Comprehensive mapping of the AOP-Wiki database: identifying biological and disease gaps

Author: Jaylet, Thomas; Coustillet, Thibaut; Smith, Nicola M.; Viviani, Barbara; Lindeman, Birgitte; Vergauwen, Lucia; Myhre, Oddvar; Yarar, Nurettin; Gostner, Johanna M.; Monfort-Lanzas, Pablo; Jornod, Florence; Holbech, Henrik; Coumoul, Xavier; Sarigiannis, Di

Publisher: Zenodo

DOI: 10.5281/zenodo.10813189

Source: https://zenodo.org/records/10813189/files/ftox-06-1285768.pdf

Comp ehensi e mapping o he
AOP-Wiki da abase: iden i ying
biological and disease gaps
Thomas Jayle
1†
, Thibau Cous ille
1†
, Nicola M. Smi h
2
,
Ba ba a Vi iani
3
, Bi gi e Lindeman
2
, Lucia Ve gauwen
4
,
Odd a Myh e
2
, Nu e in Ya a
2
, Johanna M. Gos ne
5
,
Pablo Mon o -Lanzas
5
,
6
, Flo ence Jo nod
1
, Hen ik Holbech
7
,
Xa ie Coumoul
1
, Dimos henis A. Sa igiannis
8
,
9
,
10
,
Philipp An czak
11
,
12
,
13
, Anna Bal-P ice
14
, Ellen F i sche
15
,
16
,
17
,
18
,
Eliska Kucho ska
15
, An onios K. S a idakis
10
, Robe Ba ouki
1
,
Min Ji Kim
19
, Oli ie Tabou eau
20
, Ma cin W. Wojewodzic
2
,
21
,
D ies Knapen
4
and Ka ine Audouze
1
*
1
Uni e si é Pa is Ci é, Inse m UMR-S 1124 T3S, Pa is, F ance,
2
No wegian Ins i u e o Public Heal h,
Di ision o Clima e and En i onmen , Oslo, No way,
3
Depa men o Pha macological and Biomolecula
Sciences, Uni e si à degli S udi di Milano, Milan, I aly,
4
Zeb aﬁshlab, Depa men o Ve e ina y Sciences,
Ve e ina y Physiology and Biochemis y, Uni e si y o An we p, Wil ijk, Belgium,
5
Ins i u e o Medical
Biochemis y, Medical Uni e si y o Innsb uck, Innsb uck, Aus ia,
6
Ins i u e o Bioin o ma ics, Biocen e ,
Medical Uni e si y o Innsb uck, Innsb uck, Aus ia,
7
Depa men o Biology, Uni e si y o Sou he n
Denma k, Odense, Denma k,
8
En i onmen al Enginee ing Labo a o y, Depa men o Chemical
Enginee ing, A is o le Uni e si y o Thessaloniki, Thessaloniki, G eece,
9
Na ional Hellenic Resea ch
Founda ion, A hens, G eece,
10
Science, Technology and Socie y Depa men , En i onmen al Heal h
Enginee ing, Uni e si y School o Ad anced S udies (IUSS), Pa ia, I aly,
11
Depa men II o In e nal
Medicine, Facul y o Medicine and Uni e si y Hospi al Cologne, Uni e si y o Cologne, Cologne,
Ge many,
12
Cen e o Molecula Medicine Cologne, Facul y o Medicine and Uni e si y Hospi al
Cologne, Uni e si y o Cologne, Cologne, Ge many,
13
Cologne Excellence Clus e on Cellula S ess
Responses in Aging-Associa ed Diseases (CECAD), Cologne, Ge many,
14
Eu opean Commission, Join
Resea ch Cen e (JRC), Isp a, I aly,
15
IUF-Leibniz Resea ch Ins i u e o En i onmen al Medicine,
Duesseldo , Ge many,
16
Hein ich-Heine-Uni e si y, Düsseldo , Ge many,
17
Swiss Cen e o Applied
Human Toxicology, Basel, Swi ze land,
18
DNTOX GmbH, Düsseldo , Ge many,
19
Inse m UMR-S 1124,
Uni e si é So bonne Pa is No d, Bobigny, Pa is, F ance,
20
Uni e si é Pa is Ci é, BFA, Team CMPLI, Inse m
U1133, CNRS UMR 8251, Pa is, F ance,
21
Cance Regis y o No way, NIPH, Oslo, No way
In oduc ion: The Ad e se Ou come Pa hway (AOP) concep acili a es apid
haza d assessmen o human heal h isks. AOPs a e cons an ly e ol ing, hei
numbe is g owing, and hey a e e e enced in he AOP-Wiki da abase, which is
suppo ed by he OECD. He e, we p esen a s udy ha aims a iden i ying well-
deﬁned biological a eas, as well as gaps wi hin he AOP-Wiki o u u e esea ch
needs. I does no in end o p o ide a sys ema ic and comp ehensi e summa y o
he a ailable li e a u e on AOPs bu summa izes and maps biological knowledge
and diseases ep esen ed by he al eady de eloped AOPs (wi h OECD endo sed
s a us o unde alida ion).
Me hods: Knowledge om he AOP-Wiki da abase we e ex ac ed and p epa ed
o analysis using a mul i-s ep p ocedu e. An au oma ic mapping o he exis ing
in o ma ion on AOPs (i.e., genes/p o eins and diseases) was pe o med using
bioin o ma ics ools (i.e., o e ep esen a ion analysis using Gene On ology and
DisGeNET), allowing bo h he classiﬁca ion o AOPs and he de elopmen o AOP
ne wo ks (AOPN).
