alphatims.bruker¶

This module provides functions to handle Bruker data. It primarily implements the TimsTOF class, that acts as an in-memory container for Bruker data accession and storage.

Exceptions:

PrecursorFloatError

Used to indicate that a precursor value is not an int but a float.

Classes:

TimsTOF(bruker_d_folder_name, *[, ...])

A class that stores Bruker TimsTOF data in memory for fast access.

Functions:

`add_intensity_to_bin`(query_index, ...)	Add the intensity of a query to the appropriate bin.
`calculate_dia_cycle_mask`(dia_mz_cycle, ...)	Calculate a boolean mask for cyclic push indices satisfying queries.
`centroid_spectra`(index, spectrum_indptr, ...)	Smoothen and centroid a profile spectrum (inplace operation).
`convert_slice_key_to_float_array`(key)	Convert a key to a slice float array.
`convert_slice_key_to_int_array`(data, key, ...)	Convert a key of a data dimension to a slice integer array.
`filter_indices`(frame_slices, scan_slices, ...)	Filter raw indices by slices from all dimensions.
`filter_spectra_by_abundant_peaks`(index, ...)	Filter a spectrum to retain only the most abundant peaks.
`filter_tof_to_csr`(tof_slices, push_indices, ...)	Get a CSR-matrix with raw indices satisfying push indices and tof slices.
`get_dia_push_indices`(frame_slices, ...[, ...])	Filter DIA push indices by slices from LC, TIMS and QUAD.
`indptr_lookup`(targets, queries[, ...])	Find the indices of queries in targets.
`init_bruker_dll`([bruker_dll_file_name])	Open a bruker.dll in Python.
`open_bruker_d_folder`(bruker_d_folder_name[, ...])	A context manager for a bruker dll connection to a .d folder.
`parse_decompressed_bruker_binary_type1`(...)	Parse a Bruker binary scan buffer into tofs and intensities.
`parse_decompressed_bruker_binary_type2`(...)	Parse a Bruker binary frame buffer into scans, tofs and intensities.
`parse_keys`(data, keys)	Convert different keys to a key dict with defined types.
`process_frame`(frame_id, tdf_bin_file_name, ...)	Read and parse a frame directly from a Bruker .d.analysis.tdf_bin.
`read_bruker_binary`(frames, ...[, ...])	Read all data from an "analysis.tdf_bin" of a Bruker .d folder.
`read_bruker_sql`(bruker_d_folder_name[, ...])	Read metadata, (fragment) frames and precursors from a Bruker .d folder.
`set_precursor`(precursor_index, offset_order, ...)	Sum the intensities of all pushes belonging to a single precursor.
`trim_spectra`(index, spectrum_tof_indices, ...)	Trim remaining bytes after merging of multiple pushes.
`valid_precursor_index`(precursor_index, ...)	Check if a precursor index is included in the slices.
`valid_quad_mz_values`(low_mz_value, ...)	Check if the low and high quad mz values are included in the slices.

exception alphatims.bruker.PrecursorFloatError[source]¶

Bases: TypeError

Used to indicate that a precursor value is not an int but a float.

class alphatims.bruker.TimsTOF(bruker_d_folder_name: str, *, mz_estimation_from_frame: int = 1, mobility_estimation_from_frame: int = 1, slice_as_dataframe: bool = True, use_calibrated_mz_values_as_default: int = 0, use_hdf_if_available: bool = True, mmap_detector_events: bool = True, drop_polarity: bool = True, convert_polarity_to_int: bool = True)[source]¶

Bases: object

A class that stores Bruker TimsTOF data in memory for fast access.

Data can be read directly from a Bruker .d folder. All OS’s are supported, but reading mz_values and mobility_values from a .d folder requires Windows or Linux due to availability of Bruker libraries. On MacOS, they are estimated based on metadata, but these values are not guaranteed to be correct. Often they fall within 0.02 Th, but errors up to 6 Th have already been observed!

A TimsTOF object can also be exported to HDF for subsequent access. This file format is portable to all OS’s. As such, initial reading on Windows with correct mz_values and mobility_values can be done and the resulting HDF file can safely be read on MacOS. This HDF file also provides improved accession times for subsequent use.

After reading, data can be accessed with traditional Python slices. As TimsTOF data is 5-dimensional, the data can be sliced in 5 dimensions as well. These dimensions follows the design of the TimsTOF Pro:

1 LC: rt_values, frame_indices
The first dimension allows to slice retention_time values or frames indices. These values and indices have a one-to-one relationship.

2 TIMS: mobility_values, scan_indices
The second dimension allows to slice mobility values or scan indices (i.e. a single push). These values and indices have a one-to-one relationship.

3 QUAD: quad_mz_values, precursor_indices
The third dimension focusses on the quadrupole and indirectly on the collision cell. It allows to slice lower and upper quadrupole mz values (e.g. the m/z of unfragmented ions / precursors). If set to -1, the quadrupole and collision cell are assumed to be inactive, i.e. precursor ions are detected instead of fragments. Equally, this dimension allows to slice precursor indices. Precursor index 0 defaults to all precusors (i.e. quad mz values equal to -1). In DDA, precursor indices larger than 0 point to ddaPASEF MSMS spectra. In DIA, precursor indices larger than 0 point to windows, i.e. all scans in a frame with equal quadrupole and collision settings that is repeated once per full cycle. Note that these values do not have a one-to-one relationship.

4 TOF: mz_values, tof_indices
The fourth dimension allows to slice (fragment) mz_values or tof indices. Note that the quadrupole dimension determines if precursors are detected or fragments. These values and indices have a one-to-one relationship.

5 DETECTOR: intensity_values
The fifth dimension allows to slice intensity values.

Note that all dimensions except for the detector have both (float) values and (integer) indices. For each dimension, slices can be provided in several different ways:

int:
A single int can be used to select a single index. If used in the fifth dimension, it still allows to select intensity_values

float:
A single float can be used to select a single value. As the values arrays are discrete, the smallest index with a value equal to or larger than this value is actually selected. For intensity_value slicing, the exact value is used.

slice:
A Python slice with start, stop and step can be provided. Start and stop values can independently be set to int or float. If a float is provided it conversed to an int as previously described. The step always needs to be provided as an int. Since there is not one-to-one relation from values to indices for QUAD and DETECTOR, the step value is ignored in these cases and only start and stop can be used.

