Protocol Citation: Joanna Simpson, Scott Denham, Birgit Alsbjerg, Natalie Z M Homer 2025. Sensitive targeted analysis of salivary steroids by liquid chromatography mass spectrometry for studies of infertility . protocols.io https://dx.doi.org/10.17504/protocols.io.14egn9rzpl5d/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: February 11, 2025
Last Modified: March 13, 2025
Protocol Integer ID: 119903
Keywords: steroid profiling, saliva, LC-MS/MS, assisted reproductive technologies, progesterone, IVF, ART, measuring progesterone, assessment of progesterone level, analysis of progesterone, attractive for progesterone measurement, measurement of the salivary steroid profile, progesterone measurement, analysis of salivary steroid, salivary steroid hormone, profiles multiple salivary steroid hormone, progesterone level, low blood progesterone in the luteal phase result, salivary steroid profile, multiple salivary steroid hormone, low blood progesterone, including progesterone, progesterone, free progesterone in women, superior for steroid analysis, steroid analysis, estrone in human saliva sample, liquid chromatography mass spectrometry for study, free progesterone, liquid chromatography mass spectrometry, targeted liquid chromatography mass spectrometry, salivary steroid, liquid chromatography tandem mass spectrometry, targeted liquid chromatography tandem mass spectrometry, human saliva sample, saliva sample, mass spectr
Funders Acknowledgements:
Rosetrees Trust
Grant ID: Seedcorn2023\100344
Abstract
Progesterone plays a key role in implantation and early pregnancy (Mesen et al, 2015) and studies have shown that low blood progesterone in the luteal phase result in lower pregnancy rates and an increased pregnancy loss rate following fresh and frozen embryo transfer (Alpcetin et al, 2025). As such, infertility treatments require assessment of progesterone levels to determine success, typically measured in serum following a blood-draw.
It is important to develop treatment regimens with high reproductive outcome, that are safe and painless, and minimal time.To ease discomfort and clinic visit burden then non-invasive sampling using saliva is attractive for progesterone measurement. If progesterone levels can be measured and demonstrate a constant level during daytime this could improve patient care during infertility treatment.
Methods for measuring progesterone and sex steroids are commonly immunoassay based, which can be affected by cross-reactivity and imprecision at low levels. However, mass spectrometry based methods are superior for steroid analysis in saliva (Brouillard et al, 2025), enhanced by the ability to multiplex multi-steroid analysis without issues of sensitivity and cross-reactivity (Handelsman, 2013).
In order to assess the feasibility of using saliva to profile free progesterone in women undergoing infertility treatments then a sensitive bioanalytical method that reliably measures salivary steroid hormones is needed. We developed an automated extraction and targeted liquid chromatography tandem mass spectrometry (LC-MS/MS) method that profiles multiple salivary steroid hormones, including progesterone. This method had suitable concentration ranges that allowed us to apply to saliva samples collected over a 12 hour cycle from women during the mid-luteal phase of IVF and FET cycles.
This protocol describes the extraction and targeted liquid chromatography mass spectrometry (LC-MS/MS) analysis of progesterone, 17alpha-hydroxyprogesterone, cortisol, cortisone, aldosterone, testosterone, androstenedione, dehydroepiandrosterone, 17β-estradiol and estrone in human saliva samples from women undergoing infertility treatment. It enables measurement of the salivary steroid profile. This targeted LC-MS/MS method was developed by adapting the method from Gregory et al, 2023, using updated instrumentation set up and improved isotopically labelled internal standards.
Saliva samples (200 μL) were enriched with isotopically labelled internal standards, diluted with water (0.1% formic acid v/v) and extracted alongside a (0.0025 - 400 ng) calibration curve, by automated 96-well supported liquid extraction (SLE), using dichloromethane and isopropanol as an organic solvent, on an Extrahera automated liquid handler by Biotage (Uppsala, Sweden). Extracted steroids were separated on an Acquity I-Class UPLC by Waters (UK) with gradient elution on a Kinetex C18 column (150 x 2.1 mm; 2.6 µm) by Phenomenex (UK) and a mobile phase of methanol and water (with 0.05 mM ammonium fluoride in water and methanol). The run time was 16 minutes, followed by analysis on a QTrap 6500+ mass spectrometer operated in multiple reaction mode in both positive and negative ionisation modes, scanning for 10 steroids and appropriate internal standards.
