Mass spectrometer isolation profile analyser
A method for characterising an isolation profile of a mass spectrometer includes obtaining data of an, or at least one ion species transmitted by a mass spectrometer forming an isolation profile of the mass spectrometer; normalizing the obtained data; providing the normalized data to a deep neural network trained using a plurality of previous isolation profiles; generating from the deep neural network a set of fit parameters of a curve representing a fit to the normalized data; and providing as an output, data representing the curve. The method may be used as part of a calibration procedure for the mass spectrometer.
1 . A method for characterising an isolation profile of a mass spectrometer, the method comprising the steps of:
obtaining data of an, or at least one ion species transmitted by a mass spectrometer forming an isolation profile of the mass spectrometer;
normalizing the obtained data;
providing the normalized data to a deep neural network trained using a plurality of previous isolation profiles;
generating from the deep neural network a set of fit parameters of a curve representing a fit to the normalized data;
using the fit parameters as the starting point for a curve fit function;
executing the curve fit function to generate a curve; and
providing as an output, data representing the curve.
2 . The method of claim 1 , wherein the mass spectrometer includes a quadrupole.
3 . The method of claim 1 further comprising the steps of:
measuring the full width at half height and/or centre of the generated curve.
4 . The method of claim 3 further comprising the steps of:
if the full width at half height and/or centre of the generated curve are outside of predetermined limits then:
adjusting one or more physical configurations of the mass spectrometer; and
repeating the steps of the characterisation of the isolation profile until the full width, half height and/or centre of the generated curve are within the predetermined limits.
5 . The method of claim 1 , wherein the curve fit function is a piecewise-defined function composed of three components having at least three separate parameters.
6 . The method of claim 5 , wherein the three components are based on left and right half-Gaussian functions separated by a line.
7 . The method of claim 1 , wherein the curve fit function includes any one or more functions of: Gaussian, polynomial of order greater than three, Sigmoid, Lorentzian and Pearson.
8 . The method of claim 1 , wherein the curve fit function is defined as:
f
(
x
)
=
{
x
<
c
-
w
2
,
he
-
(
x
-
x
l
)
2
2
σ
l
2
c
-
w
2
≤
x
≤
c
+
w
2
,
h
x
>
c
+
h
2
,
he
-
(
x
-
x
r
)
2
2
σ
r
2
where
x
l
=
c
-
w
2
,
x
r
=
c
+
w
2
and
σ
l
=
l
2
log
(
2
)
,
σ
r
=
r
2
log
(
2
)
and where h represents the height of the isolation profile, c is the centre of the isolation profile, l and r are the steepness of the edges of the isolation profile, w is the width of a line forming the top of the curve fit function and w+l+r is the width at half-height of the curve fit function.
9 . The method of claim 1 , further comprising the step of optimising the curve fit using a gradient descent procedure.
10 . The method of claim 1 , wherein the step of obtaining data of an, or at least one ion species transmitted by a mass spectrometer comprises generating the data by operating the mass spectrometer.
11 . The method of claim 1 , further comprising the step of identifying one or more portions of the data outside of the curve.
12 . The method of claim 11 further comprising the step of measuring an area corresponding to the identified one or more portions of data.
13 . The method of claim 12 further comprising the step of generating a ratio of the measured area to the area under the curve.
14 . The method of claim 13 , further comprising the step of cleaning the mass spectrometer when the ratio is above a threshold value.
15 . The method of claim 1 , further comprising the step of generating a quality metric of the mass spectrometer from the curve.
16 . The method of claim 1 , further comprising the step of generating a calibration value for the mass spectrometer based on the curve.
17 . A mass spectrometer, comprising:
a mass analyser;
a detector; and
means adapted to execute the steps of:
obtaining data of an, or at least one ion species transmitted by a mass spectrometer forming an isolation profile of the mass spectrometer;
normalizing the obtained data;
providing the normalized data to a deep neural network trained using a plurality of previous isolation profiles;
generating from the deep neural network a set of fit parameters of a curve representing a fit to the normalized data;
using the fit parameters as the starting point for a curve fit function;
executing the curve fit function to generate a curve; and
providing as an output, data representing the curve.
18 . The mass spectrometer of claim 17 , wherein the mass analyser includes a quadrupole and/or an Orbitrap™ mass analyser.
19 . A computer program product comprising instructions to cause a mass spectrometer to execute the steps of:
obtaining data of an, or at least one ion species transmitted by the mass spectrometer forming an isolation profile of the mass spectrometer;
normalizing the obtained data;
providing the normalized data to a deep neural network trained using a plurality of previous isolation profiles;
generating from the deep neural network a set of fit parameters of a curve representing a fit to the normalized data;
using the fit parameters as the starting point for a curve fit function;
executing the curve fit function to generate a curve; and
providing as an output, data representing the curve.