IMPORTANT NOTE: negative start, step and stop integers are not supported!

iterable:
An iterable with (mixed) floats and ints can also be provided, in a similar fashion as Numpy’s fancy indexing.

IMPORTANT NOTE: The resulting integers after float->int conversion need to be sorted in ascending order!

np.ndarray:
Multiple slicing is supported by providing either a np.int64[:, 3] array, where each row is assumed to be a (start, stop, step) tuple or np.float64[:, 2] where each row is assumed to be a (start, stop) tuple.

IMPORTANT NOTE: These arrays need to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(precursor_slices[:, :2].ravel()) >= 0) = True).

Alternatively, a dictionary can be used to define filters for each dimension (see examples).

The result of such slicing is a pd.DataFrame with the following columns:

raw_indices

frame_indices

scan_indices

precursor_indices

tof_indices

rt_values

mobility_values

quad_low_mz_values

quad_high_mz_values

mz_values

intensity_values

Instead of returning a pd.DataFrame, raw indices can be returned by setting the last slice element to “raw”.

Examples

>>> data[:100.0]
# Return all datapoints with rt_values < 100.0 seconds

>>> data[:, 450]
# Return all datapoints with scan_index = 450

>>> data[:, :, 700.: 710.]
# Return all datapoints with 700.0 <= quad_mz_values < 710.0

>>> data[:, :, :, 621.9: 191000]
# Return all datapoints with 621.9 <= mz_values and
# tof_indices < 191000

>>> data[[1, 8, 10], :, 0, 621.9: np.inf]
# Return all datapoints from frames 1, 8 and 10, which are unfragmented
# (precursor_index = 0) and with 621.9 <= mz_values < np.inf

>>> data[:, :, 999]
# Return all datapoints from precursor 999
# (for diaPASEF this is a traditional MSMS spectrum)

>>> scan_slices = np.array([[10, 20, 1], [100, 200, 10]])
>>> data[:, scan_slices, :, :, :]
# Return all datapoints with scan_indices in range(10, 20) or
# range(100, 200, 10)

>>> df = data[
...     {
...         "frame_indices": [1, 191],
...         "scan_indices": slice(300, 800, 10),
...         "mz_values": slice(None, 400.5),
...         "intensity_values": 50,
...     }
... ]
# Slice by using a dictionary

>>> data[:, :, 999, "raw"]
# Return the raw indices of datapoints from precursor 999

Methods:

`__init__`(bruker_d_folder_name, *[, ...])	Create a Bruker TimsTOF object that contains all data in-memory.
`as_dataframe`(indices, *[, raw_indices, ...])	Convert raw indices to a pd.DataFrame.
`bin_intensities`(indices, axis)	Sum and project the intensities of the indices along 1 or 2 axis.
`calculate_global_calibrated_mz_values`([...])	Calculate global calibrated_mz_values based on two calibrant ions.
`convert_from_indices`(raw_indices, *[, ...])	Convert selected indices to a dict.
`convert_to_indices`(values, *[, ...])	Convert selected values to an array in the requested dimension.
`estimate_strike_count`(frame_slices, ...)	Estimate the number of detector events, given a set of slices.
`index_precursors`([centroiding_window, ...])	Retrieve all MS2 spectra acquired with DDA.
`save_as_hdf`(directory, file_name[, ...])	Save the TimsTOF object as an hdf file.
`save_as_mgf`(directory, file_name[, ...])	Save profile spectra from this TimsTOF object as an mgf file.
`use_calibrated_mz_values_as_default`(...)	Override the default mz_values with the global calibrated_mz_values.

Attributes:

`accumulation_times`	The accumulation times of all frames.
`acquisition_mode`	The acquisition mode.
`calibrated_mz_max_value`	The maximum calibrated mz value.
`calibrated_mz_min_value`	The minimum calibrated mz value.
`calibrated_mz_values`	np.float64[:] : The global calibrated mz values.
`dia_mz_cycle`	np.float64[:, 2] : The mz_values of a DIA cycle.
`dia_precursor_cycle`	np.int64[:] : The precursor indices of a DIA cycle.
`directory`	The directory of this TimsTOF object.
`fragment_frames`	The fragment frames table.
`frame_max_index`	The maximum frame index.
`frames`	The frames table of the analysis.tdf SQL.
`intensity_corrections`	The intensity_correction per frame.
`intensity_max_value`	The maximum intensity value.
`intensity_min_value`	The minimum intensity value.
`intensity_values`	np.uint16[:] : The intensity values.
`is_compressed`	HDF array is compressed or not.
`max_accumulation_time`	The maximum accumulation time of all frames.
`meta_data`	The metadata for the acquisition.
`mobility_max_value`	The maximum mobility value.
`mobility_min_value`	The minimum mobility value.
`mobility_values`	np.float64[:] : The mobility values.
`mz_max_value`	The maximum mz value.
`mz_min_value`	The minimum mz value.
`mz_values`	np.float64[:] : The mz values.
`precursor_indices`	np.int64[:] : The precursor indices.
`precursor_max_index`	The maximum precursor index.
`precursors`	The precursor table.
`push_indptr`	np.int64[:] : The tof indptr.
`quad_indptr`	np.int64[:] : The quad indptr (tof_indices).
`quad_mz_max_value`	The maximum quad mz value.
`quad_mz_min_value`	The minimum quad mz value.
`quad_mz_values`	np.float64[:, 2] : The (low, high) quad mz values.
`raw_quad_indptr`	np.int64[:] : The raw quad indptr (push indices).
`rt_max_value`	The maximum rt value.
`rt_values`	np.float64[:] : The rt values.
`sample_name`	The sample name of this TimsTOF object.
`scan_max_index`	The maximum scan index.
`tof_indices`	np.uint32[:] : The tof indices.
`tof_max_index`	The maximum tof index.
`version`	AlphaTims version used to create this TimsTOF object.
`zeroth_frame`	A blank zeroth frame is present so frames are 1-indexed.

__init__(bruker_d_folder_name: str, *, mz_estimation_from_frame: int = 1, mobility_estimation_from_frame: int = 1, slice_as_dataframe: bool = True, use_calibrated_mz_values_as_default: int = 0, use_hdf_if_available: bool = True, mmap_detector_events: bool = True, drop_polarity: bool = True, convert_polarity_to_int: bool = True)[source]¶

Create a Bruker TimsTOF object that contains all data in-memory.