The amount of steroid in each sample was calculated using linear regression of the peak area ratio of the analytes to the isotopically labelled internal standard as determined by analysis of a calibration curve.
Ensure all training is up-to-date for operating the necessary laboratory instrumentation and equipment.
Materials
Consumables Table
A
B
C
D
Item
Supplier
Part no.
Quantity
1.75 mL glass vials with lids
Scientific Laboratory Supplies Ltd
TUB1200
10
7 mL glass vials with lids
Scientific Laboratory Supplies Ltd
TUB1220
5
Isolute SLE+ 400 96 well plate
Biotage
820-0400-P01
1
96-well plate sealing film
VWR
391-1250
1
Adhesive Plate Seal
Waters
186006336
1
Kinetex C18 (150 x 2.1 mm; 2.6 um)
Phenomenex
00F-4462-AN
1
Kinetex KrudKatcher, 0.5 um
Phenomenex
AFO-8497
1
Deep well 96 well collection plate
Biotage
121-5203
1
Deep well (2 mL) 96 well collection plate
Waters
186002482
1
Table M1 - Consumables for extraction and liquid chromatography separation of steroids
A
B
C
Item
Supplier
Article no.
Water (HPLC grade)
Fisher Scientific
C-10449380-X
Acetonitrile (LC-MS grade)
VWR
83640.320
Methanol (LC-MS grade)
VWR
83638.320
Water (LC-MS grade)
VWR
83645.320
Isopropanol (HPLC grade)
VWR
20880.320
Dichloromethane (HPLC grade)
Fisher Scientific
C-23373320-X
Cortisol
Sigma-Aldrich/Cerilliant
(C-106 ) 1 mg/mL in methanol (certified)
Cortisone
Sigma-Aldrich/Ceriliiant
(C-130) 100 µg/mL in methanol (Certified)
Aldosterone
Sigma-Aldrich/Cerilliant
A-096) 100 µg/mL in acetonitrile (certified)
Androstenedione
Sigma-Aldrich/Cerilliant
(A-075) 1 mg/mL in acetonitrile (certified)
Testosterone
Sigma-Aldrich/Cerilliant
(T-037) 1
mg/mL in acetonitrile (certified)
Dehydroepiandrosterone
Sigma-Aldrich/Cerilliant
(D-063) 1 mg/mL in methanol (certified)
17alpha-hydroxyprogesterone
Sigma-Aldrich/Cerilliant
(17OHP) H-085 1 mg/mL in methanol (certified)
Progesterone
Sigma-Aldrich/Cerilliant
(P-069) 1 mg/mL in acetonitrile (certified)
17beta-estradiol
Sigma-Aldrich/Cerilliant
(E-060) 1 mg/mL in acetonitrile (certified)
Estrone
Sigma-Aldrich/Cerilliant
(E-075) 1 mg/mL in methanol (certified)
13C3-Cortisol
Sigma-Aldrich/Cerilliant
(C-216) 100 µg/mL in methanol (certified)
13C3-cortisone
Sigma-Aldrich/Cerilliant
(C-160) 100 µg/mL in methanol (certfied)
13C3-aldosterone
Sigma-Aldrich/Cerilliant
(A-120) 10 µg/mL in acetonitrile (certified)
13C3-testosterone
Sigma-Aldrich/Cerilliant
(T-070) 100 ug/mL
in acetonitrile (certified)
13C3-androstenedione
Sigma-Aldrich/Cerilliant
(A-084) 100 ug/mL
in acetonitrile (certified)
d5-dehydroepiandrosterone
Sigma-Aldrich/Cerilliant
(D-064) 100 µg/mL in methanol
d9-progesterone
Sigma-Aldrich/Cerilliant
P-070 100 ug/mL
in acetonitrile
d8-17hydroxyprogesterone
Sigma-Aldrich/Cerilliant
(H-096) 100 µg/mL in methanol
13C3-estradiol
Sigma-Aldrich/Cerilliant
(E-073) 100 µg/mL in acetonitrile (certified)
13C3-estrone
Sigma-Aldrich/Cerilliant
E-108) 100 µg/mL in methanol (certified)
Ammonium Fluoride
Fisher Scientific
Table M2 - Chemicals and Analytical Standards
Solutions Required
0.1% formic acid (aq) (200 mL) Make up to 200 mL with Water (HPLC grade). Mix thoroughly.
