Non-Targeted Analysis at the U.S.
Environmental Protection ^
Environmental Protection Agency
Elin M. Ulrich
U.S. Environmental Protection Agency
Center for Computational Toxicology and Exposure
Research Triangle Park, NC USA
QQOQO
¦OOQa
OV1II
ollll
ooo
The views expressed in this presentation are those of the author(s) and do not necessarily represent the views or policies of the U.S. Environmental Protection Agency.
Dust samples were determined to be environmental in nature and not human subjects research. .. . .
. _ _r n_ i_ _i r-\ i i. 1 J l\l a\/qmhor 1 Jll
Office of Research and Development
November 1, 2019
-------
EPA NTA Research Contributors
4- ORD: Angela Batt, Scott Clifton, Kathy Coutros, Chris Grulke, Chris Fuller, Kristin Isaacs, Hannah
Liberatore, Charles Lowe, James McCord, Kelsey Miller, Jeff Minucci, Seth Newton, Katherine Phillips,
Tom Purucker, Ann Richard, Charlita Rosal, Jon Sobus, Mark Strynar, Adam Swank, Elin Ulrich, Ariel
Wallace, John Wambaugh, John Washington, Antony Williams
ORAU/ORISE/ASPPH: Hussein Al-Ghoul, Alex Chao, Andrew Eicher, J a rod Grossman, Johnsie Lang,
Sarah Laughlin-Toth, Jeremy Leonard, Kamel Mansouri, Aurelie Marcotte, Andrew McEachran, Dawn
Mills, Marie Russell, Randolph Singh,
Nelson Yeung
Contracts: EvoTec, General Dynamics
Information Technology
ENTACT was supported by EPA Stage
1 -3 Pathfinder Innovation Project
"Building a Network to Measure the
Totality of Chemical Exposures."
-------
Outline
-f EPA's Non-Targeted Analysis Collaborative Trial (ENTACT)
+ About NTA and ENTACT
-I- Initial results
+ APCI vs ESI
+ Predicted mass spectra
4- Applications of NTA to house dust
Existing tools and future plans for UVCBs
4- Benchmarks and Publications for Non-Targeted Analysis
-------
What is Non-Targeted Analysis?
®e®
Targeted Analysis
Standards, calibration curves
-f Suspect Screening Analysis (SSA)
Lists of compounds
Non-Targeted Analysis (NTA)
MS first principles
-------
1H
2H
12Q
13Q
14N
15N
16Q
17Q
18Q
19p
32S
33S
34S
36S
How does High Resolution MS work?
Natural
Abundance
Exact Mass
99.9885%
0.0115%
98.93%
1.07%
99.632%
0.368%
99.757%
0.038%
0.205%
100%
94.93%
0.76%
4.29%
0.02%
75.78%
24.22%
1.007825
2.014102
12.000000
13.003355
14.003074
15.000109
15.994915
16.999131
17.999159
18.998403
31.972072
32.971459
33.967868
35.967079
34.968853
36.965903
Example: Fipronil
Molecular Formula: C^F^C^FqIS^OS
Monoisotopic Mass: 435.938706
= (12.0000*12 Carbon) + (1.007825*4 Hydrogen) +
(34.968853*2 Chlorine) + (18.998403*6 Fluorine) +
(14.003074*4 Nitrogen) + (15.994915*1 Oxygen) +
(31.972072*1 Sulfur)
NH2 CI
100
-------
Non-Targeted Analysis Workflow
Experimental Acquisition Database & Library Matching Data Analysis & Computational Tools
MS1 Feature Table
Filtered Feature Table
Chemical Candidate Table
Chemical Database
MS2
Acquisition
MS1
Acquisition
MS2 Reference
Matches
MS2 Exported .mgf Files
MS2 in silico Matches
Reference MS2 Spectra
in siiico MS2 Spectra
Sample Extracts
LC-QTOF/MS
DB MS-Ready Structures
Aggregated Match Table
Reference MS-Ready
Mass/Formulae
Analytical Instruments
Chemical Databases
Computational Tools
High resolution accurate mass, mass spectrometry (QToF, Orbitrap)
CompTox Chemicals Dashboard, MassBank, PubChem
CPDat, media and retention time prediction, MetFrag, R/Python tools
-------
EPA's Non-Targeted Analysis Collaborative Trial
k Characterize current method performance characteristics (e.g., % true/false positives)
b Establish performance benchmarks and benchmark methods for SSA and NTA
b Develop reporting standards for studies using SSA and NTA methods
Increase compounds/spectra available
in reference libraries (in-house and
publicly available)
And so much more...
