PASEF™ on a timsTOF Pro defines new performance standards for shotgun proteomics with dramatic improvements in MS/MS data acquisition rates and sensitivity

PASEF™ on a timsTOF Pro defines new performance 
standards for shotgun proteomics with dramatic 
improvements in MS/MS data acquisition rates 
and sensitivity

Trapped ion mobility spectrometry coupled with quadrupole time-of-flight mass 
spectrometry (timsTOF Pro) offers a unique dimension of characterization and  
separation in complex mixtures. The timsTOF Pro instrument also enables the previously 
introduced “Parallel Accumulation Serial Fragmentation” (PASEF) method.[1]

Keywords:  

Trapped ion mobility 

spectrometry (TIMS), 

Parallel Accumulation 

Serial Fragmentation 

(PASEF)

Authors:  Markus Lubeck1, Scarlet Beck1, Heiner Koch1, Stephanie Kaspar-Schoenefeld1, Niels Goedecke1, Oliver Raether1, Nicole Drechsler1, Michael Krause1,  

Florian Meier2, Jürgen Cox2, Matthias Mann2. 
1

Bruker Daltonik GmbH, Bremen, Germany;  

2

Max Planck Institute of Biochemistry, Martinsried, Germany.

In the first publication on PASEF, a 
proof of principle experiment was 
performed using direct infusion,  
which led to the prediction 

that PASEF could achieve 

dramatic improvements in 

MS 

/ 

MS data acquisition rates 

in shotgun proteomics experi-
ments with improved sensitivity, 
with the possibility to address 
>170,000 precursors in a single 
shotgun proteomics experiment. 

This predicted performance far 
exceeded the capabilities of 
instruments available for shotgun 
proteomics, providing results with 
greater specificity and biological  
relevance. In this application note, 

we demonstrate that with further 
hardware and software develop-
ments, the promise of PASEF has 
now been realized, and that data 
dependent shotgun proteomics 
experiments performed using PASEF 
can address 166,000 independent 
precursors from a HeLa digest using 
a 90 min gradient. The dual concen-
tration effects from the separations 
by chromatography and trapped ion 
mobility spectrometry provide a gain 
of at least a factor of 10 in sensitivity  
and using just 200 ng of HeLa digest 
is sufficient to exceed what previously 
required 1-2 µg of sample.[2] The 
166,000 addressed precursors result  
in more than 35,000 unique peptide  
identifications and 5,500 protein identi- 

fications from 200 ng of HeLa digest 
injected on column. The benefits of this 
approach for proteomics applications  
are demonstrated using a timsTOF 
Pro instrument (Bruker Daltonics) 
with the PASEF acquisition mode 
connected to a nanoElute UHPLC 
(Bruker Daltonics).

Introduction

Multi-dimensional separations are 
necessary to obtain a more complete 
and accurate view of the content of 
complex proteomics samples in both 
discovery and targeted (quantitative) 
workflows, and ion mobility has been 
used as an additional separation 
device in commercial mass spectro- 

meters since the 1960s. More recently 
trapped ion mobility spectrometry 
(TIMS) has been introduced and 

coupled to QTOF mass spectrometers. 
In TIMS, ions are propelled through 
a TIMS tunnel by a gas flow. An 
electric field traps each ion at the 
position where the push that it 
experiences from the gas flow 
matches the force of the electric 

field, resulting in separation by mobility, 
a function of their three-dimensional 
size and charge in the gas phase 
(Figure 1). Ramping down the electrical 
field allows a selective release of 
ions from the TIMS tunnel according 
to their ion mobility (Ω/z), where Ω 
is the ion’s collisional cross section 
(CCS). In contrast to traditional drift 

To mass analyzer, ions of highest mobility 
(greatest collisional cross-section) eluting first

v

g

E

Parallel accumulation

TIMS scan

Figure 1: TIMS operation mode: With the 
directed forces of gas flow, v

g, and electrical 

field, E, ions are accumulated and trapped 
according to mobility. Step-wise decrease 
of the electrical field permits serial elution of 
separated ions, which are then transferred 
to the QTOF. Using the parallel accumulation 
operation mode ions are eluted from the 
second part of the TIMS device while the 
next series of ions is introduced into the first 
part of the TIMS device.

