Data processing pipeline for one Flow Injection Analysis MS (FIA-MS) sample. More...

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

Inheritance diagram for FIAMSDataProcessor:

Collaboration diagram for FIAMSDataProcessor:

Public Member Functions
	FIAMSDataProcessor ()
	Construct with the FIA-MS default parameters.

	~FIAMSDataProcessor () override=default
	Default destructor.

	FIAMSDataProcessor (const FIAMSDataProcessor &cp)=default
	Copy constructor.

FIAMSDataProcessor &	operator= (const FIAMSDataProcessor &fdp)=default
	Copy assignment.

bool	run (const MSExperiment &experiment, const float n_seconds, OpenMS::MzTab &output, const bool load_cached_spectrum=true)
	Run the full FIA-MS pipeline for one time window and (re)compute the accurate-mass-search result.

void	cutForTime (const MSExperiment &experiment, const float n_seconds, std::vector< MSSpectrum > &output)
	Append spectra of `experiment` whose retention time is strictly less than `n_seconds` to `output`.

MSSpectrum	mergeAlongTime (const std::vector< OpenMS::MSSpectrum > &input)
	Sum `input` across the time axis into a single spectrum using a per-band sliding bin size.

MSSpectrum	extractPeaks (const MSSpectrum &input)
	Smooth `input` with the configured SavitzkyGolayFilter and pick peaks via PeakPickerHiRes.

FeatureMap	convertToFeatureMap (const MSSpectrum &input)
	Wrap `input's` peaks as `Feature` objects in a new `FeatureMap`, tagging the `"scan_polarity"` meta value.

MSSpectrum	trackNoise (const MSSpectrum &input)
	Build a parallel "noise level" spectrum: per-peak m/z carried over, intensity replaced by the local noise estimate.

void	runAccurateMassSearch (FeatureMap &input, OpenMS::MzTab &output)
	Run AccurateMassSearchEngine on `input` and write the matches into `output`.

const std::vector< float > &	getMZs ()
	Return the band centres used by mergeAlongTime.

const std::vector< float > &	getBinSizes ()
	Return the per-band sliding bin sizes used by mergeAlongTime.

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

	DefaultParamHandler (const DefaultParamHandler &rhs)
	Copy constructor.

virtual	~DefaultParamHandler ()
	Destructor.

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

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

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

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

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

const std::string &	getName () const
	Non-mutable access to the name.

void	setName (const std::string &name)
	Mutable access to the name.

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

Protected Member Functions
void	updateMembers_ () override
	Recompute the per-band `mzs_` / `bin_sizes_` caches and push the `sgf:*` parameters into the SavitzkyGolayFilter member.

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

Private Member Functions
void	storeSpectrum_ (const MSSpectrum &input, const std::string &filename)
	Write `input` to `filename` as a single-spectrum mzML file via `FileHandler`.

Private Attributes
std::vector< float >	mzs_
	Per-band m/z centres consumed by mergeAlongTime; populated by updateMembers_.

std::vector< float >	bin_sizes_
	Per-band sliding bin sizes parallel to mzs_; populated by updateMembers_.

SavitzkyGolayFilter	sgfilter_
	Smoothing filter used by extractPeaks; configured from `sgf:*` parameters.

PeakPickerHiRes	picker_
	Peak picker used by extractPeaks; configured from the picker's own parameter section.

Additional Inherited Members
Static Public Member Functions inherited from DefaultParamHandler
static void	writeParametersToMetaValues (const Param &write_this, MetaInfoInterface &write_here, const std::string &key_prefix="")
	Writes all parameters to meta values.

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

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

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

std::string	error_name_
	Name that is displayed in error messages during the parameter checking.

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

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

Detailed Description

Data processing pipeline for one Flow Injection Analysis MS (FIA-MS) sample.

FIA-MS omits chromatographic separation; the entire sample is delivered in a short time window. This class consumes one MSExperiment together with a target time cut-off and performs:

truncate the experiment at n_seconds (drops spectra at RT >= n_seconds);
sum the surviving spectra along the time axis with a per-band sliding bin size bin_size[i] = mzs_[i] / (resolution * 4);
smooth the summed spectrum with SavitzkyGolayFilter;
pick peaks with PeakPickerHiRes;
run AccurateMassSearchEngine against the configured database / adduct list and write the result into the caller's MzTab object.

A cached picked spectrum is reused on subsequent runs of the same sample when load_cached_spectrum is true, skipping the cut/merge/pick steps.

Batches of FIA-MS samples are driven by FIAMSScheduler which loops over CSV rows and calls run once per ;-separated time window per row.

The workflow is inspired by Fuhrer et al. (https://pubs.acs.org/doi/10.1021/ac201267k); this is not an exact reimplementation.