OPEN ACCESS
EDITED BY
E win L. Roggen,
Independen Resea che , Lyngby, Denma k
REVIEWED BY
Hideko Sone,
Yokohama College o Pha macy, Japan
Ca he ine Wille ,
Humane Socie y In e na ional, Uni ed Kingdom
*CORRESPONDENCE
Ka ine Audouze,
[email p o ec ed]
†
These au ho s ha e con ibu ed equally o his
wo k and sha e ﬁ s au ho ship
RECEIVED 30 Augus 2023
ACCEPTED 15 Feb ua y 2024
PUBLISHED 08 Ma ch 2024
CITATION
Jayle T, Cous ille T, Smi h NM, Vi iani B,
Lindeman B, Ve gauwen L, Myh e O, Ya a N,
Gos ne JM, Mon o -Lanzas P, Jo nod F,
Holbech H, Coumoul X, Sa igiannis DA,
An czak P, Bal-P ice A, F i sche E, Kucho ska E,
S a idakis AK, Ba ouki R, Kim MJ, Tabou eau O,
Wojewodzic MW, Knapen D and Audouze K
(2024), Comp ehensi e mapping o he AOP-
Wiki da abase: iden i ying biological and
disease gaps.
F on . Toxicol. 6:1285768.
doi: 10.3389/ ox.2024.1285768
COPYRIGHT
© 2024 Jayle , Cous ille , Smi h, Vi iani,
Lindeman, Ve gauwen, Myh e, Ya a , Gos ne ,
Mon o -Lanzas, Jo nod, Holbech, Coumoul,
Sa igiannis, An czak, Bal-P ice, F i sche,
Kucho ska, S a idakis, Ba ouki, Kim,
Tabou eau, Wojewodzic, Knapen and Audouze.
This is an open-access a icle dis ibu ed unde
he e ms o he C ea i e Commons A ibu ion
License (CC BY). The use, dis ibu ion o
ep oduc ion in o he o ums is pe mi ed,
p o ided he o iginal au ho (s) and he
copy igh owne (s) a e c edi ed and ha he
o iginal publica ion in his jou nal is ci ed, in
acco dance wi h accep ed academic p ac ice.
No use, dis ibu ion o ep oduc ion is
pe mi ed which does no comply wi h hese
e ms.
F on ie s in Toxicology on ie sin.o g01
TYPE O iginal Resea ch
PUBLISHED 08 Ma ch 2024
DOI 10.3389/ ox.2024.1285768
Resul s: AOPs ela ed o diseases o he geni ou ina y sys em, neoplasms and
de elopmen al anomalies a e he mos equen ly in es iga ed on he AOP-Wiki.
An e alua ion o he h ee p io i y cases (i.e., immuno oxici y and non-geno oxic
ca cinogenesis, endoc ine and me abolic dis up ion, and de elopmen al and adul
neu o oxici y) o he EU- unded PARC p ojec (Pa ne ship o he Risk Assessmen o
Chemicals) a e p esen ed. These we e used o highligh unde - and o e - ep esen ed
ad e se ou comes and o iden i y and p io i ize gaps o u he esea ch.
Discussion: These esul s con ibu e o a mo e comp ehensi e unde s anding o
he ad e se e ec s associa ed wi h he molecula e en s in AOPs, and aid in eﬁning
isk assessmen o s esso s and mi iga ion s a egies. Mo eo e , he FAIRness
(i.e., da a which mee s p inciples o ﬁndabili y, accessibili y, in e ope abili y, and
eusabili y (FAIR)) o he AOPs appea s o be an impo an conside a ion o u he
de elopmen .
KEYWORDS
ad e se ou come pa hway, AOP ne wo k, immuno oxici y, neu o oxici y, nongeno oxic
ca cinogenesis, diseases, PARC
1 In oduc ion
In 2007, he US Na ional Resea ch Council (NRC) published a
e iew ‘Toxici y es ing in he 21s cen u y: a ision and a s a egy’,
highligh ing he ongoing pa adigm shi in oxici y es ing (K ewski
e al., 2020). Indeed, exis ing isk assessmen me hodologies and
oxicological es s a e no in line wi h he la ge numbe o no
su ﬁcien ly es ed exis ing and he apidly g owing numbe o no el
subs ances, including me aboli es p oduced in he en i onmen by
non-humans, ha need o be e alua ed. Consequen ly, he po en ial
oxici y o many subs ances is poo ly cha ac e ized. Mo eo e , he
di e si y o he o gan sys ems being a ge ed by con aminan s o
pollu an s, inc eases he need o implemen new es s. The e o e, in
his s udy, au ho s sugges ed ecommenda ions o imp o e and
accele a e chemical es ing, and emphasized he concep o ‘ oxici y
pa hways’(Ankley e al., 2010). The de elopmen and use o New
App oach Me hodologies (NAMs), which include animal and non-
animal-based me hods, ha e inc eased o assess speciﬁc endpoin s
ha indica e human and eco oxicological isks om exposu e o
en i onmen al subs ances. NAMs a e de eloped o iden i y a
oxicologically ele an esponse a di e en biological le els
(molecula , cellula , o gan, o ganism) p o oked by a chemical
exposu e (Vinken e al., 2017). NAMs include in cellulo, in i o
and in chemico sc eening, compu a ional me hods as well as high
h oughpu mul i-omics and exposomics (Babin e al., 2023;Mai e
e al., 2023) o men ion bu a ew me hodologies. Recen ly, a s udy
explo ed he p og ess in his ‘nex -gene a ion’o isk assessmen
me hodologies, and highligh ed a low, bu s ill g owing e olu ion
and use o NAMs in oxici y es ing and isk assessmen (K ewski
e al., 2020). Mo eo e , limi a ions o he use o animals in egula o y
s udies a e also an impo an componen o ake in o conside a ion,
as animal OECD es guideline s udies a e ime- and cos -in ensi e.