98:2 Dichloromethane:Isopropanol (1 L) - Add 20 mL Isopropanol (HPLC grade) to 980 mL Dichloromethane (HPLC grade). Mix thoroughly.
Methanol (HPLC grade): for preparation of calibration standard/internal standard dilutions.
Water (HPLC grade): for preparation of calibration standards.
70:30 Water:Methanol (100 mL) - Add 30 mL methanol (LC-MS grade) to 70 mL water (LC-MS grade). Mix thoroughly.
Equipment Table
A
B
C
Item
Model
Supplier
Liquid Chromatography Pumps
I-Class UPLC
Waters
mass spectrometer
QTrap 6500+
AB Sciex
Gilson Repetman
Gilson Repetman
Gilson
Deepwell plate thermoshaker
TS-DW
Grant Scientific
Liquid handling robot
Extrahera
Biotage, Sweden
SPE Dry 96 dual evaporator
SPE Dry
Biotage, Sweden
Table M3 - Equipment required for automated extraction and LC-MS/MS steroid analysis
Troubleshooting
Safety warnings
Ensure risk assessments are up to date and that all local laboratory guidelines are followed for handling chemicals and biological samples.
Ethics statement
Ensure all human samples used in the analysis have been collected following ethical approval.
Before start
Prepare the automated robot for operation.
Prepare the liquid chromatography tandem mass spectrometer (LC-MS/MS) for operation and ensure sufficient mobile phase solutions and needle wash. Prime the solvents. Check that the chromatographic column is correctly installed and is not leaking when the mobile phase is pumping through.
Preparation of human saliva samples for extraction
Remove human saliva samples from the freezer and defrost on ice
Preparation of calibration standard stock solutions
Prepare a mixed stock of 10 steroids (Table M2) - progesterone, 17alpha-hydroxyprogesterone, cortisol, cortisone, aldosterone, testosterone, androstenedione, dehydroepiandrosterone, 17β-estradiol and estrone - by using 100 µg/mL stock solutions. Do this by adding 50 µL x 100 µg/mL P4, 50 µL x 100 µg/mL 17OHP4, 50 µL x 100 µg/mL F, 50 µL x 100 µg/mL E, 50 µL x 100 µg/mL Aldo, 50 µL x 100 µg/mL T, 50 µL x 100 µg/mL A4, 50 µL x 100 µg/mL DHEA, 50 µL x 100 µg/mL E2 and 50 µL x 100 µg/mL E1 + 500 µL methanol to give a 5 µg/mL stock.
Dilute the 5 µg/mL stock Mixed STOCK by 1:10 dilution (100 µL x 5 µg/mL + 900 µL methanol ) to give 500 ng/mL stock
Dilute the 500 ng/mL mixed STOCK by 1:10 dilution (100 µL x 500 ng/mL + 900 µL methanol ) to give 50 ng/mL stock
Dilute the 50 ng/mL mixed STOCK by 1:10 dilution (100 µL x 5 µg/mL + 900 µL methanol ) to give 5 ng/mL stock
Dilute the 5 ng/mL Mixed STOCK by 1:10 dilution (100 µL x 5 µg/mL + 900 µL methanol ) to give 500 pg/mL stock
Dilute the 500 pg/mL Mixed STOCK by 1:10 dilution (100 µL x 5 µg/mL + 900 µLmethanol ) to give 50 pg/mL stock
Preparation of internal standard solution
Prepare 100 µg/mL solutions of each isotopically labelled internal standard (Table M2) (d9-progesterone, d8-17a-hydroxyprogesterone, 13C3-cortisol, 13C3-cortisone, 13C3-aldosterone, 13C3-testosterone, 13C3-androstenedione, d5-dehydroepiandrosterone, 13C3-estrone and 3C3-estradiol) in methanol.
Prepare a 5 ng/mL Working Internal Standard solution by taking 10 µL x 5 µg/mL Int Std Mix + 1990 µL methanol.
Set up of supported liquid extraction of steroids from calibration standards and samples
Label a 2 mL deep well 96-well collection plate (Table M1). Label a Supported Liquid Extraction SLE400 plate with batch details. Label a 2 mL deep well 96-well collection plate (Waters).