Configuration is via the DefaultParamHandler parameter sections "" (root), "db:", "sgf:", and "sne:" — see the constructor for the full default map.

Constructor & Destructor Documentation

◆ FIAMSDataProcessor() [1/2]

FIAMSDataProcessor ( )

Construct with the FIA-MS default parameters.

Registers the following parameters with their defaults: filename = "fiams", dir_output = "", resolution = 120000, polarity in {positive, negative} (default positive), max_mz = 1500, bin_step = 20, db:mapping = ["CHEMISTRY/HMDBMappingFile.tsv"], db:struct = ["CHEMISTRY/HMDB2StructMapping.tsv"], positive_adducts = "CHEMISTRY/PositiveAdducts.tsv" (advanced), negative_adducts = "CHEMISTRY/NegativeAdducts.tsv" (advanced), store_progress in {true, false} (default true), sgf:frame_length = 11, sgf:polynomial_order = 4, sne:window = 10.

◆ ~FIAMSDataProcessor()

~FIAMSDataProcessor ( )

overridedefault

Default destructor.

◆ FIAMSDataProcessor() [2/2]

FIAMSDataProcessor ( const FIAMSDataProcessor & cp )

default

Copy constructor.

Member Function Documentation

◆ convertToFeatureMap()

FeatureMap convertToFeatureMap ( const MSSpectrum & input )

Wrap input's peaks as Feature objects in a new FeatureMap, tagging the "scan_polarity" meta value.

For every peak in input, emits one Feature with the peak's m/z and intensity and a "scan_polarity" meta value taken from the polarity parameter ("positive" or "negative"). The returned map is unsorted.

Parameters

[in] input Picked spectrum.

Returns: FeatureMap with one feature per input peak; scan_polarity meta value set on every feature.

◆ cutForTime()

void cutForTime	(	const MSExperiment &	experiment,
		const float	n_seconds,
		std::vector< MSSpectrum > &	output
	)

Append spectra of experiment whose retention time is strictly less than n_seconds to output.

output is appended to (existing entries are preserved). The RT comparison is strict (< ), so spectra at exactly n_seconds are dropped.

Parameters

[in]	experiment	Source experiment.
[in]	n_seconds	Upper retention-time bound (exclusive).
[in,out]	output	Receives the retained spectra.

◆ extractPeaks()

MSSpectrum extractPeaks ( const MSSpectrum & input )

Smooth input with the configured SavitzkyGolayFilter and pick peaks via PeakPickerHiRes.

The filter and picker are class members configured through sgf:* and the picker's own parameter section; their state is set up by the constructor and refreshed by updateMembers_ when sgf:* parameters change. input is not modified — a local copy is filtered in place before picking.

Parameters

[in] input Smoothed/picked input spectrum.

Returns: Spectrum of picked peaks.

◆ getBinSizes()

const std::vector< float > & getBinSizes ( )

Return the per-band sliding bin sizes used by mergeAlongTime.

Each bin size is computed as bin_size[i] = mzs_[i] / (resolution * 4).

Returns: Const reference to the vector of sliding bin sizes, parallel to getMZs.

◆ getMZs()

const std::vector< float > & getMZs ( )

Return the band centres used by mergeAlongTime.

The list contains {bin_step, 2*bin_step, ..., (n-1)*bin_step} where n = max_mz / bin_step.

Returns: Const reference to the vector of band centres. Refreshed by updateMembers_ when parameters change.

◆ mergeAlongTime()

MSSpectrum mergeAlongTime ( const std::vector< OpenMS::MSSpectrum > & input )

Sum input across the time axis into a single spectrum using a per-band sliding bin size.

For each consecutive pair (mzs_[i], mzs_[i+1]), calls SpectrumAddition::addUpSpectra(input, bin_sizes_[i], false) and keeps only the peaks whose m/z falls in [mzs_[i], mzs_[i+1]). Iteration stops one short of the end (i < mzs_.size() - 1), so the upper-most band is not emitted — configure max_mz to extend past your highest expected peak. The returned spectrum is sorted by m/z position.

Parameters

[in] input Spectra to sum (typically the output of cutForTime).

Returns: Single summed spectrum, sorted by m/z.

◆ operator=()

FIAMSDataProcessor & operator= ( const FIAMSDataProcessor & fdp )

default

Copy assignment.

◆ run()

bool run	(	const MSExperiment &	experiment,
		const float	n_seconds,
		OpenMS::MzTab &	output,
		const bool	load_cached_spectrum = `true`
	)

Run the full FIA-MS pipeline for one time window and (re)compute the accurate-mass-search result.