tube separation by size-to-charge, the 
larger ions elute first, followed by ions 
of decreasing cross section. In this  
study, a timsTOF Pro instrument 

(Bruker Daltonics) was used, with 
the eluted ions further analyzed 
via quadrupole time-of-flight mass 
spectrometry (Figure 2). The spectra  
generated during the TIMS elution 
cycle is termed a TIMS scan. The 
two dimensional separation of ions  
(1/K

0 and m / z) summed across a TIMS 

scan can be visualized as a TIMS MS 
heat map, with mobility and m / z as the 
y and x axes, respectively (Figure 3). 

Likewise, particular ions may be 

isolated in three dimensions according 
to their liquid chromatography (LC) 
retention time, TIMS elution time and 
m / z (in this study, by a quadrupole 
filter) and subjected to collision 
induced dissociation. In the generated 

PASEF MS / MS heat map, fragments 
may be easily aligned with their 

precursors in the TIMS MS heat map, 
given that they appear at the same 
ion mobility (Figure 3). 

While this orthogonal separation can 
provide a deeper look into complex pro-
teomics samples, its utility for peptide 
identification may be challenged in 
many instruments due to insufficient 
sensitivity resulting from low ion 

abundance (leading to weak spectra 
from subsequent collision induced 
fragmentation), as well as high scan-
ning speeds required to be compatible 
with the typical 100 ms TIMS scan 
times. Through the implementation 
of the “Parallel Accumulation Serial 
Fragmentation” (PASEF) method on 
a fast scanning QTOF instrument, 
fragment ion spectra from isolated 
precursor ions may be acquired more 

rapidly without sacrificing spectral 
quality. By using a dual TIMS analyzer, 
ions may be nearly continuously 
queued for accumulation, sorting and 
elution by mobility, allowing a duty 
cycle near 100% (Figure 1). During 
the timeframe of each TIMS scan, the 
quadrupole switches its (mass) iso- 
lation position several times, enabling 
the collection of MS/MS spectra from 
multiple precursors (Figure 4). Spectra  
for low abundant precursors may be 
summed to improve signal intensities  
to increase both the number and 
confidence of peptide identifications.  
The use of ion mobility separation 
prior to MS / MS results in improved 
signal-to-noise ratios, as the targeted 
ion species is time focused and 

subsequently transferred to the 

collision cell only during the periods of  
time they elute from the TIMS device. 
Additionally, by virtue of this additional 

Second generation  
dual TIMS  analyzer

TIMS synchronized  
quadrupole mass filter

UHR-TOF with high  
efficiency technology

High speed  
collision cell

Figure 2. Ion optics of the timsTOF Pro instrument including a dual TIMS analyzer and QTOF mass spectrometer.

dimension of separation, precursors 
of the same m / z that co-elute from 
the LC may be distinguishable which 
is not possible with mass based 
selection alone.  

Two dimensional (m / z-mobility) data 
dependent precursor acquisition

Another challenge in fast data 

acquisition is the computational time 
required to determine the precursors 
that will be selected for MS/MS. 

Maximizing the number of quality 
MS/MS spectra, an essential step 
in successful shotgun proteomics 

experiments, requires consideration of 
relative ion intensities and resolutions, 
the latter of which is increased (thus 
increasing the number of viable 

precursors) with the second dimension 
of separation.  With higher speed data 
acquisition capabilities, lower intensity 
targets may be fragmented multiple  
times. A rather simple and rapid 

algorithmic approach[3] was developed 
to efficiently schedule the subsequent 
fragmentation scans based on 
the initial detection of precursor 
ions during the MS scan in both 
the m / z and mobility dimensions. 
Even in very complex proteomics 

samples, computational time does not 
exceed 1 ms, easily supporting high  
throughput data collection.