100000000 a
93552981'
10000000
1000000
100000
10000
1000
100
Peisl BYL, Schymanski EL, Wilmes P (2018) Anal. Chim. Acta
1037:13-27. https://doi.Org/10.1016/i.aca.2017.12.034
59000000
1200000
_ 775500
* * I 306622
747000 707000 x
2160803
574826
CD
Q
E
c
O
Compound databases
599700
A
74461
267376
72268
26500
44076
28000
A A
2200 A 1380814034
* _ 6599
3500
1000
a
5
CD
10000 8853
5912 5465
6748
4167
A
5061
92
u_
A
1200
o
c/)
2
CO
CL
z
CD
A
778
CD
a
A
3308
to
CO
<
EI-MS databases
MS/MS databases
a Compounds/structures - Spectra
-------
ENTACT Sample Overview
Part 1. Ten ToxCast mixtures Part 2. Three standard exposure relevant extracts
95, 185 or 365 substances/mixture Unaltered Fortified
>
%
""f
NISTSRM 1957-
Organic Contaminants in Non-fortified Human Serum
Part 3. Individual ToxCast standards
1,269 ENTACT; 4,685 ToxCast all
Oregon State University-
Outdoor air exposed silicone wrist-bands
9
9
«
NIST SRM 2585-
Organic Contaminants in House Dust
tfrgani
Hous
o
©
J
-------
ENTACT Mixtures- Brainchild of C. Grulke
400
(/)
350
CO
o
E
300
CD
sz
O
»¦
250
o
i_
0
200
E
3
150
100
50
0
5 NTA method "controls"
Grade A replicate 90 set
Grade A test substances
Grade A isomers & isobaric compounds
Challenge set of isobaric & <80% purity
10 Prepared Mixtures:
1,939 total spiked substances
1,269 unique substances:
1 -> spiked 11 times
4 -> spiked 10 times
-> spiked 4 times
-> spiked 3 times
388 -> spiked 2 times
786 -> spiked 1 time
8
10
Mixture Number
UI rich EM, et al. (2019) ABC 411:853-866. doi: 10.1007/s00216-018-1435-6
-------
Exposure Sample Prep
SRM 2585 dust
SRM 1957 serum
Silicone bands
Preparation
Quantity
extracted
Extraction and
final solvent
Extraction
method
Cleanup
Evaporation
Final samples
Blank
NA
3 g dust unaltered
3 g dust spiked
Methanol
Vortex, sonication,
centrifuge
LC-Si SPE cartridges
N2 blowdown to 15 mL
final volume
400 |jL in 30 vials
Method blank
Reconstituted 2 vials with 10.7
mL Dl water, shaking, vortexing
9.75 mL unaltered
9.75 mL spiked
Cold acetonitrile, formic acid
Protein crash, vortex,
centrifuge
NA
N2 blowdown to 19.5 mL final
volume
400 |jL in 30 vials
Method blank
Rinse, conditioning,
deployment in semi-rural
outdoor environment 18 days
8 bands unaltered
8 bands spiked
Ethyl acetate, 13 surrogates
added
Orbital shaker
NA
TurboVap to 12 mL final volume
400 |jL in 30 vials
Solvent blank
-------
ENTACT Instrument Methods: GC + Other
Lab # Chromatography
Mobile phase
MS type
MS/MS
1
Agilent GCxGC, Restek Rxi-5ms (30 m x 0.25 mm x
0.25 |jm) + Restek Rxi-17Sil MS (0.6 m x 0.25 mm x
0.25 |jnri)
Helium
Leco HRT+ ToF in El and
CI for confirmation
NA
2
Agilent GC, Agilent J&W VF-5MS (30 m x 0.25 mm x
0.25 |jnri)