Fu he mo e, i is inc easingly e hically ques ionable ( es ing one
subs ance may equi e se e al housand animals) in addi ion o he
unce ain ies in me hodologies, e alua ion, egula ion, and
ex apola ion. Al hough s ill assumed o be he mos p o ec i e ypes
o s udy, in i o animal s udy p edic i i y o he p o ec ion o human
heal h is some imes insu ﬁcien gi en he di e ences in o gan unc ion/
complexi y, exposu e, de elopmen al iming, oxicokine ics and
oxicodynamics be ween humans and o he mammals (Tsuji and
C o on, 2012). In he ecen assessmen o Ex ended One-
Gene a ion Rep oduc i e Toxici y S udies (EOGRTS; OECD TG
443) conduc ed wi h 55 indus ial chemical subs ances unde
REACH (Eu opean Chemicals Agency, 2023), me hodological issues
we e p esen ed conce ning he de elopmen al immuno- and
neu o oxici y coho s ha appea o be pa icula ly demanding in
e ms o p oﬁciency, i.e., deﬁciencies we e equen ly ound ha
hinde ed he in e p e a ion o he esul s. This, again, highligh s he
need o me hods’de elopmen o suppo isk assessmen o chemicals
by conside ing human biology and p o iding a deep mechanis ic
unde s anding o he oxicological e en s leading o ad e se e ec s.
To acili a e he use o NAMs in a egula o y con ex , a new
concep was p oposed by Ankley in 2010, named Ad e se Ou come
Pa hways (AOP) (Ankley e al., 2010). The AOP amewo k
desc ibes and o ganizes exis ing biological knowledge o
humans and wildli e om he pe u ba ion by a s esso leading
o an ad e se e ec , in a s uc u ed manne . An AOP s a s wi h a
Molecula Ini ia ing E en (MIE) (ini ia ed by a p o o ypical
s esso , he la e no being pa o he AOP), linked o a
sequence o biological Key E en s (KE) ha ul ima ely lead o an
Ad e se Ou come (AO) a he le el o an indi idual, popula ion o
ecosys em. One main objec i e is o suppo chemical isk
assessmen based on mechanis ic easoning and egula o y
oxicology. In 2013, he O ganiza ion o Economic Co-ope a ion
and De elopmen (OECD) s a ed an ini ia i e o he o maliza ion
o AOP de elopmen , esul ing in guidance documen s (OECD,
2017;OECD, 2018). Today, AOPs a e becoming mo e and mo e
accep able as a amewo k o help chemical sa e y assessmen and
egula o y oxicology by suppo ing a sys ema ic way o p edic ing
AOs based on accumula ed mechanis ic knowledge (S ingen e al.,
2021). To ollow he egula o y amewo k p oposed by he
Eu opean Union (EU) o educe le els o en i onmen al s esso s
(e.g., Endoc ine Dis up ing Chemicals (EDC)), esea ch-d i en
app oaches need o be de eloped, which include AOP-in o med
In eg a ed App oaches o Tes ing and Assessmen (IATA) as an
impe a i e ea u e o he isk assessmen p ocess. The Eu opean
Clus e o Imp o e Iden iﬁca ion o Endoc ine Dis up o s
(EURION, h ps://eu ion-clus e .eu)(S ee e al., 2021)as
F on ie s in Toxicology on ie sin.o g02
Jayle e al. 10.3389/ ox.2024.1285768
ini ia ed a ew yea s ago wi h he aim o p opose new and e ec i e
me hods o EDC es ing, ela ed o se e al heal h ou comes such as
me abolic diso de s (Audouze e al., 2020) o de elopmen al
neu o oxici y (DNT) (Lupu e al., 2020). Likewise, he ASPIS
clus e (h ps://aspis-clus e .eu/) uses AOP-based app oaches o
implemen a ion o no el s a egies o animal- ee sa e y assessmen
o (non-endoc ine) chemicals. Fu he mo e, he Pa ne ship o he
Assessmen o Risks om Chemicals ini ia i e (PARC) has he
ambi ion o de elop a obus isk assessmen o new gene a ion
chemicals o be e p o ec heal h and he en i onmen (h ps://
www.eu-pa c.eu/)(Ma x-S oel ing e al., 2023). The de elopmen o
AOPs is a key ea u e o PARC. P e iously, se e al s udies ha e
desc ibed bo h he de elopmen o AOPs and hei po en ial uses o
isk assessmen (Villeneu e e al., 2014;Leis e al., 2017;Villeneu e
e al., 2018a;Knapen e al., 2018;Vinken, 2018;Baja d e al., 2023).
