PMID- 28074598
OWN - NLM
STAT- MEDLINE
DCOM- 20180220
LR  - 20181113
IS  - 1099-1263 (Electronic)
IS  - 0260-437X (Print)
IS  - 0260-437X (Linking)
VI  - 37
IP  - 7
DP  - 2017 Jul
TI  - Prediction of skin sensitization potency using machine learning approaches.
PG  - 792-805
LID - 10.1002/jat.3424 [doi]
AB  - The replacement of animal use in testing for regulatory classification of skin 
      sensitizers is a priority for US federal agencies that use data from such 
      testing. Machine learning models that classify substances as sensitizers or 
      non-sensitizers without using animal data have been developed and evaluated. 
      Because some regulatory agencies require that sensitizers be further classified 
      into potency categories, we developed statistical models to predict skin 
      sensitization potency for murine local lymph node assay (LLNA) and human 
      outcomes. Input variables for our models included six physicochemical properties 
      and data from three non-animal test methods: direct peptide reactivity assay; 
      human cell line activation test; and KeratinoSens assay. Models were built to 
      predict three potency categories using four machine learning approaches and were 
      validated using external test sets and leave-one-out cross-validation. A 
      one-tiered strategy modeled all three categories of response together while a 
      two-tiered strategy modeled sensitizer/non-sensitizer responses and then 
      classified the sensitizers as strong or weak sensitizers. The two-tiered model 
      using the support vector machine with all assay and physicochemical data inputs 
      provided the best performance, yielding accuracy of 88% for prediction of LLNA 
      outcomes (120 substances) and 81% for prediction of human test outcomes (87 
      substances). The best one-tiered model predicted LLNA outcomes with 78% accuracy 
      and human outcomes with 75% accuracy. By comparison, the LLNA predicts human 
      potency categories with 69% accuracy (60 of 87 substances correctly categorized). 
      These results suggest that computational models using non-animal methods may 
      provide valuable information for assessing skin sensitization potency. Copyright 
      (c) 2017 John Wiley & Sons, Ltd.
CI  - Copyright (c) 2017 John Wiley & Sons, Ltd.
FAU - Zang, Qingda
AU  - Zang Q
AD  - ILS, Research Triangle Park, NC, 27709, USA.
FAU - Paris, Michael
AU  - Paris M
AD  - ILS, Research Triangle Park, NC, 27709, USA.
FAU - Lehmann, David M
AU  - Lehmann DM
AD  - EPA/NHEERL/EPHD/CIB, Research Triangle Park, NC, 27709, USA.
FAU - Bell, Shannon
AU  - Bell S
AD  - ILS, Research Triangle Park, NC, 27709, USA.
FAU - Kleinstreuer, Nicole
AU  - Kleinstreuer N
AD  - NIH/NIEHS/DNTP/NICEATM, Research Triangle Park, NC, 27709, USA.
FAU - Allen, David
AU  - Allen D
AD  - ILS, Research Triangle Park, NC, 27709, USA.
FAU - Matheson, Joanna
AU  - Matheson J
AD  - US Consumer Product Safety Commission, Bethesda, MD, 20814, USA.
FAU - Jacobs, Abigail
AU  - Jacobs A
AD  - FDA/CDER, Silver Spring, MD, 20993, USA.
FAU - Casey, Warren
AU  - Casey W
AD  - NIH/NIEHS/DNTP/NICEATM, Research Triangle Park, NC, 27709, USA.
FAU - Strickland, Judy
AU  - Strickland J
AD  - ILS, Research Triangle Park, NC, 27709, USA.
LA  - eng
GR  - HHSN273201500010C/ES/NIEHS NIH HHS/United States
PT  - Journal Article
PT  - Research Support, N.I.H., Extramural
DEP - 20170110
PL  - England
TA  - J Appl Toxicol
JT  - Journal of applied toxicology : JAT
JID - 8109495
RN  - 0 (Hazardous Substances)
SB  - IM
MH  - Animal Testing Alternatives/*methods
MH  - Biological Assay/*methods
MH  - Dermatitis, Allergic Contact/*etiology/*immunology
MH  - Hazardous Substances/*toxicity
MH  - Humans
MH  - *Machine Learning
MH  - Models, Statistical
MH  - Skin/*drug effects
MH  - United States
PMC - PMC5435511
MID - NIHMS842013
OTO - NOTNLM
OT  - KeratinoSens
OT  - Skin sensitization potency
OT  - allergic contact dermatitis (ACD)
OT  - direct peptide reactivity assay (DPRA)
OT  - h-CLAT (human cell line activation test)
OT  - integrated decision strategy (IDS)
OT  - machine learning
OT  - murine local lymph node assay (LLNA)
COIS- Conflict of Interest The authors declare that there are no conflicts of interest.
EDAT- 2017/01/12 06:00
MHDA- 2018/02/21 06:00
PMCR- 2018/07/01
CRDT- 2017/01/12 06:00
PHST- 2016/10/13 00:00 [received]
PHST- 2016/10/26 00:00 [revised]
PHST- 2016/11/01 00:00 [accepted]
PHST- 2017/01/12 06:00 [pubmed]
PHST- 2018/02/21 06:00 [medline]
PHST- 2017/01/12 06:00 [entrez]
PHST- 2018/07/01 00:00 [pmc-release]
AID - 10.1002/jat.3424 [doi]
PST - ppublish
SO  - J Appl Toxicol. 2017 Jul;37(7):792-805. doi: 10.1002/jat.3424. Epub 2017 Jan 10.