Helium
Agilent Triple Quad in El
NA
3
Agilent GC, Agilent HP-5ms Ultra Inert (30 m x 0.25
mm x 0.25 |jm)
Helium
Agilent ToF in El
NA
4
Agilent GCxGC, Restek Rtx-5MS (35 m x 0.25 mm x
0.25 |jm)+Restek Rxi-17 (0.79 m x 0.1 mm x 0.1 |jm)
Helium
Leco Pegasus 4D
NA
9b
Direct infusion
NA
Thermo Velos Pro +21T
FT-ICR in ESI +/-
NA
Ulrich EM, et al. (2019) ABC 411:853-866. doi:10.1007/s00216-018-1435-6
-------
ENTACT Instrument Methods: LC-ToF
Lab#
Chromatography
Mobile phase
MS type
MS/MS
5
Agilent LC, Agilent Zorbax Eclipse
Plus C8 (50 x 2.1 mm, 1.8 |jm)
Water, methanol, ammonium
formate (AF)
Agilent 6530 QToF, ESI +/-
Data dependent
10, 20, 40 collision
6
Agilent LC, ZORBAX Eclipse Plus
C18 (100 x 2.1 mm, 1.8 |jm)
Water, methanol, ammonium
acetate, acetic acid
Agilent 6530 QToF, ESI +/-
Data dependent
10, 20, 40 collision
7
Agilent LC, Agilent Zorbax RRHD
Eclipse Plus C18 (150 x 2.1 mm,
1.8 |jm)
Water, acetonitrile, formic
acid (FA)
Agilent 6550 QToF, ESI +/-
Data dependent
10, 20, 40 collision
8
Agilent LC, Agilent Zorbax Eclipse
Plus C18 (100 x 2.1 mm, 1.8 |jm)
Water, methanol, ammonium
acetate
Agilent 6550 QToF, ESI +/-
Data dependent
10, 20, 40 collision
9a
Direct infusion
NA
Agilent 6560 QToF,
nESI +/-&APPI +/-
NA, but drift tube ion
mobility spectrometry
10
Dionex LC, Waters XSelect HSS T3
(150 x 3 mm, 3.5 |jm)
Water, acetonitrile, FA
Bruker maXis II TOF, ESI+,
Bruker maXis II UHR QToF
in ESI+
Data dependent
35 collision energy
11
Waters LC, ACQUITY UPLC BEH
C18 (50 x 2.1 mm,1.7 |jm)
Water, methanol, AF
Waters Xevo G2-XS QToF
in ESI +/-
Data independent,
Low = 4,
High = ramp 10-45
Ulrich EM, et al. (2019) ABC 411:853-866. doi:10.1007/s00216-018-1435-6
-------
ENTACT Instrument Methods: LC-Orbitrap
Lab # Chromatography
Mobile phase
MS type
MS/MS
12a
Direct infusion
NA
Thermo Orbitrap Elite in
nESI +/-
NA
12b
Thermo LC, Thermo Accucore C30
(150 x 2.1 mm, 2.6 |jm)
Water, acetonitrile,
isopropanol, AF, FA
Thermo Q Exactive in
HESI +/-
Data dependent
30, 60 collision
13
Dionex LC, MAC-MOD Analytical
ACE Excel C18-PFP (125 x 3 mm, 2
^m)
Water, acetonitrile, FA or
ammonium hydroxide
Thermo Q Exactive in ESI
+/-, APCI +/-
Data dependent
15, 30, 45 collision
14
Dionex LC, Waters XBridge BEH C18
(50 x2.1 mm, 3.5 |jm)
Water, methanol, FA
Thermo Q Exactive in
HESI +/-
Data dependent, 50
collision varied by m/z
15
Thermo LC, Thermo Hypersil GOLD
aQ C18 Polar Endcapped
(100 x 2.1 mm, 1.9 |jm)
Water, acetonitrile, FA
Thermo Q Exactive in
HESI +/-
Data independent,
stepped 30
16
Dionex LC, Waters Atlantis T3
(150 x 3 mm, 3 |jm)
Water, methanol,
isopropanol, FA
Thermo Q Exactive Plus
in ESI +/-
Data Independent,
varies 15-120;
Dependent 20, 50, 90