E idence-based app oaches (Audouze e al., 2021) and inno a i e
ools ha e been p oposed o help in he cons uc ion o AOPs such
as he AOP-helpFinde ool ha is based on a iﬁcial in elligence
(h p://aop-helpﬁnde .u-pa is-sciences. /)(Ca aillo e al., 2019;
Jo nod e al., 2022;Jayle e al., 2023). AOP-helpFinde
au oma ically sc eens he a ailable li e a u e and p o ides a e y
impo an sou ce o knowledge ha can be used o AOP
de elopmen (Benoi e al., 2022;Jayle e al., 2022). Since 2014,
all de eloped AOPs a e s o ed in he AOP-Wiki da abase (h ps://
aopwiki.o g/) ha is pa o he AOP-knowledge-base (AOP-KB;
h ps://aopkb.oecd.o g) se up by he OECD. Fo each p oposed
AOP, KEs should be measu able, meaning ha associa ed es
me hods should need o be men ioned i hey exis , o de eloped
i hey do no . The e o e, he de elopmen o AOPs ollows he
cu en a ailabili y o es me hods. Fo example, a ecen s udy
ocusing on EDCs, highligh ed ha alida ed in i o and in i o
es s used by he EU o iden i y he po en ial oxici y o EDCs do no
assess all he endoc ine pa hways (Zgheib e al., 2021). F om hese
ﬁndings, an online ool has been de eloped o e alua e he s a us o
hei es s unde de elopmen be o e submission o he OECD o
endo semen (h ps:// eaded es .u-pa is-sciences. /).
The cu en s udy aims in a holis ic app oach ac oss he whole
con en o he AOP-Wiki o iden i y gaps in he cu en AOPs
wi hin he AOP-Wiki da abase o u u e esea ch needs. I does no
in end o p o ide a sys ema ic and comp ehensi e summa y o he
a ailable li e a u e, bu a he summa izes and maps he biological
knowledge and diseases ep esen ed by he al eady de eloped AOPs
(OECD endo sed s a us o s ill unde alida ion).
2 Ma e ial and me hods
The p oposed app oach, o explo e in o ma ion ela ed o AOPs
om he AOP-Wiki (i.e., a he gene/p o ein le els and he disease
le el), is a mul i-s ep p ocedu e illus a ed in Figu e 1.
2.1 Compila ion and p epa a ion o AOP
in o ma ion om he AOP-Wiki da abase
Fo bo h mapping app oaches (genes/p o eins and ad e se
ou comes), he AOP-Wiki da abase was used. The AOP-Wiki
compiles da a o all exis ing AOPs submi ed by esea che s and a e
ei he unde de elopmen o ha e been endo sed by he OECD. By he
ime o ou analysis, i con ained 403 unique AOPs, bu con inues o
inc ease apidly due o global e o s based on c owdsou ced
collabo a ion. Howe e , un il now only 29 AOPs ha e been endo sed
by he OECD, he o he s being unde de elopmen o unde e alua ion.
In his s udy, all in o ma ion ela ed o AOPs and any key e en (MIE,
KE, AO) we e downloaded om he AOP-Wiki da abase on he 22nd o
May 2023 ( e sion 2.6; h ps://aopwiki.o g/in o_pages/5).
FIGURE 1
Wo kﬂow o he p ocedu e de eloped o assess he global mapping o exis ing Ad e se Ou come Pa hways (AOPs). Fo each e alua ion, a
comp ehensi e wo kﬂow was de eloped: (A) Wo kﬂow o AOP and he biological space, (B) Wo kﬂow o AOP and disease space and (C) Wo kﬂow o
he h ee case s udies. MIE: Molecula Ini ia ing E en , KE: Key E en , AO: Ad e se Ou come, AOPN: Ad e se Ou come Pa hway Ne wo k, HGNC–HUGO
gene Nomencla u e Commi ee, ICD-11–In e na ional Classiﬁca ion o Diseases 11 h Re ision, GO: Gene On ology.
F on ie s in Toxicology on ie sin.o g03
Jayle e al. 10.3389/ ox.2024.1285768
2.1.1 Da a se o he gene/p o ein space
explo a ion
To ex ac he comple e se o genes and p o eins in ol ed in exis ing
AOPs, we pe o med a comp ehensi e e alua ion o he 1371 biological
e en s (MIEs, KEs and AOs) de i ed om he 403 AOPs con ained in
he AOP-Wiki da abase. Conside ing he in e -species cha ac e is ics o
AOPs and he collabo a i e na u e o he AOP-Wiki, i was c ucial o
conside possible edundancies in gene and p o ein anno a ions
including synonyms and o hologs. To ensu e he uniqueness o
biological en i ies and elimina e synonyms, a me iculous manual
e iﬁca ion p ocess was conduc ed (e.g., he NR1I3 ecep o is
labeled as And os ane ecep o , CAR, o NR1I3 depending on he
AOP au ho s). Then, o acili a e da a in eg a ion and in e p e a ion, as
well as o a oid po en ial conﬂic s a ising om in e -species genes
du ing subsequen en ichmen s eps, we decided o e ain only human
genes and map hem o hei co esponding HGNC gene nomencla u e
(h ps://www.genenames.o g/)(Seal e al., 2023). Fo non-human genes,
we pe o med a sea ch o he p esence o human o hologs using he
HGNC Compa ison o O holog P edic ions (HCOP) ool (h ps://
www.genenames.o g/ ools/hcop/), aiming o p ese e maximum
in o ma ion o subsequen analyses. The comple e lis o ex ac ed
genes and p o eins is p esen ed in Supplemen a y Table S1. Genes
mapped o hei HGNC symbols we e subsequen ly used o he
explo a ion o biological pa hways using Gene On ology (GO) and
o disease associa ions using DisGeNET (as de ailed in Sec ions 2.2.1
and 2.2.2, espec i ely).