Design and prepare batch of standards and saliva samples in Microsoft Excel template, following a column-wise plate map design as below (Table S1).
Preparation of calibration standard curve and samples
12
Prepare calibration standards directly into the 96-well deep well plate using the following table for volumes of each stock concentration, into a final volume of 200 µLwater.
A
B
C
D
Standard name
Amount (ng)
STD Mix Vol (uL)
Vol water (uL)
0 STD
0
0
200
0.00250 STD
0.00250
5 uL x 500 pg/mL
195
0.00500 STD
0.00500
10 uL x 500 pg/mL
190
0.01000 STD
0.0100
20 uL x 500 pg/mL
180
0.0250 STD
0.0250
5 uL x 5 ng/mL
195
0.0500 STD
0.0500
10 uL x 5 ng/mL
190
0.100 STD
0.100
20 uL x 5 ng/mL
180
0.250 STD
0.250
5 uL x 50 ng/mL
195
0.500 STD
0.500
10 uL x 50 ng/mL
190
1.00 STD
1.00
20 uL x 50 ng/mL
180
2.50 STD
2.50
5 uL x 500 ng/mL
195
5.00 STD
5.00
10 uL x 500 ng/mL
190
10.0 STD
10.0
20 uL x 500 ng/mL
180
25.0 STD
25.0
5 uL x 5 ug/mL
195
50.0 STD
50.0
10 uL x 5 ug/mL
190
100 STD
100.0
20 uL x 5 ug/mL
180
200 STD
250
5 uL x 50 ug/mL
195
400 STD
400.0
8 uL x 50 ug/mL
192
Table 2 - Calibration standard preparation table
13
Aliquot 200 µLsaliva sample into the correct well according to the plate map design.
Supported liquid extraction of steroids from calibration standards and saliva samples
14
Using a multi-step pipette enrich the plate containing calibration standards with WIS by adding 20 µLx 5 ng/mL Working Internal Standard into each calibration standard, including 0 std and each sample (human saliva), except for the double blank and solvent blank.
15
Using the Extrahera liquid handling robot, set up with the batch labelled SLE400 extraction plate and the deep well extraction plate, containing the calibration standards and samples. Programme Extrahera to aliquot200 µL 0.1% formic acid in water (v/v) into each well of the 96-well deep well plate containing the samples and standards.
16
Programme the Extrahera to transfer 400 µL of liquid from each well (containing sample and the diluent, into a 400 µL volume Supported Liquid Extraction plate (SLE400), pre-placed into the deck on the Extrahera, with a deep well Waters 2 mL deep well collection plate below, pre-labelled with the batch details and date of extraction.
17
Allow the diluted sample to adsorb onto the SLE extraction bed for 00:05:00 before eluting with 600 µL x 98:2 (v/v) dichloromethane/isopropanol and repeating twice more, each time collecting the eluent into the collection plate
18
Dry down the eluent collected into the 2 mL collection plate using the SPE Dry down for 96-well plates under nitrogen.
19
Resuspend in 100 µL x 70:30 water/methanol, seal the plate with a zone-free plate seal and shake on ThermoShaker for 00:05:00 at 300 rpm
5m
20
Place the plate in the autosampler for LC-MS/MS or store at-20 °Cuntil ready for analysis.
Steroid measurement by LC-MS/MS
21
Set up an acquisition batch in Analyst software using the electronic excel file of the calibration standards and sample list. Set to inject 20 µL per sample and use a method of chromatographic separation as described in steps 22 and 23 and mass spectrometer settings as outlined in steps 24 and 25.