17
Waters LC, Thermo Hypersil Gold aQ
C18 Polar Endcapped
Water, methanol, FA, AF
Thermo Q Exactive Plus
in HESI +/-, APCI +/-
NA
(200 x2.1 mm, 1.9 |jm)
Ulrich EM, et al. (2019) ABC 411:853-866. doi: 10.1007/s00216-018-1435-
-------
ENTACT Initial Results: Mixtures
Reported
vs Actual
<75%
>75 to <125%
>125%
59/180
34/180
87/180
499
500
501
502
503
504
505
506
507
508
Mix 1
Mix 2
Mix 3
Mix 4
Mix 5
Mix 6
Mix 7
Mix 8
Mix 9
Mix 10
Actual
95
95
95
95
185
185
365
365
95
365
1
2
3
4
5
6
7
8
9
10
11
12a
12b
13
14
15
16
17
-------
True Positive Rates (TPR)
Blinded Results
Unblinded Results
Mix
Spiked
Substances
Spiked
Isomers
Spiked
Isobars
True
Positives
TPR
Adj. TPR
True
Positives
TPR
Adj. TPR
1
95
2
2
33
0.35
0.35
46
0.48
0.49
2
95
2
2
12 (35)
0.13(0.37)
0.13(0.38)
19(53)
0.20 (0.56)
0.20 (0.57)
3
95
0
0
26
0.27
0.27
47
0.49
0.49
4
95
0
0
44 (36)
0.46 (0.38)
0.46 (0.38)
58 (58)
0.61 (0.61)
0.61 (0.61)
5
185
2
4
66
0.36
0.36
103
0.56
0.56
6
185
2
2
81
0.44
0.44
103
0.56
0.56
7
365
18
26
156
0.43
0.45
225
0.62
0.65
8
365
2
6
144
0.39
0.40
195
0.53
0.54
9
95
52
67
18
0.19
0.42
19
0.20
0.44
10
364
207
257
19(80)
0.05 (0.22)
0.12 (0.51)
31 (107)
0.09 (0.29)
0.20 (0.68)
t
Note: results in () are based
on a 2nd analysis using altered How good are we?
methods (ACN vs. MeOH)
How good can we be?
Sobus JR, etal. (2019) ABC 411 (4):835-851. doi:10.1007/s00216-018-1526-4
-------
Reproducibility of Results
1.0-
0.5--
0.0-
n = 5 (100%)
1
J
Unadjusted o Adjusted
+
oo
X
CXJIX)
n = 36 (63%)
1
a spiked 11 times
4 -> spiked 10 times
-> spiked 4 times
33 -> spiked 3 times
388 -> spiked 2 times
Sobus JR, etal. (2019) ABC 411 (4):835-851. doi:10.1007/s00216-018-1526-4
-------
ENTACT Initial Results: Method Coverage
Not detected
/
195
ESI+
GC with electron impact
LC with electrospray +/-
Ulrich EM, et al. (2019) ABC 411:853-866. doi: 10.1007/s00216-018-1435-6
-------
ENTACT Initial Results
O) 400
ro .2> 300
o 200
Q)
o
o ^1e+4
c
O =*
° o
¦S5
o
CO
Chemical Space
O 200
~_ 100
CO
_D
O
CO
0
ro
T oxcast
ENTACT
GC only
M ESI+ only
F^l ESI-only
Not detected
95th
90th
75th
50th
25th
10th
5th
% ile
~
_L
T
~
-------
ENTACT Initial Results: Multimedia Samples
<75%
>75 to <125%
>125%
- not reported
10/54
2/54
10/54
32/54
Dust
Fort. Dust
Serum
Fort. Serum
Band
Fort. Band
Actual
?
365
?
95
?
185
1
2
-
-
-
-
-
-
3
-
-
-
-
-
-
4
923
1026
-
-
616
680
5
-
-
-
-
-
-
6
7
-
-
-
-
-
-
8
87
236
31
92
46
124
9
277
259
206
222
243
313
10
150
270
31
54
58
101
11
-
-
-
-
-
-
12a
917
1009
638
614
-
-
12b
772
861
94
145
298
557
13
120
124
41
52
24
76
14
188
389
90
178
100
88
15
-
-
-
-
-
-
16
-
-
-
-
-
-
17
-
-
-
-
-
-
-------
Dust Spiking Experiment
|
wmi
%ID REFERENCE IMlE
'1
1!