Parameters

bruker_d_folder_name (str) – The full file name to a Bruker .d folder. Alternatively, the full file name of an already exported .hdf can be provided as well.
mz_estimation_from_frame (int) – If larger than 0, mz_values from this frame are read as default mz_values with the Bruker library. If 0, mz_values are being estimated with the metadata based on “MzAcqRangeLower” and “MzAcqRangeUpper”. IMPORTANT NOTE: MacOS defaults to 0, as no Bruker library is available. Default is 1.
mobility_estimation_from_frame (int) – If larger than 0, mobility_values from this frame are read as default mobility_values with the Bruker library. If 0, mobility_values are being estimated with the metadata based on “OneOverK0AcqRangeLower” and “OneOverK0AcqRangeUpper”. IMPORTANT NOTE: MacOS defaults to 0, as no Bruker library is available. Default is 1.
slice_as_dataframe (bool) – If True, slicing returns a pd.DataFrame by default. If False, slicing provides a np.int64[:] with raw indices. This value can also be modified after creation. Default is True.
use_calibrated_mz_values (int) – If not 0, the mz_values are overwritten with global calibrated_mz_values. If 1, calibration at the MS1 level is performed. If 2, calibration at the MS2 level is performed. Default is 0.
use_hdf_if_available (bool) – If an HDF file is available, use this instead of the .d folder. Default is True.
mmap_detector_events (bool) – Do not save the intensity_values and tof_indices in memory, but use an mmap instead. Default is True
drop_polarity (bool) – The polarity column of the frames table contains “+” or “-” and is not numerical. If True, the polarity column is dropped from the frames table. this ensures a fully numerical pd.DataFrame. If False, this column is kept, resulting in a pd.DataFrame with dtype=object. Default is True.
convert_polarity_to_int (bool) – Convert the polarity to int (-1 or +1). This allows to keep it in numerical form. This is ignored if the polarity is dropped. Default is True.

property accumulation_times¶

The accumulation times of all frames.

Type: np.ndarray

property acquisition_mode¶

The acquisition mode.

Type: str

as_dataframe(indices: numpy.ndarray, *, raw_indices: bool = True, frame_indices: bool = True, scan_indices: bool = True, quad_indices: bool = False, tof_indices: bool = True, precursor_indices: bool = True, rt_values: bool = True, rt_values_min: bool = True, mobility_values: bool = True, quad_mz_values: bool = True, push_indices: bool = True, mz_values: bool = True, intensity_values: bool = True, corrected_intensity_values: bool = True, raw_indices_sorted: bool = False)[source]¶

Convert raw indices to a pd.DataFrame.

Parameters

indices (np.int64[:]) – The raw indices for which coordinates need to be retrieved.
raw_indices (bool) – If True, include “raw_indices” in the dataframe. Default is True.
frame_indices (bool) – If True, include “frame_indices” in the dataframe. Default is True.
scan_indices (bool) – If True, include “scan_indices” in the dataframe. Default is True.
quad_indices (bool) – If True, include “quad_indices” in the dataframe. Default is False.
tof_indices (bool) – If True, include “tof_indices” in the dataframe. Default is True.
precursor_indices (bool) – If True, include “precursor_indices” in the dataframe. Default is True.
rt_values (bool) – If True, include “rt_values” in the dataframe. Default is True.
rt_values_min (bool) – If True, include “rt_values_min” in the dataframe. Default is True.
mobility_values (bool) – If True, include “mobility_values” in the dataframe. Default is True.
quad_mz_values (bool) – If True, include “quad_low_mz_values” and “quad_high_mz_values” in the dict. Default is True.
push_indices (bool) – If True, include “push_indices” in the dataframe. Default is True.
mz_values (bool) – If True, include “mz_values” in the dataframe. Default is True.
intensity_values (bool) – If True, include “intensity_values” in the dataframe. Default is True.
corrected_intensity_values (bool) – If True, include “corrected_intensity_values” in the dataframe. Default is True.
raw_indices_sorted (bool) – If True, raw_indices are assumed to be sorted, resulting in a faster conversion. Default is False.

Returns

A dataframe with all requested columns.

Return type

pd.DataFrame

bin_intensities(indices: numpy.ndarray, axis: tuple)[source]¶

Sum and project the intensities of the indices along 1 or 2 axis.

Parameters

indices (np.int64[:]) – The selected indices whose coordinates need to be summed along the selected axis.
axis (tuple) – Must be length 1 or 2 and can only contain the elements “rt_values”, “mobility_values” and “mz_values”.

Returns

np.float64[ – An array or heatmap that express the summed intensity along the selected axis.

Return type

], np.float64[:, 2]

calculate_global_calibrated_mz_values(calibrant1: tuple = (922.009798, 1.1895, slice(0, 1, None)), calibrant2: tuple = (1221.990637, 1.382, slice(0, 1, None)), mz_tolerance: float = 10, mobility_tolerance: float = 0.1) → None[source]¶

Calculate global calibrated_mz_values based on two calibrant ions.

Parameters

calibrant1 (tuple) – The first calibrant ion. This is a tuple with (mz, mobility, precursor_slice) foat values. Default is (922.009798, 1.1895, slice(0, 1)).
calibrant2 (tuple) – The first calibrant ion. This is a tuple with (mz, mobility, precursor_slice) foat values. Default is (1221.990637, 1.3820, slice(0, 1)).
mz_tolerance (float) – The tolerance window (in Th) with respect to the uncalibrated mz_values. If this is too large, the calibrant ion might not be the most intense ion anymore. If this is too small, the calibrant ion might not be contained. Default is 10.
mobility_tolerance (float) – The tolerance window with respect to the uncalibrated mobility_values. If this is too large, the calibrant ion might not be the most intense ion anymore. If this is too small, the calibrant ion might not be contained. Default is 0.1.

property calibrated_mz_max_value¶

The maximum calibrated mz value.

Type: float

property calibrated_mz_min_value¶

The minimum calibrated mz value.

Type: float

property calibrated_mz_values¶

np.float64[:] : The global calibrated mz values.

Type: np.ndarray

convert_from_indices(raw_indices, *, frame_indices=None, quad_indices=None, scan_indices=None, tof_indices=None, return_raw_indices: bool = False, return_frame_indices: bool = False, return_scan_indices: bool = False, return_quad_indices: bool = False, return_tof_indices: bool = False, return_precursor_indices: bool = False, return_rt_values: bool = False, return_rt_values_min: bool = False, return_mobility_values: bool = False, return_quad_mz_values: bool = False, return_push_indices: bool = False, return_mz_values: bool = False, return_intensity_values: bool = False, return_corrected_intensity_values: bool = False, raw_indices_sorted: bool = False) → dict[source]¶

Convert selected indices to a dict.

