak of the spectrum being filtered, the entries of blacklist_ its decision read (see filterSpectrum_())

Kept from spectrum to spectrum, so that the lists keep their memory.

◆ blacklist_writes_

std::vector<std::vector<BlacklistEntry> > blacklist_writes_

protected

for each peak of the spectrum being filtered that passes, the entries of blacklist_ it sets (see filterSpectrum_())

◆ exp_centroided_

MSExperiment exp_centroided_

protected

centroided experimental data

◆ exp_centroided_mapping_

White2Original exp_centroided_mapping_

protected

mapping of peak indices from a 'white' experiment exp_centroided_white_ to its original experiment exp_centroided_

◆ exp_centroided_white_

MSExperiment exp_centroided_white_

protected

"white" centroided experimental data

subset of all peaks of exp_centroided_ which are not blacklisted in blacklist_

◆ intensity_cutoff_

double intensity_cutoff_

protected

intensity cutoff

◆ isotopes_per_peptide_max_

size_t isotopes_per_peptide_max_

protected

maximum number of isotopic peaks per peptide

◆ isotopes_per_peptide_min_

size_t isotopes_per_peptide_min_

protected

minimum number of isotopic peaks per peptide

◆ mz_tolerance_

double mz_tolerance_

protected

m/z shift tolerance

◆ mz_tolerance_unit_in_ppm_

bool mz_tolerance_unit_in_ppm_

protected

unit for m/z shift tolerance (ppm - true, Da - false)

◆ patterns_

std::vector<MultiplexIsotopicPeakPattern> patterns_

protected

list of peak patterns

◆ peptide_similarity_

double peptide_similarity_

protected

peptide similarity

◆ rt_band_

double rt_band_

protected

RT range used for filtering.

Classes

Public Types

Public Member Functions

Protected Types

Protected Member Functions

Protected Attributes

Additional Inherited Members

Detailed Description

Class Documentation

◆ OpenMS::MultiplexFiltering::BlacklistEntry

Member Typedef Documentation

◆ PeakFilter

◆ White2Original

Constructor & Destructor Documentation

◆ MultiplexFiltering()

Member Function Documentation

◆ blacklistEntries_()

◆ blacklistPeak_() [1/2]

◆ blacklistPeak_() [2/2]

◆ checkForSignificantPeak_()

◆ filterAveragineModel_()

◆ filterPeakPositions_()

◆ filterPeptideCorrelation_()

◆ filterSpectrum_()

◆ getBlacklist()

◆ getCentroidedExperiment()

◆ updateWhiteMSExperiment_()

Member Data Documentation

◆ averagine_similarity_

◆ averagine_similarity_scaling_

◆ averagine_type_

◆ blacklist_

◆ blacklist_reads_

◆ blacklist_writes_

◆ exp_centroided_

◆ exp_centroided_mapping_

◆ exp_centroided_white_

◆ intensity_cutoff_

◆ isotopes_per_peptide_max_

◆ isotopes_per_peptide_min_

◆ mz_tolerance_

◆ mz_tolerance_unit_in_ppm_

◆ patterns_

◆ peptide_similarity_

◆ rt_band_