12585
Organic
Contaminant"
House Dust
•91
Sonication
Methanol
A. EN TACT and B. 4x dust
spiked before extraction
C ©
EN TACT mixture
Siiica > Methanol
•17338?
Solvent spike
Extract Spike
A vs. B- Concentration issues
B vs. C- Recovery
C vs. - Matrix suppression
vs. - Dilution issues
Newtori, S. et al., "Examining NTA performance and
potential using fortified and reference house dust as part
EMTACT," in preparation for ABC.
-------
180
160
140
120
100
80
60
40
20
0
Losses at Each Stage
Mixture Analysis vs. Spiked Solvent
vs. - Dilution issues (29)
vs. - Matrix suppression (34)
C vs. B- Recovery (18)
B vs. A- Concentration issues (33)
Mixture Analysis
Spiked
Solvent
Post-Extraction
Spike
Pre-Extraction
Spi ke
Trial Dust
Newton, S.R. et al., ' Examining NTA
performance and potential using fortified
and reference house dust as part EN TACT,"
in preparation for ABC.
-------
ENTACT LC Ionization Comparison
Singh, R.R. et al., "Expanded coverage of NT-LC-HRMS using atmospheric pressure chemical
ionization: A case study with ENTACT mixtures," in preparation for ABC.
Odds
ratio
ToxPrint Substructure
Structure with ToxPrint
None
148
ring :fused_[6_6]_naphthalene (APCI)
7.607
chain :alkaneCyclic_pentyl_C5
00
APCI-
ESI-
chain:alkaneLinear_hexyl_C6
and _octyl_C8
00
bond:COH_alcohol_sec-alkyl
8.543
APCI +
ESI +
119
bond:COH_alcohol_pri-alkyl
6.891
chain :alkaneCyclic_ethyl_C2_
(connect_noZ)
6.891
187
463
bond:COH_alcohol_generic
3.256
-------
In silico fragmentation: CFM-ID Scoring
a
a
<
Precursor 1
Experimental
Spectrum at CE 10
ra Experimental
^ Spectrum at CE 10
Q.
<
Experimental
Spectrum at CE 10
ra Experimental
Spectrum at CE 20
Experimental
Spectrum at CE 40
Predicted
Spectrum at
CE 10
A
A
ID
H
Candidate 1
Predicted
Spectrum at
CE 20
B
B
B
E
Predicted
Spectrum at
CE 40
c
c
c
Score =
A
Score =
A+B+C
Score =
A+B+C
Score =
D+E+F
Score =
G+H+I
Score =
A+B+C+D+E+
F+O+H+I
./ S«
Chao, A., "In silico MS/MS spectra for identifying unknowns: A critical examination
using CFM-ID algorithms and ENTACT mixture samples," submitted to ABC.
-------
Reference vs. CFM
ID
Libraries
PCDL CFM-ID
88 HI 77
101
"Pass" Compounds
Chao, A., "In silico MS/MS spectra for identifying unknowns: A critical examination
Using CFM-ID algorithms and ENTACT mixture samples," submitted to ABC.