2.1.2 Da a se o he disease space explo a ion
F om he 403 exis ing AOPs, we ex ac ed 194 unique AOs. In he
AOP-Wiki da abase, some e en s can be ca ego ized di e en ly
depending on he AOP hey a e in ol ed in, acco ding o he
de elope he same e en can be labeled as ei he an MIE, a KE o
an AO (e.g., e en ID 759 ‘Inc eased, Kidney ailu e’was labeled as an
AO in he AOP ID 33 and 447 while i was labeled as a KE in he AOP
ID 377 and 413). No cu a ion was done o assess and coun he ac ual
numbe o AOs sca e ed in all he AOPs, i.e., da a ela ed o AOP-
Wiki e en s is ep esen ed as i was in he AOP-Wiki, so ha he e en
ID 759 ‘Inc eased, K. ailu e’wascoun ed as occu ing wice as an AO.
Then, each compiled AO was manually checked by expe s o a oid
edundancy o in o ma ion as some may be synonyms. Fo example,
he AO ID 1458 ‘Pulmona y ﬁb osis’and AO ID 1276 ‘lung ﬁb osis’
we e me ged in o one unique AO. The name chosen among all he
synonyms o an AO was a bi a y, o ins ance, we kep ‘lung ﬁb osis’
in he abo e example, so ha ‘lung ﬁb osis’encloses he numbe o
occu ences o AO ID 1458 and AO ID 1276. A e me ging
synonyms, a o al o 149 AOs we e kep o he classiﬁca ion
(Supplemen a y Table S2 shee 1 ICD_11 classiﬁca ion and shee
6synonyms). These AOs we e in ol ed in 379 AOPs among he
403 cu en ly p esen in he AOP-Wiki, meaning ha 24 AOPs did
no ha e any AO e en s lis ed.
2.2 Mapping he AOPs o he biological
space using gene/p o ein in o ma ion
2.2.1 Gene and pa hways
An o e ep esen a ion analysis (ORA) was pe o med on
22 May 2023, using he Gene On ology (GO) da abase
(Ashbu ne e al., 2000), which classiﬁes genes based on hei
Cellula Componen (CC), Molecula Func ion (MF), and
Biological P ocess (BP) anno a ions. The analysis was
conduc ed wi h he g:P oﬁle ool ( e sion e109_e.g.,56_p17_
1d3191d) (Raud e e e al., 2019) using a Fishe ’s one- ailed es
(o hype geome ic es ). The Benjamini–Hochbe g False
Disco e y Ra e (FDR) me hod was applied o adjus he
p- alues, and an adjus ed p- alue h eshold o 0.05 was used o
selec en iched GO e ms. The esul s o he BP, CC, and MF
ca ego ies a e p o ided in Supplemen a y Table S3. To simpli y he
isualiza ion and in e p e a ion o he ORA esul s, he Re iGO
ool was used (Supek e al., 2011). By applying a clus e ing
algo i hm ha inco po a es seman ic simila i y measu es,
Re iGO educed da a edundancy while acili a ing he
g aphical ep esen a ion o he mos ele an e ms h ough
mul idimensional scaling (MDS).
2.2.2 Genes and diseases
Each selec ed HGNC symbol was in es iga ed o disease
associa ions using he DisGeNET da abase, a comp ehensi e
pla o m con aining one o he la ges publicly a ailable
collec ions o genes and a ian s associa ed wi h human
diseases (Piñe o e al., 2020). The pu pose o his analysis was
o iden i y po en ial disease associa ions and gain u he insigh s
in o he biological implica ions o ou gene lis . We u ilized he
disease_en ichmen unc ion om he disgene 2 R package o
sea ch o disease associa ions (Piñe o e al., 2020). The esul s
we e ﬁl e ed based on a False Disco e y Ra e (FDR) h eshold o
0.05 o selec s a is ically signiﬁcan associa ions. In addi ion o he
disease en ichmen analysis, a second analysis was pe o med
using he gene2disease unc ion om he disgene 2 R package,
wi h he lis o genes om he AOP and he cu a ed DisGeNET
da abase. This analysis allowed us o explo e he ela ionship
be ween he genes and hei co esponding Medical Subjec
Headings (MeSH) disease classes. A isualiza ion was c ea ed o
display he associa ions be ween he genes and he MeSH disease
classes, using a cho d diag am o be e unde s and hese
ela ionships. To enhance cla i y and ocus on he mos
ele an associa ions, only ela ionships wi h g ea e han 60%
in ol emen we e kep in he ﬁnal isualiza ion.
2.3 Mapping o he AOP o he disease space
using he ad e se ou come in o ma ion
2.3.1 Disease en ichmen
To ge an o e iew o exis ing da a ob ained om he AOP-Wiki
and o acili a e u he analysis, all selec ed AOs om 2.1.2 we e
au oma ically classiﬁed using he 11 h In e na ional Classiﬁca ion o
Diseases sys em (ICD-11) p o ided by he Wo ld Heal h
O ganiza ion (h ps://icd.who.in /b owse11/l-m/en). The
classiﬁca ion ou come was u he checked and cu a ed by
expe s wi hin he ﬁeld o human oxicology. ICD-11 e sion was
launched 1s Janua y 2022 and con ains a o al o 24 disease
ca ego ies. To ha e he mos comple e o e iew o he diseases
p esen in he di e en AOPs al eady de eloped, he unclassiﬁed
AOs using ICD-11 we e checked by expe s, and, i needed, new
classes we e c ea ed.
F on ie s in Toxicology on ie sin.o g04
Jayle e al. 10.3389/ ox.2024.1285768
2.3.2 Iden iﬁca ion o gaps
To a oid any in e p e a ion bias conce ning AOPs ha ha e
been o e - o insu ﬁcien ly app aised, we ca ied ou a o mal
s a is ical analysis o in es iga e he biological a eas ha a e mos
s udied, and hose whe e esea ch needs o ﬁll in he gaps. Wi h no
p econcei ed assump ions ega ding he AOPs dis ibu ion om he
AOP-Wiki da abase, ICD-11 ca ego ies ha we e s a is ically oo
common o oo a e wi hin he AOP space we e in es iga ed.