Parameters

raw_indices (np.int64[:], None) – The raw indices for which coordinates need to be retrieved.
frame_indices (np.int64[:], None) – The frame indices for which coordinates need to be retrieved.
quad_indices (np.int64[:], None) – The quad indices for which coordinates need to be retrieved.
scan_indices (np.int64[:], None) – The scan indices for which coordinates need to be retrieved.
tof_indices (np.int64[:], None) – The tof indices for which coordinates need to be retrieved.
return_raw_indices (bool) – If True, include “raw_indices” in the dict. Default is False.
return_frame_indices (bool) – If True, include “frame_indices” in the dict. Default is False.
return_scan_indices (bool) – If True, include “scan_indices” in the dict. Default is False.
return_quad_indices (bool) – If True, include “quad_indices” in the dict. Default is False.
return_tof_indices (bool) – If True, include “tof_indices” in the dict. Default is False.
return_precursor_indices (bool) – If True, include “precursor_indices” in the dict. Default is False.
return_rt_values (bool) – If True, include “rt_values” in the dict. Default is False.
return_rt_values_min (bool) – If True, include “rt_values_min” in the dict. Default is False.
return_mobility_values (bool) – If True, include “mobility_values” in the dict. Default is False.
return_quad_mz_values (bool) – If True, include “quad_low_mz_values” and “quad_high_mz_values” in the dict. Default is False.
return_push_indices (bool) – If True, include “push_indices” in the dict. Default is False.
return_mz_values (bool) – If True, include “mz_values” in the dict. Default is False.
return_intensity_values (bool) – If True, include “intensity_values” in the dict. Default is False.
return_corrected_intensity_values (bool) – If True, include “corrected_intensity_values” in the dict. Default is False.
raw_indices_sorted (bool) – If True, raw_indices are assumed to be sorted, resulting in a faster conversion. Default is False.

Returns

A dict with all requested columns.

Return type

dict

convert_to_indices(values: numpy.ndarray, *, return_frame_indices: bool = False, return_scan_indices: bool = False, return_tof_indices: bool = False, side: str = 'left', return_type: str = '')[source]¶

Convert selected values to an array in the requested dimension.

Parameters

values (float, np.float64[...], iterable) – The raw values for which indices need to be retrieved.
return_frame_indices (bool) – If True, convert the values to “frame_indices”. Default is False.
return_scan_indices (bool) – If True, convert the values to “scan_indices”. Default is False.
return_tof_indices (bool) – If True, convert the values to “tof_indices”. Default is False.
side (str) – If there is an exact match between the values and reference array, which index should be chosen. See also np.searchsorted. Options are “left” or “right”. Default is “left”.
return_type (str) – Alternative way to define the return type. Options are “frame_indices”, “scan_indices” or “tof_indices”. Default is “”.

Returns

An array with the same shape as values or iterable or an int which corresponds to the requested value.

Return type

np.int64[…], int

Raises

PrecursorFloatError – When trying to convert a quad float other than np.inf or -np.inf to precursor index.

property dia_mz_cycle¶

np.float64[:, 2] : The mz_values of a DIA cycle.

Type: np.ndarray

property dia_precursor_cycle¶

np.int64[:] : The precursor indices of a DIA cycle.

Type: np.ndarray

property directory¶

The directory of this TimsTOF object.

Type: str

estimate_strike_count(frame_slices: numpy.ndarray, scan_slices: numpy.ndarray, precursor_slices: numpy.ndarray, tof_slices: numpy.ndarray, quad_slices: numpy.ndarray) → int[source]¶

Estimate the number of detector events, given a set of slices.

Parameters

frame_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(frame_slices[:, :2].ravel()) >= 0) = True).
scan_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(scan_slices[:, :2].ravel()) >= 0) = True).
precursor_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(precursor_slices[:, :2].ravel()) >= 0) = True).
tof_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(tof_slices[:, :2].ravel()) >= 0) = True).
quad_slices (np.float64[:, 2]) – Each row of the array is assumed to be (lower_mz, upper_mz) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(quad_slices.ravel()) >= 0) = True).

Returns

The estimated number of detector events given these slices.

Return type

int

property fragment_frames¶

The fragment frames table.

Type: pd.DataFrame

property frame_max_index¶

The maximum frame index.

Type: int

property frames¶

The frames table of the analysis.tdf SQL.

Type: pd.DataFrame

index_precursors(centroiding_window: int = 0, keep_n_most_abundant_peaks: int = - 1) → tuple[source]¶

Retrieve all MS2 spectra acquired with DDA.

IMPORTANT NOTE: This function is intended for DDA samples. While it in theory works for DIA sample too, this probably has little value.

Parameters

centroiding_window (int) – The centroiding window to use. If 0, no centroiding is performed. Default is 0.
keep_n_most_abundant_peaks (int) – Keep the n most abundant peaks. If -1, all peaks are retained. Default is -1.

Returns

The spectrum_indptr array, spectrum_tof_indices array and spectrum_intensity_values array.

Return type

tuple (np.int64[:], np.uint32[:], np.uint32[:])

property intensity_corrections¶

The intensity_correction per frame.

Type: np.ndarray

property intensity_max_value¶

The maximum intensity value.

Type: float

property intensity_min_value¶

The minimum intensity value.

Type: float

property intensity_values¶

np.uint16[:] : The intensity values.

Type: np.ndarray

property is_compressed¶

HDF array is compressed or not.

Type: bool

property max_accumulation_time¶

The maximum accumulation time of all frames.

Type: float

property meta_data¶

The metadata for the acquisition.

Type: dict

property mobility_max_value¶

The maximum mobility value.

Type: float

property mobility_min_value¶

The minimum mobility value.

Type: float

property mobility_values¶

np.float64[:] : The mobility values.

Type: np.ndarray

property mz_max_value¶

The maximum mz value.

Type: float

property mz_min_value¶

The minimum mz value.

Type: float

property mz_values¶

np.float64[:] : The mz values.

Type: np.ndarray

property precursor_indices¶

np.int64[:] : The precursor indices.

Type: np.ndarray

property precursor_max_index¶

The maximum precursor index.