Experimental

A complex tryptic peptide mixture 
derived from HeLa cells was diluted 
with 0.1% formic acid (FA) in water 
to a concentration of 200 ng/µL. A 
nanoElute UHPLC (Bruker Daltonics)  
was coupled to the timsTOF Pro mass 
spectrometer. The peptide mixture 
(200 ng) was loaded onto a C18 column 
(25 cm X 75 µm 1.6 µm, IonOpticks, 
Australia). Chromatographic separation 
was carried out using a linear gradient  
of 2-37% of buffer B (0.1% FA in 
ACN) at a flow rate of 400 nL/min 
over 90 min. MS data was collected 
over a m / z range of 100 to 1700, and 
MS / MS range of 100 to 1700. During 
MS / MS data collection, each TIMS 
cycle was 1.1 s and included 1 MS + an 
average of 10 PASEF MS / MS scans. 
The acquired data were submitted 

to the MASCOT search engine for 
peptide identifications.

Rapid data collection...

As may be expected for this type of 
sample, the base peak chromatogram 
(BPC) (Figure 5, top) indicates an 
extremely complex collection of pep-
tide species. Considering one narrow 
time slice of the LC gradient (61.7 min), 
the power of the PASEF method is 
readily seen. During this example 

scan, 41 different precursors were 
fed into the scheduling algorithm and 
addressed by a total of 10 PASEF 
MS / MS scans. By examination of a 
single TIMS elution (time) heat map 
(Figure 5, bottom left), rapid quadru- 
pole isolation window switching for 
the first 12 precursors is evident. 
Additional precursors are targeted in 
subsequent segments of the elution 
gradient. All precursors analyzed 
within this example PASEF cycle are 
indicated (Figure 5, bottom right), 
including the resequencing of targets 
of lower abundance. Within this single 
PASEF cycle (1.1 s), the equivalent  
of 119 separate MS 

/ 

MS events 

(if collected without PASEF on a 

timsTOF mass spectrometer) were 
conducted.

…while maintaining necessary 
sensitivity

Important proteomic features are 
often hidden within low intensity 
peaks of complex samples, peaks 
which might not be targeted by 

instruments with slower MS 

/ 

MS 

acquisition rates. The sensitivity 
(through the TIMS time focusing and  
the ability to sum spectra of low 

abundance targets) and separation 
power (through differences in ion 
mobility) of the PASEF approach is 

Figure 3. Illustration of TIMS MS (left) and PASEF MS / MS (right) heat maps resulting from TIMS separations. Although the data generated may be very  
complex, the characteristic ion mobility of targeted precursors and their fragments facilitates easy data alignment. Note that precursors of the same m / z  
(yellow and green) may be clearly separated by their mobility.

1/K

0

[Vs/cm2]

1/K

0

[Vs/cm2]

m / z

m / z

PASEF MS/MS Heat Map

TIMS MS Heat Map

Figure 5. LC MS/MS run, 200 ng HeLa digest,  
90 min gradient.   A : Base peak chromatogram. 

B : TIMS MS heat map at 61.7 min, with targeted  

precursors indicated. Bottom left, first PASEF 
MS/MS analysis event, precursors circled in 
orange. Bottom right, full PASEF MS/MS cycle, 
with each of the ten colored lines indicating 
targets from one TIMS elution segment. The 
most abundant precursors were sequenced 
in previous scan cycles and were dynamically 
excluded from re-sequencing.

m / z

Time (minutes)

10

40

70

30

60

90

20

50

80

100

1/K

0

400

1.4

0.8

0.9

1.0

1.1

1.2

1.3

1000

800

1400

600

1200

A

B

1/K

0

400

1.4

0.8

0.9

1.0

1.1

1.2

1.3

1000

800

1400

600

1200

Figure 4. Illustration of the PASEF method in 
comparison with the standard TIMS MS / MS  
operation mode, with the same 100 ms times-
cale. Using PASEF (lower panel), the quadrupole  
switches its isolation position several times during 
each ion mobility scan, with multiple precursors 
isolated and subsequently transferred to the 
collision cell. In contrast, with the standard  
TIMS MS / MS approach (upper panel), only one 
precursor from each TIMS scan is selected.