The e o e, we ﬁ s pe o med a Chi-Squa e Goodness o Fi (χ
2
GoF) es (Pea son, 1900). Pa icula ly, we es ed whe he he
dis ibu ion o he ICD-11 ca ego ies was uni o m (H
0
),
i.e., whe he all ca ego ies we e equally dis ibu ed ac oss exis ing
AOPs in he AOP-Wiki. I he e is no disc epancy (i.e., all classes a e
s udied in he same p opo ion), he ull dis ibu ion should be
uni o m. To a oid bias in he χ
2
GoF es , AOPs belonging o he
‘unclassiﬁed’home-made amily class and hose belonging o six
ca ego ies including ‘Condi ions ela ed o sexual heal h’,‘Fac o s
inﬂuencing heal h s a us o con ac wi h heal h se ices’,
‘Symp oms, signs o clinical ﬁndings, no elsewhe e classiﬁed’,
‘Inju y, poisoning o ce ain o he consequences o ex e nal
causes’,‘Ex e nal causes o mo bidi y o mo ali y’and ‘Dec ease,
Popula ion g ow h a e’we e emo ed. Such ca ego ies a e
imp ecise and may include e ms ha a e oo di e gen om
each o he , in addi ion o hos ing non-humans AOs. They a e
also no expec ed o be equally dis ibu ed compa ed o speciﬁc
ca ego ies o a pa icula disease ype. The e o e, he dis ibu ion
was es ed on 304 AOPs sp ead o e 18 classes. Subsequen ly, we
conduc ed a pos hoc es o in es iga e which classes we e o e - o
unde - ep esen ed. In pa icula , we compu ed Habe man’s
esiduals o de e mine which classes ha e con ibu ed he mos
o he ejec ion o H
0
, leading o a non-uni o m dis ibu ion o ICD-
11 ca ego ies in AOP-Wiki (Ag es i, 2007;Sha pe, 2015). These so-
called s anda dized esiduals
i
we e compu ed o each ca ego y
acco ding o he ollowing o mula (Eq. 1):
iobse edi−expec edi


N×pi×1−pi

(1)
whe e ‘obse ed
i
’is he ac ual numbe o obse ed AOPs belonging
o a speciﬁc class i, ‘expec ed
i
’ he numbe o expec ed AOPs
belonging o he class i unde he null hypo hesis (expec ed
i
= 304/
18 ≈17 ∀i since H
0
s a es ha he dis ibu ion should be uni o m),
N he numbe o AOPs conside ed in he es (N = 304) and p
i
he
p obabili y ha an AOP d awn a andom om he 304 conside ed
o he es belongs o he class i unde he null hypo hesis (p
i
=1/
18 ∀i since H
0
s a es ha he dis ibu ion should be uni o m). The
la ge he esidual in absolu e alue, he g ea e he con ibu ion o
he ca ego y o he magni ude o he χ
2
alue ob ained and he mo e
he ca ego y was o e - o unde -s udied ( he esidual sign indica es
he di ec ion o he gap).
2.4 Case s udies
The h ee case s udies ha we e p io i ized wi hin he PARC p ojec
we e e alua ed in dep h by p esen ing AOPN om exis ing da a in he
AOP-Wiki. This iden iﬁes he s a e-o - he a o chosen cases, also
gi ing an o e iew in which a eas u u e de elopmen should ake place.
These h ee cases co espond o Immuno oxici y and Non-Geno oxic
Ca cinogenesis (Case 1), Endoc ine and Me abolic Dis up ion (Case 2),
and De elopmen al and Adul Neu o oxici y (Case 3).
The AOP-DB RDF ool (Ma ens e al., 2022) was used o ex ac
he da a o he subcases wi h cus omized que ies o each one.
Fu he mo e, key e en s common o all subcases we e connec ed,
when possible, ollowed by a g aphical isualiza ion o he AOP
ne wo k wi h Cy oscape 3.10.0 (Shannon e al., 2003).
3 Resul s
3.1 Ex ac ion and p epa a ion o he AOP
in o ma ion om he AOP-Wiki da abase
F om he AOP-Wiki da abase, we ex ac ed in o ma ion o
403 di e en AOPs, ha include 1371 unique ID e en s (Figu e 2).
O he 1371 unique ID e en s, 204 we e labeled as MIEs
(i.e., e en s which a e always e e enced as MIEs wi hin all
AOPs), 937 we e labeled as KEs and 158 as AOs. Some o he
e en s belonged o se e al ypes, as hey a e deﬁned depending
on he AOP hey a e in ol ed in and we e some imes a MIE o KE
(n = 36), KE o AO (n = 35), and e en an MIE, KE o AO in one case
(e.g., ID 1194‘Inc ease, DNA damage’which is an MIE in he AOP
ID 293, a KE in AOP ID 200 and an AO in he AOP ID 294)
(Figu e 2A). The exis ence o such e en s wi h di e en s a uses
depending on he AOP is a way o ex ending he concep o AOPs o
AOPNs (Knapen e al., 2018). The in e connec ion be ween he
same e en belonging o se e al AOPs is one o he pilla s o he
AOP-Wiki philosophy, which seeks o connec AOPs. Fo example,
he e en ID 1194 is used abou 11 imes, which allows i o be
equen ly ci ed (all s a es KE, MIE, AO- Rank 10). Ou o he
403 AOPs su eyed, 379 ha e a leas one AO. Among he 379 AOPs
wi h a leas 1 AO, only 28 ha e he ‘WPHA/WNT Endo sed’s a us
(~7%) (Figu e 2B). The AOP ID 21 ‘A yl hyd oca bon ecep o
ac i a ion leading o ea ly li e s age mo ali y, ia inc eased COX-2’
has his s a us bu did no ha e any AO epo ed, b inging he o al
numbe o OECD-endo sed AOPs o 29.