Type: int

property precursors¶

The precursor table.

Type: pd.DataFrame

property push_indptr¶

np.int64[:] : The tof indptr.

Type: np.ndarray

property quad_indptr¶

np.int64[:] : The quad indptr (tof_indices).

Type: np.ndarray

property quad_mz_max_value¶

The maximum quad mz value.

Type: float

property quad_mz_min_value¶

The minimum quad mz value.

Type: float

property quad_mz_values¶

np.float64[:, 2] : The (low, high) quad mz values.

Type: np.ndarray

property raw_quad_indptr¶

np.int64[:] : The raw quad indptr (push indices).

Type: np.ndarray

property rt_max_value¶

The maximum rt value.

Type: float

property rt_values¶

np.float64[:] : The rt values.

Type: np.ndarray

property sample_name¶

The sample name of this TimsTOF object.

Type: str

save_as_hdf(directory: str, file_name: str, overwrite: bool = False, compress: bool = False, return_as_bytes_io: bool = False)[source]¶

Save the TimsTOF object as an hdf file.

Parameters

directory (str) – The directory where to save the HDF file. Ignored if return_as_bytes_io == True.
file_name (str) – The file name of the HDF file. Ignored if return_as_bytes_io == True.
overwrite (bool) – If True, an existing file is truncated. If False, the existing file is appended to only if the original group, array or property does not exist yet. Default is False.
compress (bool) – If True, compression is used. This roughly halves files sizes (on-disk), at the cost of taking 3-6 longer accession times. See also alphatims.utils.create_hdf_group_from_dict. If False, no compression is used Default is False.
return_as_bytes_io – If True, the HDF file is only created in memory and returned as a bytes stream. If False, the file is written to disk. Default is False.

Returns

The full file name or a bytes stream containing the HDF file.

Return type

str, io.BytesIO

save_as_mgf(directory: str, file_name: str, overwrite: bool = False, centroiding_window: int = 5, keep_n_most_abundant_peaks: int = - 1)[source]¶

Save profile spectra from this TimsTOF object as an mgf file.

Parameters

directory (str) – The directory where to save the mgf file.
file_name (str) – The file name of the mgf file.
overwrite (bool) – If True, an existing file is truncated. If False, nothing happens if a file already exists. Default is False.
centroiding_window (int) – The centroiding window to use. If 0, no centroiding is performed. Default is 5.
keep_n_most_abundant_peaks (int) – Keep the n most abundant peaks. If -1, all peaks are retained. Default is -1.

Returns

The full file name of the mgf file.

Return type

str

property scan_max_index¶

The maximum scan index.

Type: int

property tof_indices¶

np.uint32[:] : The tof indices.

Type: np.ndarray

property tof_max_index¶

The maximum tof index.

Type: int

use_calibrated_mz_values_as_default(use_calibrated_mz_values: int) → None[source]¶

Override the default mz_values with the global calibrated_mz_values.

Calibrated_mz_values will be calculated if they do not exist yet.

Parameters: use_calibrated_mz_values (int) – If not 0, the mz_values are overwritten with global calibrated_mz_values. If 1, calibration at the MS1 level is performed. If 2, calibration at the MS2 level is performed.

property version¶

AlphaTims version used to create this TimsTOF object.

Type: str

property zeroth_frame¶

A blank zeroth frame is present so frames are 1-indexed.

Type: bool

alphatims.bruker.add_intensity_to_bin(query_index: int, intensities: numpy.ndarray, parsed_indices: numpy.ndarray, intensity_bins: numpy.ndarray) → None[source]¶

Add the intensity of a query to the appropriate bin.

IMPORTANT NOTE: This function is decorated with alphatims.utils.pjit. The first argument is thus expected to be provided as an iterable containing ints instead of a single int.

Parameters

query_index (int) – The query whose intensity needs to be binned The first argument is thus expected to be provided as an iterable containing ints instead of a single int.
intensities (np.float64[:]) – An array with intensities that need to be binned.
parsed_indices (np.int64[:], np.int64[:, :]) – Description of parameter parsed_indices.
intensity_bins (np.float64[:]) – A buffer with intensity bins to which the current query will be added.

alphatims.bruker.calculate_dia_cycle_mask(dia_mz_cycle: numpy.ndarray, quad_slices: numpy.ndarray, dia_precursor_cycle: numpy.ndarray = None, precursor_slices: numpy.ndarray = None)[source]¶

Calculate a boolean mask for cyclic push indices satisfying queries.

Parameters

dia_mz_cycle (np.float64[:, 2]) – An array with (upper, lower) mz values of a DIA cycle (per push).
quad_slices (np.float64[:, 2]) – Each row of the array is assumed to be (lower_mz, upper_mz) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(quad_slices.ravel()) >= 0) = True).
dia_precursor_cycle (np.int64[:]) – An array with precursor indices of a DIA cycle (per push).
precursor_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(precursor_slices[:, :2].ravel()) >= 0) = True).

Returns

A mask that determines if a cyclic push index is valid given the requested slices.

Return type

np.bool_[:]

alphatims.bruker.centroid_spectra(index: int, spectrum_indptr: numpy.ndarray, spectrum_counts: numpy.ndarray, spectrum_tof_indices: numpy.ndarray, spectrum_intensity_values: numpy.ndarray, window_size: int)[source]¶

Smoothen and centroid a profile spectrum (inplace operation).

IMPORTANT NOTE: This function will overwrite all input arrays.

IMPORTANT NOTE: This function is decorated with alphatims.utils.pjit. The first argument is thus expected to be provided as an iterable containing ints instead of a single int.

Parameters

index (int) – The push index whose intensity_values and tof_indices will be centroided.
spectrum_indptr (np.int64[:]) – An index pointer array defining the (untrimmed) spectrum boundaries.
spectrum_counts (np. int64[:]) – The original array defining how many distinct tof indices each spectrum has.
spectrum_tof_indices (np.uint32[:]) – The original array containing tof indices.
spectrum_intensity_values (np.float64[:]) – The original array containing intensity values.
window_size (int) – The window size to use for smoothing and centroiding peaks.

alphatims.bruker.convert_slice_key_to_float_array(key)[source]¶

Convert a key to a slice float array.

NOTE: This function should only be used for QUAD or DETECTOR dimensions.

