FragmentIonLikelihoodModel::PeakLists Class Reference

Compact peak lists of the spectra of one run: per peak only its m/z and its intensity-rank bin. More...

#include <OpenMS/ANALYSIS/ID/FragmentIonLikelihoodModel.h>

Public Member Functions
void	reset (Size spectra)
	Remove all peak lists and make room for spectra empty ones.
void	clear ()
	Remove all peak lists and release their memory.
Size	size () const
	Number of peak lists (spectra)
void	assign (Size index, const MSSpectrum &spectrum)
	Store the peaks of spectrum as list index, replacing what was there.
Size	peaks (Size index) const
	Number of peaks of list index.
double	mz (Size index, Size peak) const
	m/z of peak peak of list index
Size	rankBin (Size index, Size peak) const
	Intensity-rank bin (see rankBin()) of peak peak of list index.
Int	findNearest (Size index, double mz, double tolerance) const
	The peak of list index nearest to mz, if it lies within [mz - tolerance, mz + tolerance].
Size	totalPeaks () const
	Number of peaks over all lists.
Size	memoryUsage () const
	Bytes held by the peak lists (peak data, list headers)

Private Attributes
std::vector< std::vector< float > >	mz_
	m/z per peak, ascending, per spectrum
std::vector< std::vector< std::uint8_t > >	rank_bins_
	intensity-rank bin per peak, per spectrum

Detailed Description

Compact peak lists of the spectra of one run: per peak only its m/z and its intensity-rank bin.

Five bytes per peak (m/z as float, rank bin as byte) instead of a copy of the spectra, so the peak lists of a whole run can be kept from preprocessing until the PSMs are annotated. The float m/z deviates from the double by at most 0.06 ppm, far below any fragment tolerance.

Member Function Documentation

assign()

void assign	(	Size	index,
		const MSSpectrum &	spectrum
	)

Store the peaks of spectrum as list index, replacing what was there.

The intensity ranks are taken among the peaks of spectrum (1 = most intense; ties by ascending m/z). Different indices may be assigned concurrently.

Exceptions

Exception::IndexOverflow	if index is not below size()
Exception::IllegalArgument	if spectrum is not sorted by m/z

Referenced by NuXLIonLikelihood::setSpectra().

clear()

void clear

(

)

Remove all peak lists and release their memory.

findNearest()

Int findNearest	(	Size	index,
		double	mz,
		double	tolerance
	)		const

The peak of list index nearest to mz, if it lies within [mz - tolerance, mz + tolerance].

Same rule as MSSpectrum::findNearest(mz, tolerance): of two equally near peaks the lower one.

Returns

the peak index, or -1

memoryUsage()

Size memoryUsage

(

)

const

Bytes held by the peak lists (peak data, list headers)

mz()

double mz ( Size  index, Size  peak  ) const

inline

m/z of peak peak of list index

peaks()

Size peaks ( Size  index ) const

inline

Number of peaks of list index.

Referenced by NuXLIonLikelihood::matchHypothesis_().

rankBin()

Size rankBin ( Size  index, Size  peak  ) const

inline

Intensity-rank bin (see rankBin()) of peak peak of list index.

reset()

void reset

(

Size

spectra

)

Remove all peak lists and make room for spectra empty ones.

Referenced by NuXLIonLikelihood::setSpectra().

size()

Size size ( ) const

inline

Number of peak lists (spectra)

Referenced by NuXLIonLikelihood::matchHypothesis_().

totalPeaks()

Size totalPeaks

(

)

const

Number of peaks over all lists.

Member Data Documentation

mz_

std::vector<std::vector<float> > mz_

private

m/z per peak, ascending, per spectrum

rank_bins_

std::vector<std::vector<std::uint8_t> > rank_bins_

private

intensity-rank bin per peak, per spectrum