3.2 Global mapping o he AOP o he
biological space using gene in o ma ion
3.2.1 Da a se o he biological space explo a ion
A o al o 245 genes o p o eins we e manually ex ac ed om
he 1371 e en s con ained wi hin he 403 AOPs. A e emo ing
duplica es esul ing om he collabo a i e na u e o he AOP-Wiki
(e.g., see 2.1.1 o NR1I3), a o al o 220 unique en i ies we e kep .
Then, all 220 genes we e mapped o hei co esponding symbols o
acili a e GO and DisGeNET en ichmen analysis. Among hese,
149 genes we e di ec ly associa ed wi h a HGNC gene symbol
( ep esen ed in g een in Supplemen a y Table S1). Addi ional
14 en i ies we e manually e iﬁed by expe s (e.g., when iso o m
in o ma ion was missing) and success ully mapped ( ep esen ed in
o ange in Supplemen a y Table S1). Howe e , 57 en i ies could no
be mapped ( ep esen ed in ed in Supplemen a y Table S1). Some o
hese genes did no co espond o human genes and lacked a human
o holog (e.g., Vi ellogenin), esul ing in hei exclusion om he
F on ie s in Toxicology on ie sin.o g05
Jayle e al. 10.3389/ ox.2024.1285768

en ichmen analysis. On he o he hand, some genes could no be
mapped due o insu ﬁcien p ecision du ing he AOP de elopmen .
Fo ins ance, among he mos equen ly unmapped en i ies, p o ein
amilies such as Cy okines o Caspases we e ound in 7 and 6 AOPs,
espec i ely.
Consequen ly, he o al numbe o genes p ope ly associa ed
wi h hei HGNC symbols has been aised o 163, being he s a ing
poin o bo h GO and DisGeNET analysis.
3.2.2 Gene and pa hways explo a ion
Among he genes p ominen ly ep esen ed in ou ﬁnal lis , (n =
163), he Angio ensin Con e ing Enzyme 2 (ACE2) and A yl
Hyd oca bon Recep o (AHR o AhR) a e he mos equen ly
men ioned in he AOP-Wiki, appea ing in mo e han 17 dis inc
AOPs (Supplemen a y Figu e S1). The p ominen p esence o hese
p o eins in nume ous oxici y pa hways can be explained by hei
impo an biological oles. Fo example, ACE2, being a mul i unc ional
p o ein (Tu ne , 2015), is ound in a ious AOPs ela ed o di e se
o gans (e.g., pulmona y ﬁb osis in AOP ID 319, enal dys unc ion in
AOP ID 384, and ca diac dys unc ion in AOP ID 427). Mo eo e , a
wide ange o AOPs including ACE2 a e ela ed o he SARS-CoV-
2 i us (COVID-19) (h ps://www.ciao-co id.ne /aops), which has been
ex ensi ely s udied in ecen yea s (AOP IDs 379, 426, 430, 468), as he
i us uses ACE2 as a cellula en y poin o in ec ion (Beye s ed e al.,
2021). The AhR is a cy osolic ansc ip ion ac o ha plays a signiﬁcan
ole in a ious cellula p ocesses, including xenobio ic me abolism. I is
p ima ily associa ed wi h en i onmen al pollu an s, such as dioxins and
polycyclic a oma ic hyd oca bons. Ac i a ion o he AhR is linked o
me abolic diso de s (e.g., AOP ID 57) and he p og ession o speciﬁc
cance ypes, which explains i s p esence in a ious cance ela ed AOPs
(AOP IDs 416, 417, 420, and 439) (Wang e al., 2020). Howe e , he
AhR is also egula ed by endogenous compounds including yp ophan
me aboli es, by mic oo ganisms o ul a iole ligh (F i sche e al., 2007;
Rannug, 2022). This will likely inc ease u he he numbe o AOPs in
which i is in ol ed.
A e he GO en ichmen o he 163 mapped genes, a o al o
963 signiﬁcan e ms (adjus ed p- alue <0.05) we e ob ained o BP,
87 o MF, and 60 o CC (Supplemen a y Table S3). Among he mos
ele an BPs iden iﬁed using Re iGO, a subs an ial ep esen a ion o
me abolism- ela ed p ocesses was obse ed (e.g., glucose homeos asis,
egula ion o choles e ol s o age, s e oid me abolic p ocess, lipop o ein
anspo ). Addi ionally, he e is a no able p esence o BPs commonly
associa ed wi h cance , such as cellula egula ion (e.g., cell
di e en ia ion, egula ion o apop o ic p ocess, egula ion o
miRNA ansc ip ion, egula ion o cell p oli e a ion), as well as
p ocesses ela ed o oxida i e s ess and inﬂamma ion (Figu e 3).