% Compounds
IDed
PCDL
53%
CFM-ID Top 1 Hit
50%
PCDL +
CFM-ID Top 1 Hit
73%
-------
ENTACT Summary and Future Work
4- # features in mixtures » intentionally added substances
4-195 substances not detected by GC or LC-ESI methods, 37 detected by all
4 EPA true positive rate 12-45% blinded and 20-65% unblinded
4- Environmental samples more complex, losses for multiple reasons
4 APCI complementary to ESI, ToxPrints help predict ionization mode success
4 CFM-ID MS2 spectra predicted for DSSTox at 3 LC/ESI energies + GC/EI
4 Unblinded performance for CFM-ID
4- 34% (without formula filtering), 50% (with formula filtering) identified as the top hit
4 73% of compounds identified by reference library match or as the top hit (formula filtering)
4 Additional work on serum and bands
4 Adding GC-Orbitrap and GC-QTOF to cover more volatile chemical space
4 Cross laboratory comparison coming soon
-------
Outline
-f EPA's Non-Targeted Analysis Collaborative Trial (ENTACT)
+ About NTA and ENTACT
-I- Initial results
+ APCI vs ESI
+ Predicted mass spectra
4- Applications of NTA to house dust
Existing tools and future plans for UVCBs
4- Benchmarks and Publications for Non-Targeted Analysis
-------
Children's Exposure: Introduction
Background:
Dust is an important route of exposure, especially for children
Non-targeted analysis (NTA) to understand the exposome (including non-chemical stressors)
Non-chemical stressors include stressors from the built, natural, and social environments
+ Built- building materials, pest problems
+ Social- dietary habits, technology access
+ Natural- access to/use of natural environment
Goal: Understand factors affecting exposure to stressors using dust
Objectives:
Explore variability within NTA dust results from: Time (April, June, October); Room (N = 13);
Size fraction (<150 |jm, 150 |jm - 1mm); Extraction reps (N =3); Analysis reps (N = 3)
Identify differences between dust in homes with (N = 5) and without (N = 4) children
Identify chemical indicators of non-chemical stressors
Work by D. Mills
-------
entry H
dining room
Single Home Variability Study- Clustering
April
3 Living room
7. Hallway +
connector
10. Guest
bedroom
a
I laundry
2. Hallway
6. Master
bedroom
13. Breakfast
Nook
5. Master
closet
+ bath
9. Guest
bath +
Work by D. Mills
June
October
-------
Chemical
Patterns
Master Bath only- PCP?
849.2411@19.12
TDCPP
CI
1
9
1
0 P=0 CI
CI CI
loglO(Abundance)
Termiticide (doorways), flea top spot (dog)
Fipronil
"
IE
~ Not found
TDCPP
Flame retardant in PUF, dissipating?
1 2 3
-F
X
Pyriproxyfen
Similar pattern at all time points, dog?
Black pepper-KITCHEN!
Piperine
BuBzP
Hi
Butylbenzylphthalate
It's everywhere! Construction materials?
1 2 3
[7^^ttT
-------
House Dust with Children vs. No-children
> Samples collected by homeowners using own vacuum in entire home
> Number of children varied from 1-4
> Ages ranged from 3.5 months to 18 years old at time of dust collection
> 5 homes had dogs, 1 home had cats, 3 homes had no pets (N = 9)
> Samples from a variety of states
mapchart.net/usa.html
MUmCYOONK
-------
Children vs. No-children- Clustering
Homes without children
Homes with children
Work by D. Mills
-------
House Dust with Children vs. No-children- Differences
Unique to homes with children (5/5)
+ Sodium copper chlorophyllin:
+ Found mainly personal care products (PCPs)
O Na
O
DECAF
COFFEE
Increased abundance in homes with children
+ Caffeine: 1551 Fold Change (FC)
+ Oleic acid: fats & oils, PCPs, soaps, fragrances 16 FC
+ TDCPP: flame retardant, pesticide 23 FC
+ Bis(2-ethylhexyl) decanedioate: plasticizer used in PCPs/toys 60 FC
HOT
CHOCOLATE
GREEN
TEA
©
SHOT OF
ESPRESSO
mg caffeine/cup
//
(v mfl
Is
INSTANT COFFEE
CAN OF
COLA
BREWED COFFEE
Work by D. Mills
Images from: www.srnutrachem.cn/productgrouplist-802498034/Herb.htrnl
www.irishtimes.com/life-and-style/health-family/caffeine-view-it-as-a-potent-drug-1.1891374
-------
House Dust with Children vs. No-children
Images from: www.southcrop.org/food-trends/tobacco-stocks-plurige-after-fda-proposes-riicotirie-cuts/
www.webmd.com/drugs/2/drug-8858/vancomycin-oral/details
Chemical indicators of non-chemical stressors:
4- Diet/eating practices:
+ Hesperidin: found in citrus fruits
+ Mauritianin: found in cereal/cereal products
4- Cigarette smoking:
+ Nicotine
4* Outdoor time:
Umbelliferone: used in sunscreen
4- Access to medicine:
Vancomycin: antibiotic
+ Pimecrolimus: eczema
Note: Some chemicals tentatively identified
Hesperidin
PlantMedicineNews.com
-------
Applications of NTA to Indoor Dust- Summary
Single home variability study
Less variability across extraction replicates than rooms (as expected)
More variability for large particles
Spatial patterns possibly indicative of chemical use; patterns inconsistent between time points
At higher level, floors cluster together; hierarchical clusters inconsistent between time points
Non-chemical stressors in homes with/without children
Dust from homes cluster based on whether children inhabit home
There are features that are unique to homes with children; some features enriched
Chemical indicators of diet, outdoor activity, smoking, health care non-chemical stressors
detected (identities not confirmed)
-------
Outline
-f EPA's Non-Targeted Analysis Collaborative Trial (ENTACT)
+ About NTA and ENTACT
-I- Initial results
+ APCI vs ESI
+ Predicted mass spectra
4- Applications of NTA to house dust
Existing tools and future plans for UVCBs
4- Benchmarks and Publications for Non-Targeted Analysis
-------
Understanding Exposure to UVCBs
4- Problem
Many commercial products are or contain chemical substances of unknown or variable composition (UVCB)
with no definite molecular formula. UVCB substances generally cannot be characterized using existing
chemical exposure estimation methods. Thus, new methods are needed to further categorize and
characterize UVCB exposure.