22
Set up the liquid chromatography system and fit with a Phenomenex Krud Katcher and a Phenomenex 150 x 2.1 mm; 2.6 µm Kinetex C18 liquid chromatography column, using mobile phase A - water with 0.05 mM ammonium fluoride and mobile phase B - methanol with 0.05 mM ammonium fluoride at 0.3 mL/min and 50 °C diverting to the mass spectrometer at 0.2 mins and returning to waste at 15.9 mins
23
Set up chromatographic gradient as below (Table S3) with a run time of00:16:00 per sample
A
B
C
D
Time (min)
Flow (mL/min)
A (%)
B (%)
Initial
0.3
50
50
4
0.3
50
50
9
0.3
25
75
10
0.3
0
100
12
0.3
0
100
12.1
0.3
50
50
16.00
0.3
50
50
Table S3 - Chromatographic gradient details. A - water w/ 0.05 mM ammonium fluoride; B - methanol w/ 0.05 mM ammonium fluoride. 50oC. Kinetex C18 (150 x 2.1 mm; 2.6 µm)
24
Set up the mass spectrometer for Multiple Reaction Monitoring (MRM) method in positive mode, with electrospray ionisation as below, with divert of LC flow into the mass spectrometer set at 1 minute and 15.9 minutes.
A
B
Instrument
Sciex QTrap 6500+
Source, Ionisation Mode
IonDrive Turbo V Source, ESI
Scan Mode, Polarity
MRM, Positive and Negative
Resolution (Q1/Q3)
unit/unit
Mass range
Low mass
Pause Time
5.007 ms
Acquisition time
16.0 min
Delay time
0 sec
Curtain Gas (CUR) (N2)
30 units
Collision Gas (CAD) (N2)
Medium
IonSpray Voltage (IS) (Positive)
5500 V / -4500 V
Temperature (TEM)
600 °C
Ion Source Gas 1 (GS1) (Air)
40 units
Ion Source Gas 2 (GS2) (Air)
60 units
Entrance Potential (EP) (Positive)
10 V
Probe position (x – axis)
5
Probe position (y – axis)
2
Table S4 - Mass Spectrometry source settings for positive and negative ion electrospray ionsiation on QTrap 6500+ mass spectrometer
25
Set up the mass spectrometer to monitor for the following multiple reaction monitoring (MRM) transitions for each steroid and each isotopically labelled steroid in positive and negative mode (Table S5).
A
B
C
D
D
E
F
G
Q1 Mass (Da)
Q3 Mass (Da)
Scan time (msec)
Polarity
Steroid Name
DP (V)
CE (V)
CXP (V)
363.1
121.2
10
+
Cortisol 1
66
31
12
363.1
91.0
10
+
Cortisol 2
76
83
10
361.1
163.1
10
+
Cortisone 1
81
31
26
361.1
77.1
10
+
Cortisone 2
81
107
10
289.1
97.0
10
+
Testosterone 1
101
29
12
289.1
109.2
10
+
Testosterone 2
101
31
6
287.1
97.0
10
+
Androstenedione 1
61
27
14
287.1
78.9
10
+
Androstenedione 2
61
67
10
271.1
235.1
10
+
Dehydroepiandrosterone 1
106
17
12
271.1
188.1
10
+
Dehydroepiandrosterone 2
106
17
12
315.0
97.1
10
+
Progesterone 1
96
23
10
315.0
109.1
10
+
Progesterone 2
96
27
10
333.1
109.1
10
+
17a-hydroxyprogesterone 1
66
31
12
333.1
96.9
10
+
17a-hydroxyprogesterone 2
66
29
12
359.1
188.9
10
-
Aldosterone 1
-70
-24
-21
359.1
331
10
-
Aldosterone 2
-70
-22
-35
269.1
144.9
10
-
Estrone 1
-150
-48
-15
269.1
142.9
10
-
Estrone 2
-150
-70
-15
271.0
144.9
10
-
Estradiol 1
-110
-52
-21
271.0
182.9
10
-
Estradiol 2
-110
-52
-19
292.1
100.0
10
+
13C3-Testosterone
96
29
12
290.2
100.1
10
+
13C3-Androstenedione
31
27
12
294.1
258.2
10
+
d5-dehydroepiandrosterone
21
13
28
324.1
100.0
10
+
d9-progesterone
151
31
15
339.2
96.9
10
+
d8-17hydroxyprogesterone
66
29
12
367.2
121.1
10
+
13C3-cortisol
80
29
16
364.2
166.0
10
+
13C3-cortisone
81
31
26
362.0
192.0
10
-
13C3-aldosterone
-70
-24
-12
272.1
147.8
10
-
13C3-estrone
-110
-52
-21
274.0
147.9
10
-
13C3-estradiol
-110
-48
-29
Table S5 - Multiple reaction monitoring (MRM) settings for each steroid, including quantitative (1) and qualitative (2) ions for each steroid. DP - declustering potential, CE - collision energy, CXP - collision exit potential
26
Check the retention times of the steroids are as expected, as shown in the chromatogram in Table S6:
Expected result
Retention times; aldosterone at 2.6 mins, cortisol at 3.5 mins, cortisone at 2.9 mins, estrone at 7.2 mins, 17beta-estradiol at 7.0 mins, androstenedione at 6.9 mins, testosterone at 7.6 mins, dehydroepiandrosterone at 7.9 mins, 17alpha-hydroxyprogesterone at 8.0 mins andprogesterone at 9mins
27
Inject a mid-level standard. Check the chromatography and each steroid retention time is consistent with expected times and peak area response is as expected. Once satisfied then set the batch of samples to analyse, injecting 20 µL per sample.