TIMS

T

IMS

M

S

 / 

M

S

P

ar

all

el

 A

cc

u

m

u

la

ti

o

n

S

eri

al

 F

ra

g

m

en

ta

tio

n

Quadrupole

CID

Spectrum Scan Time

100 

ms

m / z

m / z

m / z

m / z

100 

ms

m / z

TIMS tunnel

TIMS tunnel

TIMS tunnel

TIMS tunnel

TIMS tunnel

m / z

m / z

m / z

m / z

m / z

Figure 6. Collection of precursors at m / z  
887.96, 888.94, and 889.78 indicated on a TIMS 
MS heat map, with the number of summed 
MS / MS spectra for each indicated. The base 
peak chromatogram for this narrow m / z range  
is shown in the bottom of the heat map.

Figure 7. Example of low intensity peptide 
matches from HeLa cell analysis via PASEF 
differ by only 2 Da. While their isotopic patterns 
overlap in m / z, they are separated in the mobility 
dimension and the PASEF MS / MS results yield 
confident identifications.

1/K

0

M

ob

ili

ty

1.4

0.8

0.7

0.9

1.0

1.1

1.2

1.3

890

885

900

905 m / z

800

895

Precursor
Mass: 

m / z 888.94  2+

CCS: 

464.71 Å2

Intensity: 21219

Fragment
Repititions: 3
Mascot Ions Score: 62
Expectation Value:  6.7e-7

Precursor
Mass: 

m / z 889.78  3+

CCS: 

614.83 Å2

Intensity: 11844

Fragment
Repititions: 4
Mascot Ions Score: 72
Expectation Value:  6.9e-8

Precursor
Mass: 

m / z 887.96  2+

CCS: 

498.32 Å2

Intensity: 10355

Fragment
Repititions: 5
Mascot Ions Score: 35
Expectation Value:  3.0e-5

demonstrated in Figure 6. Summed 
MS 

/ 

MS spectra from three low 

intensity precursors of similar m / z in 
a dense region of the spectrum were 
readily identified with good sequence 
coverage, high MASCOT scores and 
low expectation values (probabilities 
to be a random match). Clearly, high 
performance of the (subsequent) 
QTOF is also critical, and the timsTOF 
Pro mass spectrometer provides high 
resolution, ppm accurate mass and high 
isotopic fidelity (True Isotopic Pattern, 
or TIP™). The benefit of this combined 
approach for shotgun proteomics 

applications is clearly indicated by 
the increase in confident peptide 
identifications (Figure 8). In the same 
sample (200 ng HeLa digest, eluted 
from the LC with a 90 min gradient), 
nearly twice as many peptides were 
identified using the PASEF approach 
as compared to analysis without 
trapped ion mobility spectrometry 
separation (using the same instru-
mentation, however, with the TIMS 
device switched off). 

7A

7A

7B

7B

7C

7C

LC-Settings

LC-System

nanoElute UHPLC system (Bruker Daltonics)

Column

25 cm X 75 µm 1.6 µm C18 column  
(IonOpticks, Australia)

LC flow rate

400 nL/min

LC elution conditions

2% to 37% (100% ACN, 0.1% FA)

Column oven (Sonation) temperature 50°C

Source

CaptiveSpray Ion Source (Bruker Daltonics)

Ionization

ESI (+)

Table 1 and 2: Instrumental Details

MS-Settings

MS-System

timsTOF Pro mass spectrometer (Bruker Daltonics)

TIMS elution

100 ms accumulation and ramp time (100% duty cycle)

Cycle time

1.1 s cycle with 1 MS + 10 PASEF MS / MS

Scan range

100 – 1700 m / z

Conclusions

“Parallel Accumulation Serial Fragmentation” (PASEF) on a timsTOF Pro 
instrument (Bruker Daltonics) has been shown to successfully deliver 
significantly higher speed and sensitivity in a data-dependent shotgun 
proteomics workflow, as demonstrated by the analysis of 166,000  
independent precursors from a HeLa digest using a 90 min gradient. By 
applying PASEF, the number of identifications could be increased to more 
than 40,000 unique peptide identifications and 5,500 protein identifications 
from only 200 ng of HeLa digest injected on column. The number of  
precursors able to be targeted with this approach is dramatically increased 
by virtue of the rapid sample separation by both chromatography 

and trapped ion mobility spectrometry. These exceptional acquisition  
rates also support improved sensitivity, as low abundant precursors  
may be targeted several times. This unique combination of features 
provides a new standard for shotgun proteomics perfromance, enabling 
researchers to find and identify more biologically relevant proteins. 