Pipeline:

Compute the cached picked-spectrum path as {dir_output}/{filename}_picked_{int(n_seconds)}.mzML.
If load_cached_spectrum is true and the path exists, load it via FileHandler with MZML pinned and take its first spectrum as the picked spectrum (skip steps 3-5).
Otherwise: call cutForTime → mergeAlongTime → extractPeaks.
If store_progress is "true" (and we recomputed in step 3), store the merged spectrum to {dir_output}/{filename}_merged_{int(n_seconds)}.mzML and the picked spectrum to the cached path.
Estimate signal-to-noise via trackNoise and convert the picked spectrum to a FeatureMap via convertToFeatureMap.
Always store the signal-to-noise spectrum to {dir_output}/{filename}_signal_to_noise_{int(n_seconds)}.mzML — note that this is independent of the store_progress setting.
Call runAccurateMassSearch and write {dir_output}/{filename}_{int(n_seconds)}.mzTab via MzTabFile::store.

Progress is logged via OPENMS_LOG_INFO (cache load start/finish or calculation start/finish).

Parameters

[in]	experiment	Input MS experiment (centroided MS1 is assumed).
[in]	n_seconds	Time cut-off (seconds). Cast to `int` when forming the per-run filename suffix, so sub-second offsets collapse.
[out]	output	Receives the accurate-mass-search results from runAccurateMassSearch.
[in]	load_cached_spectrum	If `true` and the cached picked-spectrum file exists, reuse it instead of re-running the cut / merge / pick steps. Default `true`.

Returns: true when the picked spectrum was loaded from the cached file, false when it was recomputed.

◆ runAccurateMassSearch()

void runAccurateMassSearch	(	FeatureMap &	input,
		OpenMS::MzTab &	output
	)

Run AccurateMassSearchEngine on input and write the matches into output.

Configures AccurateMassSearchEngine with:

ionization_mode = "auto" (polarity is detected from the feature map, not taken from the class's polarity parameter);
mass_error_value = 1e6 / (resolution * 2) ppm;
db:mapping / db:struct / positive_adducts / negative_adducts forwarded from the class parameters;
keep_unidentified_masses = "false" — only identified masses are reported. Then calls init followed by run.

Parameters

[in,out]	input	Feature map to search; consumed by `AccurateMassSearchEngine::run`.
[out]	output	Accurate-mass-search results in `MzTab` form.

◆ storeSpectrum_()

void storeSpectrum_	(	const MSSpectrum &	input,
		const std::string &	filename
	)

private

Write input to filename as a single-spectrum mzML file via FileHandler.

Used by run to persist the merged spectrum, picked spectrum, and signal-to-noise spectrum.

Parameters

[in]	input	Spectrum to store.
[in]	filename	Output path for the mzML file.

Exceptions

Exception Propagates exceptions from FileHandler::storeExperiment on I/O failure (no in-class guard).

◆ trackNoise()

MSSpectrum trackNoise ( const MSSpectrum & input )

Build a parallel "noise level" spectrum: per-peak m/z carried over, intensity replaced by the local noise estimate.

Wraps SignalToNoiseEstimatorMedianRapid with window sne:window. Returns an empty spectrum when input is empty. The output's intensity field holds the noise estimate at the corresponding m/z (not a S/N ratio).

Parameters

[in] input Picked spectrum.

Returns: Spectrum with the same m/z values as input, intensities replaced by per-peak noise estimates.

◆ updateMembers_()

void updateMembers_ ( )

overrideprotectedvirtual

Recompute the per-band mzs_ / bin_sizes_ caches and push the sgf:* parameters into the SavitzkyGolayFilter member.

Invoked by DefaultParamHandler when any parameter changes (notably max_mz, bin_step, resolution, sgf:frame_length, sgf:polynomial_order).

Reimplemented from DefaultParamHandler.

Member Data Documentation

◆ bin_sizes_

std::vector<float> bin_sizes_

private

Per-band sliding bin sizes parallel to mzs_; populated by updateMembers_.

◆ mzs_

std::vector<float> mzs_

private

Per-band m/z centres consumed by mergeAlongTime; populated by updateMembers_.

◆ picker_

PeakPickerHiRes picker_

private

Peak picker used by extractPeaks; configured from the picker's own parameter section.

◆ sgfilter_

SavitzkyGolayFilter sgfilter_

private

Smoothing filter used by extractPeaks; configured from sgf:* parameters.

Public Member Functions

Protected Member Functions

Private Member Functions

Private Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ FIAMSDataProcessor() [1/2]

◆ ~FIAMSDataProcessor()

◆ FIAMSDataProcessor() [2/2]

Member Function Documentation

◆ convertToFeatureMap()

◆ cutForTime()

◆ extractPeaks()

◆ getBinSizes()

◆ getMZs()

◆ mergeAlongTime()

◆ operator=()

◆ run()

◆ runAccurateMassSearch()

◆ storeSpectrum_()

◆ trackNoise()

◆ updateMembers_()

Member Data Documentation

◆ bin_sizes_

◆ mzs_

◆ picker_

◆ sgfilter_