Method specific data evaluation of LC-MS/MS data
28
Use the data analysis parameters to assess the peak area of the chromatograms for each steroid in the Steroid analytes and their assigned internal standards (Table S6)
A
B
C
D
Steroid Name
Abbreviation
Retention Time (min)
Internal Standard
Progesterone
P4
9.0
d9-P4
Cortisol
F
3.5
13C3F
Cortisone
E
2.9
13C3E
Androstenedione
A4
6.9
13C3A4
Testosterone
T
7.6
13C3T
Dehydroepiandorsterone
DHEA
7.9
d5-DHEA
Aldosterone
Aldo
2.6
13C3-Aldo
Estrone
E1
7.2
13C3-E1
17beta-estradiol
E2
7.0
13C3-E2
17alpha-hydroxyprogesterone
17OHP4
8.0
d817OHP4
Internal Standards
13C3-cortisol
13C3F
3.5
Int Std
13C3-cortisone
13C3E
2.8
Int Std
13C3-Androstenedione
13C3A4
6.9
Int Std
13C3-Testosterone
13C3T
7.6
Int Std
d9-progesterone
d9P4
8.9
Int Std
d8-17hydroxyprogesterone
d817OHP4
7.9
Int Std
Table S6 - Method specific summary of retention time and specific internal standard of the steroids
Data Evaluation of steroid profiling LC-MS/MS data
29
Use MultiQuant software and Microsoft Excel to evaluate the LC-MS/MS steroid profiling data, by defining calibration standard levels, ensuring accuracy of the calibration standards and linear regression > 0.99. Use the Table above, to calculate the concentration of steroids in each sample, as detailed in the 'MultiQuant and Excel' protocol below. Remember to account for the volume of sample extracted and express as ng/mL.
Arik Alpcetin, S. I., Ince, O., Akcay, B., Cevher Akdulum, M. F., Demirdag, E., Erdem, A., & Erdem, M. (2025). Comparison of Individualized Rescue Luteal Phase Support Strategies with Vaginal and Combined Vaginal & Subcutaneous Progesterone Administration in Artificial Frozen-Thawed Blastocyst Embryo Transfer Cycles Based on Serum Progesterone levels. Frontiers in endocrinology, 15, 1503008. https://doi.org/10.3389/fendo.2024.1503008
Brouillard, A., Davignon, L. M., Cernik, R., Giguère, C. É., Findlay, H., Juster, R. P., Lupien, S. J., & Marin, M. F. (2025). Comparing immunoassay and mass spectrometry techniques for salivary sex hormone analysis. Psychoneuroendocrinology, 174, 107379. https://doi.org/10.1016/j.psyneuen.2025.107379
Gregory, S., Denham, S. G., Lee, P., Simpson, J. P., & Homer, N. Z. M. (2023). Using LC-MS/MS to Determine Salivary Steroid Reference Intervals in a European Older Adult Population. Metabolites, 13(2), 265. https://doi.org/10.3390/metabo13020265
Handelsman, D. J., & Wartofsky, L. (2013). Requirement for mass spectrometry sex steroid assays in the Journal of Clinical Endocrinology and Metabolism. The Journal of clinical endocrinology and metabolism, 98(10), 3971–3973. https://doi.org/10.1210/jc.2013-3375
Mesen, T. B., & Young, S. L. (2015). Progesterone and the luteal phase: a requisite to reproduction. Obstetrics and gynecology clinics of North America, 42(1), 135–151. https://doi.org/10.1016/j.ogc.2014.10.003