4- Actions and Outputs
+ Enumerate UVCB chemical structures and add them to databases
+ Data processing tools- Kendrick Mass Defect analyses, envihomolog (Eawag), others?
+ NTAof UVCBs- identification, fingerprints, composition, and classification models
• Potential candidates: Palm oil, beeswax, chloro-paraffin and hydrocarbon waxes,
kerosine, eucalyptus oil
+ EDA of UVCBs-AOPs, identification, exposure models
-------
4ft JV United States
Environmental Protection Home Advanced Search Batch Search Lists v Predictions Downloads
Agency
DETAILS
Palm oil
8002-75-3 | DTXSID2027674
Searched by Approved Name.
RELATED SUBSTANCES
SYNONYMS
LINKS
BtOACnVTTY
~ EXPOSURE
HAZARD
COMMENTS
PROPERTIES
> LITERATURE
https://comptox.epa.gov/dashboard/
A. Williams
Presence in Lists
Federal
LISThazardous Substances Data Bank EPA: High Production Volume list EPA: Toxicity Values Version 5 (Aug 2018)
US State
None.
International
None.
Other
EPA|ENDOCRINE: Endocrine Disrupter Screening Program (EDSP) Universe of Chemicals
EPAJTSCA: TSCA Inventory, active non-confidential portion
EPA|CPDAT: Chemical a-nd Products Database
Record Information
-
ft Citation: U.S. Environmental Protection Agency. Chemistry Dashboard. https://comptox.epa.gw/dashbOGrd/DTXSID2027674 (accessed May 31st, 2019), Palm oil
Data Quality:
Level 1: Expert curated, highest confidence in accuracy and consistency of unique chemical identifiers
i
i
I Level 2: Expert curated, unique chemical identifiers using muftiple sources
I
I
Level 3: Programmaticaily curated from high quality EPA source, unique chemical identifiers have no conflicts in ChernlD and PubChem
Level 4: Programmatica ly curated from Chen ID, unique chemical identifiers have no conflicts in PubChem
Levei 5: Programmatica'ly curated from ACToR or PubChem, unique chemical :dentifiers with !ow confidence, single public source
Quality Control Notes
-------
Enumeration of UVCBs
• Generating base case SMILES in RChemMass
• https://qithub.com/schvmane/RChenriMass/
23 # try with one 1,2 variant
24 genSmilesl2PPGl <- "[R]0"
25 RltON_12PPGl <- "(OC(C)O"
26 nRltoN_PPG <- "(0-15)"
27
28 smiles_12PPGl_Otol6 <- buildSmiles(genSmilesPPG,RltoN_12PPGl,nRltoN_PPG)
29 smiles_12PPGl_0tol6
30 write.table(smiles_12PPGl_0tol6,"smiles_12PPGl_0tol6.txt",row.names = F, col.names = F,quote = F)
31
> smiles_12PPGl_0tol6 <- buildSmiles(genSmilesPPG,RltoN_12PPGl,nRltoN_PPG)
> smiles_12PPGl_0tol6
[1] "OC(C)CO"
[2] "OC(C)COC(C)CO"
[3] "OC(C)COC(C)COC(C)CO"
[4 ] "OC (C) COC (C) COC (C) coc (C) CO "
[5] "OC(C)COC(C)COC(C)COC(C)COC(C)CO"
[6] "OC(C)COC(C)COC(C)COC(C)COC(C)COC(C)CO"
[7] "OC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) CO "
[8] "OC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)CO"
[9] "0C(C)C0C(C)C0C(C)C0C(C)C0C(C)C0C(C)C0C(C)C0C(C)C0C(C)C0"
[10] "OC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)CO"
[11] "OC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)CO"
[12] "OC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) CO "
[13] "OC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)CO"
[14] "OC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)COC(C)CO"
[15] "OC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) CO "
[16 ] "OC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) COC (C) CO "
-------
Kendrick Mass Defect Plots
IUPAG
Mass Defect
©J
0.5
0.4
0,3
0.2
0.1 L
400
,f ¦ »«
/ ' ¦ •'/
I l.l.:.,
i . i •. '
< ¦ . * j
* i ' • 'I
'» • %',* • / «
¦ : i ' ' i
• i • ! i
< ' i ' •
. I
, 1 •
I .