Parameters: key (slice, int, float, None, iterable) – The key that needs to be converted.
Returns: Each row represent a a (start, stop) slice.
Return type: np.float64[:, 2]
Raises: ValueError – When the key is an np.ndarray with more than 2 columns.

alphatims.bruker.convert_slice_key_to_int_array(data: alphatims.bruker.TimsTOF, key, dimension: str)[source]¶

Convert a key of a data dimension to a slice integer array.

Parameters

data (alphatims.bruker.TimsTOF) – The TimsTOF objext for which to get slices.
key (slice, int, float, None, iterable) – The key that needs to be converted.
dimension (str) – The dimension for which the key needs to be retrieved

Returns

Each row represent a a (start, stop, step) slice.

Return type

np.int64[:, 3]

Raises

ValueError – When the key contains elements other than int or float.
PrecursorFloatError – When trying to convert a quad float to precursor index.

alphatims.bruker.filter_indices(frame_slices: numpy.ndarray, scan_slices: numpy.ndarray, precursor_slices: numpy.ndarray, tof_slices: numpy.ndarray, quad_slices: numpy.ndarray, intensity_slices: numpy.ndarray, frame_max_index: int, scan_max_index: int, push_indptr: numpy.ndarray, precursor_indices: numpy.ndarray, quad_mz_values: numpy.ndarray, quad_indptr: numpy.ndarray, tof_indices: numpy.ndarray, intensities: numpy.ndarray)[source]¶

Filter raw indices by slices from all dimensions.

Parameters

frame_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(frame_slices[:, :2].ravel()) >= 0) = True).
scan_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(scan_slices[:, :2].ravel()) >= 0) = True).
precursor_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(precursor_slices[:, :2].ravel()) >= 0) = True).
tof_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(tof_slices[:, :2].ravel()) >= 0) = True).
quad_slices (np.float64[:, 2]) – Each row of the array is assumed to be (lower_mz, upper_mz) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(quad_slices.ravel()) >= 0) = True).
intensity_slices (np.float64[:, 2]) – Each row of the array is assumed to be (lower_mz, upper_mz) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(intensity_slices.ravel()) >= 0) = True).
frame_max_index (int) – The maximum frame index of a TimsTOF object.
scan_max_index (int) – The maximum scan index of a TimsTOF object.
push_indptr (np.int64[:]) – The self.push_indptr array of a TimsTOF object.
precursor_indices (np.int64[:]) – The self.precursor_indices array of a TimsTOF object.
quad_mz_values (np.float64[:, 2]) – The self.quad_mz_values array of a TimsTOF object.
quad_indptr (np.int64[:]) – The self.quad_indptr array of a TimsTOF object.
tof_indices (np.uint32[:]) – The self.tof_indices array of a TimsTOF object.
intensities (np.uint16[:]) – The self.intensity_values array of a TimsTOF object.

Returns

The raw indices that satisfy all the slices.

Return type

np.int64[:]

alphatims.bruker.filter_spectra_by_abundant_peaks(index: int, spectrum_indptr: numpy.ndarray, spectrum_counts: numpy.ndarray, spectrum_tof_indices: numpy.ndarray, spectrum_intensity_values: numpy.ndarray, keep_n_most_abundant_peaks: int)[source]¶

Filter a spectrum to retain only the most abundant peaks.

IMPORTANT NOTE: This function will overwrite all input arrays.

IMPORTANT NOTE: This function is decorated with alphatims.utils.pjit. The first argument is thus expected to be provided as an iterable containing ints instead of a single int.

Parameters

index (int) – The push index whose intensity_values and tof_indices will be centroided.
spectrum_indptr (np.int64[:]) – An index pointer array defining the (untrimmed) spectrum boundaries.
spectrum_counts (np. int64[:]) – The original array defining how many distinct tof indices each spectrum has.
spectrum_tof_indices (np.uint32[:]) – The original array containing tof indices.
spectrum_intensity_values (np.float64[:]) – The original array containing intensity values.
keep_n_most_abundant_peaks (int) – Keep only this many abundant peaks.

alphatims.bruker.filter_tof_to_csr(tof_slices: numpy.ndarray, push_indices: numpy.ndarray, tof_indices: numpy.ndarray, push_indptr: numpy.ndarray) → tuple[source]¶

Get a CSR-matrix with raw indices satisfying push indices and tof slices.

Parameters

tof_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(tof_slices[:, :2].ravel()) >= 0) = True).
push_indices (np.int64[:]) – The push indices from where to retrieve the TOF slices.
tof_indices (np.uint32[:]) – The self.tof_indices array of a TimsTOF object.
push_indptr (np.int64[:]) – The self.push_indptr array of a TimsTOF object.

Returns

(np.int64[ – An (indptr, values, columns) tuple, where indptr are push indices, values raw indices, and columns the tof_slices.

Return type

], np.int64[:], np.int64[:],)

alphatims.bruker.get_dia_push_indices(frame_slices: numpy.ndarray, scan_slices: numpy.ndarray, quad_slices: numpy.ndarray, scan_max_index: int, dia_mz_cycle: numpy.ndarray, dia_precursor_cycle: numpy.ndarray = None, precursor_slices: numpy.ndarray = None, zeroth_frame: bool = True)[source]¶

Filter DIA push indices by slices from LC, TIMS and QUAD.

Parameters

frame_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(frame_slices[:, :2].ravel()) >= 0) = True).
scan_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(scan_slices[:, :2].ravel()) >= 0) = True).
quad_slices (np.float64[:, 2]) – Each row of the array is assumed to be (lower_mz, upper_mz) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(quad_slices.ravel()) >= 0) = True).
scan_max_index (int) – The maximum scan index of a TimsTOF object.
dia_mz_cycle (np.float64[:, 2]) – An array with (upper, lower) mz values of a DIA cycle (per push).
dia_precursor_cycle (np.int64[:]) – An array with precursor indices of a DIA cycle (per push).
precursor_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(precursor_slices[:, :2].ravel()) >= 0) = True).
zeroth_frame (bool) – Indicates if a zeroth frame was used before a DIA cycle.

Returns

The raw push indices that satisfy all the slices.

Return type

np.int64[:]

alphatims.bruker.indptr_lookup(targets: numpy.ndarray, queries: numpy.ndarray, momentum_amplifier: int = 2)[source]¶

Find the indices of queries in targets.

This function is equivalent to “np.searchsorted(targets, queries, “right”) - 1”. By utilizing the fact that queries are also sorted, it is significantly faster though.