Discussion

The recent introduction of trapped 
ion mobility spectrometry (TIMS) 
has added an innovative dimension 
to quadrupole time-of-flight (QTOF). 
The incorporation of the previously 
described “Parallel Accumulation 
Serial Fragmentation” (PASEF)[1] 
method within Bruker’s proprietary 
timsTOF Pro instrument significantly  
magnifies these benefits. In this 
experiment, more than 160,000 
independent precursors from a HeLa 
digest, separated using a 90 min LC 
gradient, were addressed, with very 
fast MS / MS data acquisition rates.

The timsTOF Pro used for this study 
provides critical advances in separation 
and sensitivity. Trapped ion mobility 
spectrometry separation alone leads 
to improved spectral quality by virtue 
of the reduction of background noise. 
The use of two regions of separation 
within the TIMS device enables 100% 
duty cycle so that no ions are lost, ions 
are accumulated in the first region 
of the TIMS device, while they are 
scanned out into the QTOF from the 
second region. The quadrupole of the 
timsTOF Pro can switch its isolation 
window on the millisecond time scale, 
enabling the QTOF to address >10  
precursors in a typical 100 msec 

TIMS scan. The sensitivity is improved 
through the additional time focusing  
achieved by the TIMS, and additional  
sensitivity for low abundance precur-
sors is achieved by addressing the 
same precursor in multiple PASEF 
scans.

Total 
number  
of MS/MS 
spectra

Precursors 
targeted

Peptide  
spectrum  
matches

Unique peptide 
sequences, FDR 
< 1%, (Mascot 
with Percolator)

Protein 
groups 
(Mascot with 
Percolator)

TIMS-PASEF 610,000

166,000

73,000

39,000

5,200

TIMS off

105,000

105,000

29,000

20,000

3,200

Figure 8. Protein ID results. Approximately six times as many MS / MS spectra can be collected in the 
same time frame, with nearly double the number of unique peptide sequences and 1.5 times as many 
protein groups represented.

B

ru

ke

r D

al

to

ni

cs i

s c

on

tin

ua

ly i

m

pr

ov

in

g i

ts p

ro

du

ct

s a

nd r

es

er

ve

s t

he r

ig

ht 

to

 c

ha

ng

e sp

ec

ifi

ca

tion

s w

ith

out

 n

ot

ic

e.

 © B

ru

ke

r D

al

to

ni

cs 0

9

-2

01

7, L

C

M

S

-1

31

, 1

8

55

15

0, R

ev

. 0

1

For research use only. Not for use in diagnostic procedures.

[email protected] – www.bruker.com

References

[1] Meier, et al. (2015) Parallel Accumulation - Serial Fragmentation (PASEF): Multiplying 

Sequencing Speed and Sensitivity by Synchronized Scans in a Trapped Ion Mobility Device.  
J Proteome Res. 14(12):5378-87

[2] Beck, et al (2015) The Impact II, a Very High-Resolution Quadrupole Time-of-Flight Instrument 

(QTOF) for Deep Shotgun Proteomics. Mol Cell Proteomics. 14(7):2014-29.

[3] Lubeck et al. Evaluation of a peptide selection algorithm for trapped ion mobility with parallel 

accumulation serial fragmentation (tims PASEF) on a QTOF instrument. Poster presented at: 
65th ASMS Conference on Mass Spectrometry and Allied Topics, June 6, 2017; Indianapolis, IN.

Th

e

N

ew

 S

ta

nd

ar

d f

or Shotgun

 P

ro

te

o

m

ic

s

Bruker Daltonics GmbH & Co. KG

Bremen · Germany
Phone +49 (0)421-2205-0

Bruker Scientific LLC

Billerica, MA · USA 
Phone +1 (978) 663-3660