!
I ».
I . "
i
.. ' i
¦V:
• i
i . '
i » i . i , i . ;
i . i •
i , i . i . i , i •. :
, : * », : !
i». «. ». •«.
' f, * f **f, '*1.
«.» * I
i
. 1 ,
i , ;
* *
. i
. 1 .
*
4, * *
: «,
, 1
• . : ».
: »,
, 1 .
• K
1 ..
iV'
* :
K Z
i».
» u »
¦¦ *
• •
:
1, •
, " «.
i. •»
" * ».
' »
•
« * •,
¦ i ¦
. * i
" ( /
• »
• t .
#
m
- I
¦# *
* m + * «
* *
• r ~
• *
~* I
% 9 *~. * ~!» *
* .i " » .
«>
450 500 550
Nominal IUPAC Hass
600
Plot of IUPAC mass defect vs nominal IUPAC
mass for ions of odd nominal mass
On the nominal mass axis:
Periodicity of 2 Da (H2, rings + double bonds)
Periodicity of 14 Da (CH2, chain length)
On the mass defect axis:
Periodicity of 0.015 65 Da (H2)
These spacings make it possible to determine
molecular "class" and "type" simultaneously
over a wide mass range from a single display
Hughey CA, etal. (2001) Analytical Chemistry, 73 (19):4676-4681. doi:10.1021/ac010560w
-------
Homologous Series and UVCBs in samples
o
o
o
CM
O
O
LO
O
O
o
o
o
LO
^S-OH
o
0-i zC
9' '19
~T~
0
~T~
5
10
I
15
Retention time
I
20
58.04 0
• 48.01
47.02 •
40.03 «
37.04 • 37.06
• 36
35.03 •
• 32.02
30.05 •
• 28.03
27.02 5
( 26.02
24.03 rn 24
22.03 A
• 20.03
18 .04 A
I 17.02
• 10.03
9 #
• 8.02
7.01 •
25
I
30
aquatic research Vooo
nontarget
http://www.envihomolog.eawag.ch/
E. Schymanski et al. 2014, ES&T
DO I: 10.1021 /es4044374
M. Loos & H. Singer, 2017. J. Cheminf.
DOI: 10.1186/sl 3321-017-0197-z
-------
Outline
-f EPA's Non-Targeted Analysis Collaborative Trial (ENTACT)
+ About NTA and ENTACT
-I- Initial results
+ APCI vs ESI
+ Predicted mass spectra
4- Applications of NTA to house dust
Existing tools and future plans for UVCBs
4- Benchmarks and Publications for Non-Targeted Analysis
-------
Benchmarking and Publications for
Non-targeted Analysis (BP4NTA)
A diverse working group
consisting of
representatives from all
sectors dedicated to
improving NTA through
benchmark methods
and performance,
standardized definitions
and publication standards.
Join us! Informational mtg
at SETAC Toronto
November 4 @
Rm 703 or phone
Monday,
2:15pm,
Contact Ben Place (benjamin.place@nist.gov) or Elin Ulrich (ulrich.elin@epa.gov) for more information
------- |