Parameters

targets (np.int64[:]) – A sorted list of index pointers where queries needs to be looked up.
queries (np.int64[:]) – A sorted list of queries whose index pointers needs to be looked up.
momentum_amplifier (int) – Factor to add momentum to linear searching, attempting to quickly discard empty range without hits. Invreasing it can speed up searching of queries if they are sparsely spread out in targets.

Returns

The indices of queries in targets.

Return type

np.int64[:]

alphatims.bruker.init_bruker_dll(bruker_dll_file_name: str = '/home/docs/checkouts/readthedocs.org/user_builds/alphatims/checkouts/latest/alphatims/ext/timsdata.so')[source]¶

Open a bruker.dll in Python.

Five functions are defined for this dll:

tims_open: [c_char_p, c_uint32] -> c_uint64

tims_close: [c_char_p, c_uint32] -> c_uint64

tims_read_scans_v2: [c_uint64, c_int64, c_uint32, c_uint32, c_void_p, c_uint32] -> c_uint32

tims_index_to_mz: [c_uint64, c_int64, POINTER(c_double), POINTER(c_double), c_uint32] -> None

tims_scannum_to_oneoverk0: Same as “tims_index_to_mz”

Parameters: bruker_dll_file_name (str) – The absolute path to the timsdata.dll. Default is alphatims.utils.BRUKER_DLL_FILE_NAME.
Returns: The Bruker dll library.
Return type: ctypes.cdll

alphatims.bruker.open_bruker_d_folder(bruker_d_folder_name: str, bruker_dll_file_name='/home/docs/checkouts/readthedocs.org/user_builds/alphatims/checkouts/latest/alphatims/ext/timsdata.so') → tuple[source]¶

A context manager for a bruker dll connection to a .d folder.

Parameters

bruker_d_folder_name (str) – The name of a Bruker .d folder.
bruker_dll_file_name (str, ctypes.cdll) – The path to Bruker’ timsdata.dll library. Alternatively, the library itself can be passed as argument. Default is alphatims.utils.BRUKER_DLL_FILE_NAME, which in itself is dependent on the OS.

Returns

The opened bruker dll and identifier of the .d folder.

Return type

tuple (ctypes.cdll, int).

alphatims.bruker.parse_decompressed_bruker_binary_type1(decompressed_bytes: bytes, scan_indices_: numpy.ndarray, tof_indices_: numpy.ndarray, intensities_: numpy.ndarray, scan_start: int, scan_index: int) → int[source]¶

Parse a Bruker binary scan buffer into tofs and intensities.

Parameters

decompressed_bytes (bytes) – A Bruker scan binary buffer that is already decompressed with lzf.
scan_indices (np.ndarray) – The scan_indices_ buffer array.
tof_indices (np.ndarray) – The tof_indices_ buffer array.
intensities (np.ndarray) – The intensities_ buffer array.
scan_start (int) – The offset where to start new tof_indices and intensity_values.
scan_index (int) – The scan index.

Returns

The number of peaks in this scan.

Return type

int

alphatims.bruker.parse_decompressed_bruker_binary_type2(decompressed_bytes: bytes) → tuple[source]¶

Parse a Bruker binary frame buffer into scans, tofs and intensities.

Parameters: decompressed_bytes (bytes) – A Bruker frame binary buffer that is already decompressed with pyzstd.
Returns: The scan_indices, tof_indices and intensities present in this binary array
Return type: tuple (np.uint32[:], np.uint32[:], np.uint32[:]).

alphatims.bruker.parse_keys(data: alphatims.bruker.TimsTOF, keys) → dict[source]¶

Convert different keys to a key dict with defined types.

NOTE: Negative slicing is not supported and all indiviudal keys are assumed to be sorted, disjunct and strictly increasing

Parameters

data (alphatims.bruker.TimsTOF) – The TimsTOF objext for which to get slices.
keys (tuple) – A tuple of at most 5 elemens, containing slices, ints, floats, Nones, and/or iterables. See alphatims.bruker.convert_slice_key_to_int_array and alphatims.bruker.convert_slice_key_to_float_array for more details.

Returns

The resulting dict always has the following items:

”frame_indices”: np.int64[:, 3]
”scan_indices”: np.int64[:, 3]
”tof_indices”: np.int64[:, 3]
”precursor_indices”: np.int64[:, 3]
”quad_values”: np.float64[:, 2]
”intensity_values”: np.float64[:, 2]

Return type

dict

alphatims.bruker.process_frame(frame_id: int, tdf_bin_file_name: str, tims_offset_values: numpy.ndarray, scan_indptr: numpy.ndarray, intensities: numpy.ndarray, tof_indices: numpy.ndarray, frame_indptr: numpy.ndarray, max_scan_count: int, compression_type: int, max_peaks_per_scan: int) → None[source]¶

Read and parse a frame directly from a Bruker .d.analysis.tdf_bin.

Parameters

frame_id (int) – The frame number that should be processed. Note that this is interpreted as 1-indixed instead of 0-indexed, so that it is compatible with Bruker.
tdf_bin_file_name (str) – The full file name of the SQL database “analysis.tdf_bin” in a Bruker .d folder.
tims_offset_values (np.int64[:]) – The offsets that indicate the starting indices of each frame in the binary. These are contained in the “TimsId” column of the frames table in “analysis.tdf_bin”.
scan_indptr (np.int64[:]) – A buffer containing zeros that can store the cumulative number of detections per scan. The size should be equal to max_scan_count * len(frames) + 1. A dummy 0-indexed frame is required to be present for len(frames). The last + 1 allows to explicitly interpret the end of a scan as the start of a subsequent scan.
intensities (np.uint16[:]) – A buffer that can store the intensities of all detections. It’s size can be determined by summing the “NumPeaks” column from the frames table in “analysis.tdf_bin”.
tof_indices (np.uint32[:]) – A buffer that can store the tof indices of all detections. It’s size can be determined by summing the “NumPeaks” column from the frames table in “analysis.tdf_bin”.
frame_indptr (np.int64[:]) – The cumulative sum of the number of detections per frame. The size should be equal to len(frames) + 1. A dummy 0-indexed frame is required to be present for len(frames). The last + 1 allows to explicitly interpret the end of a frame as the start of a subsequent frame.
max_scan_count (int) – The maximum number of scans a single frame can have.
compression_type (int) – The compression type. This must be either 1 or 2. Should be treieved from the global metadata.
max_peaks_per_scan (int) – The maximum number of peaks per scan. Should be retrieved from the global metadata.

alphatims.bruker.read_bruker_binary(frames: numpy.ndarray, bruker_d_folder_name: str, compression_type: int, max_peaks_per_scan: int, mmap_detector_events: Optional[bool] = None) → tuple[source]¶

Read all data from an “analysis.tdf_bin” of a Bruker .d folder.

Parameters

frames (pd.DataFrame) – The frames from the “analysis.tdf” SQL database of a Bruker .d folder. These can be acquired with e.g. alphatims.bruker.read_bruker_sql.
bruker_d_folder_name (str) – The full path to a Bruker .d folder.
compression_type (int) – The compression type. This must be either 1 or 2.
max_peaks_per_scan (int) – The maximum number of peaks per scan. Should be treieved from the global metadata.
mmap_detector_events (bool) – Do not save the intensity_values and tof_indices in memory, but use an mmap instead. Default is True

Returns

The scan_indptr, tof_indices and intensities.

Return type

tuple (np.int64[:], np.uint32[:], np.uint16[:]).

alphatims.bruker.read_bruker_sql(bruker_d_folder_name: str, add_zeroth_frame: bool = True, drop_polarity: bool = True, convert_polarity_to_int: bool = True) → tuple[source]¶

Read metadata, (fragment) frames and precursors from a Bruker .d folder.

Parameters

bruker_d_folder_name (str) – The name of a Bruker .d folder.
add_zeroth_frame (bool) – Bruker uses 1-indexing for frames. If True, a zeroth frame is added without any TOF detections to make Python simulate this 1-indexing. If False, frames are 0-indexed. Default is True.
drop_polarity (bool) – The polarity column of the frames table contains “+” or “-” and is not numerical. If True, the polarity column is dropped from the frames table. this ensures a fully numerical pd.DataFrame. If False, this column is kept, resulting in a pd.DataFrame with dtype=object. Default is True.
convert_polarity_to_int (bool) – Convert the polarity to int (-1 or +1). This allows to keep it in numerical form. This is ignored if the polarity is dropped. Default is True.

Returns

(str, dict, pd.DataFrame, pd.DataFrame, pd.DataFrame). The acquisition_mode, global_meta_data, frames, fragment_frames and precursors. For diaPASEF, precursors is None.

Return type

tuple

alphatims.bruker.set_precursor(precursor_index: int, offset_order: numpy.ndarray, precursor_offsets: numpy.ndarray, quad_indptr: numpy.ndarray, tof_indices: numpy.ndarray, intensities: numpy.ndarray, spectrum_tof_indices: numpy.ndarray, spectrum_intensity_values: numpy.ndarray, spectrum_indptr: numpy.ndarray, spectrum_counts: numpy.ndarray) → None[source]¶

Sum the intensities of all pushes belonging to a single precursor.

IMPORTANT NOTE: This function is decorated with alphatims.utils.pjit. The first argument is thus expected to be provided as an iterable containing ints instead of a single int.

Parameters

precursor_index (int) – The precursor index indicating which MS2 spectrum to determine.
offset_order (np.int64[:]) – The order of self.precursor_indices, obtained with np.argsort.
precursor_offsets (np.int64[:]) – An index pointer array for precursor offsets.
quad_indptr (np.int64[:]) – The self.quad_indptr array of a TimsTOF object.
tof_indices (np.uint32[:]) – The self.tof_indices array of a TimsTOF object.
intensities (np.uint16[:]) – The self.intensity_values array of a TimsTOF object.
spectrum_tof_indices (np.uint32[:]) – A buffer array to store tof indices of the new spectrum.
spectrum_intensity_values (np.float64[:]) – A buffer array to store intensity values of the new spectrum.
spectrum_indptr (np.int64[:]) – An index pointer array defining the original spectrum boundaries.
spectrum_counts (np. int64[:]) – An buffer array defining how many distinct tof indices the new spectrum has.

alphatims.bruker.trim_spectra(index: int, spectrum_tof_indices: numpy.ndarray, spectrum_intensity_values: numpy.ndarray, spectrum_indptr: numpy.ndarray, trimmed_spectrum_tof_indices: numpy.ndarray, trimmed_spectrum_intensity_values: numpy.ndarray, new_spectrum_indptr: numpy.ndarray) → None[source]¶

Trim remaining bytes after merging of multiple pushes.

IMPORTANT NOTE: This function is decorated with alphatims.utils.pjit. The first argument is thus expected to be provided as an iterable containing ints instead of a single int.

Parameters

index (int) – The push index whose intensity_values and tof_indices will be trimmed.
spectrum_tof_indices (np.uint32[:]) – The original array containing tof indices.
spectrum_intensity_values (np.float64[:]) – The original array containing intensity values.
spectrum_indptr (np.int64[:]) – An index pointer array defining the original spectrum boundaries.
trimmed_spectrum_tof_indices (np.uint32[:]) – A buffer array to store new tof indices.
trimmed_spectrum_intensity_values (np.float64[:]) – A buffer array to store new intensity values.
new_spectrum_indptr (np.int64[:]) – An index pointer array defining the trimmed spectrum boundaries.

alphatims.bruker.valid_precursor_index(precursor_index: int, precursor_slices: numpy.ndarray) → bool[source]¶

Check if a precursor index is included in the slices.

Parameters

precursor_index (int) – The precursor index to validate.
precursor_slices (np.int64[:, 3]) – Each row of the array is assumed to be a (start, stop, step) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(precursor_slices[:, :2].ravel()) >= 0) = True).

Returns

True if the precursor index is present in any of the slices. False otherwise.

Return type

bool

alphatims.bruker.valid_quad_mz_values(low_mz_value: float, high_mz_value: float, quad_slices: numpy.ndarray) → bool[source]¶

Check if the low and high quad mz values are included in the slices.

NOTE: Just a part of the quad range needs to overlap with a part of a single slice.

Parameters

low_mz_value (float) – The lower mz value of the current quad selection.
high_mz_value (float) – The upper mz value of the current quad selection.
quad_slices (np.float64[:, 2]) – Each row of the array is assumed to be (lower_mz, upper_mz) tuple. This array is assumed to be sorted, disjunct and strictly increasing (i.e. np.all(np.diff(quad_slices.ravel()) >= 0) = True).

Returns

True if some part of the quad overlaps with some part of some slice. False if there is no overlap in the range.

Return type

bool