Among he mos equen ly obse ed genes in me abolism- ela ed
p ocesses, examples include AhR and LXR in AOPs associa ed wi h
s ea osis (e.g., AOP IDs 34,57, 58 and 232), PPARαin AOPs desc ibing
lipid me abolism al e a ions (e.g., AOP ID 166), and ERαin an AOP
ela ed o obesi y (AOP ID 493). Fo p ocesses in ol ed in umo
p og ession (cellula egula ion, oxida i e s ess, inﬂamma ion, e c.),
nume ous genes a e implica ed in umo p og ession (e.g., ERs
con ibu e o dis up ions in p oli e a ion, apop osis, and/o
oxida i e s ess in AOP IDs 167 and 200). AhR ac i a ion leads o
inc eased inﬂamma ion and/o apop osis in AOP IDs 419 and 439,
dis up ions in cell p oli e a ion in AOP ID 420, and he accumula ion
o ROS in AOP IDs 418 and 420). This end was conﬁ med by he
esul s ob ained o MF. Among he mos ele an en ichmen esul s,
we pa icula ly ound ‘choles e ol binding’,‘lipid binding’,‘cy okine
ecep o binding’,‘oxido educ ase ac i i y’,and‘molecula unc ion
egula o ac i i y’, which align wi h he iden iﬁed BPs ela ed o
me abolism and cellula egula ion (Supplemen a y Figu e S2).
Addi ionally, he e a e esul s such as ‘nuclea ecep o ac i i y’and
‘signaling ecep o ac i a o ac i i y’, which co espond o genes and
p o eins equen ly ound ac oss all AOPs, such as s e oid ecep o s
(and ogens and es ogens), LXRs, PXR, CAR, and o he s.
3.2.3 Gene and disease explo a ion
To complemen he wo k on biological pa hways pe o med
abo e, a disease en ichmen analysis was conduc ed using he
DisGeNET da abase, based on he 163 mapped genes
(Supplemen a y Table S1). The goal o his analysis was o gain
u he insigh in o he biological implica ions o he gene lis o
expand ou unde s anding o disease associa ions wi h AOPs. Fo
he ﬁ s analysis, a disease en ichmen analysis was pe o med, and a
o al o 1215 signiﬁcan ly en iched diseases (FDR <0.05) we e
iden iﬁed. These 1215 diseases g oup in o 165 unique DisGeNET
disease classes, based on he MeSH hie a chy. The comple e lis o
signiﬁcan ly en iched diseases can be ound in Supplemen a y Table
S4, and Supplemen a y Figu e S3. In he second analysis, he
gene2disease unc ion was used o explo e gene-disease
FIGURE 2
The AOP-Wiki da a. Cu en da a ela ed o submi ed AOPs (endo sed o no ) in he AOP-Wiki da abase, as o 22 May 2023. (A) G een ci cle: e en s
ha a e conside ed as a MIE wi hin all AOPs whe e hey a e epo ed, ed ci cle: e en s ha a e conside ed as AO wi hin all AOPs whe e hey a e epo ed,
o ange c escen : e en s ha a e some imes KE, some imes AO. (B) Ou e ci cle: o al numbe o AOPs in he AOP-wiki, in e media e ci cle: numbe o
AOPs wi h a leas one AO, inne ci cle: o al numbe o OECD-endo sed AOP (‘WPHA/WNT Endo sed’s a us) wi h a leas one AO.
F on ie s in Toxicology on ie sin.o g06
Jayle e al. 10.3389/ ox.2024.1285768
associa ions and isualize he ela ionships be ween he genes in he
lis and he MeSH disease classes hey a e associa ed wi h. Gene-
disease associa ions we e compu ed indi idually, yielding
pe cen ages o measu e gene in ol emen wi hin speciﬁc disease
classes ela i e o o e all disease associa ions. Subsequen
in es iga ion ocused on associa ions whe e gene in ol emen
su passed a 60% h eshold, emphasizing p onounced connec ions
be ween genes and speciﬁc diseases. The esul ing cho d diag am
(Supplemen a y Figu e S4) e ec i ely illus a es he associa ions
be ween DisGeNET diseases and HGNC genes. In e es ingly, he
genes ha a e s ongly connec ed wi h he DisGeNET disease classes
a e di e en om hose ha a e mos equen ly ound in he AOPs.
This sugges s ha many o he DisGeNET disease classes
ep esen ed in he gene-disease associa ions a e no adequa ely
cap u ed by he exis ing AOPs, highligh ing po en ial a eas whe e
he AOP amewo k could beneﬁ om u he de elopmen . One
eason o his migh be axa speciﬁci ies. Fo example, AOP ID
41 ‘Sus ained AhR Ac i a ion leading o Roden Li e Tumou s’is a
e y well deﬁned AOP es ablished in oden s which is no di ec ly
ans e able o he human sys em. A e applying he 60% h eshold
ﬁl e , only 8 genes we e ela ed o 6 disease classes (Supplemen a y
Figu e S4). This highligh s he alue o pe o ming a disease
en ichmen analysis o unco e no el associa ions and be e
unde s and he links be ween he AOPs and a ious diseases. By
iden i ying diseases ha a e signiﬁcan ly en iched in ela ion o he
genes in ol ed in AOPs, we gain insigh s in o po en ial AOs ha
may no ha e been p e iously conside ed. This in o ma ion can
con ibu e o a mo e comp ehensi e unde s anding o he ad e se
e ec s associa ed wi h he molecula e en s in AOPs and aid in
eﬁning isk assessmen and mi iga ion s a egies.
3.3 Global mapping o he AOP o he
disease space using AO in o ma ion
3.3.1 P elimina y eedback o MIEs, KEs and AOs
The mos s udied MIEs a e bo h he ac i a ion o he AhR (ID
18) and a deposi ion o ene gy (ID 1686) wi h 17 occu ences
each, ollowed by ‘Binding o ACE2’(15 occu ences)
(Supplemen a y Figu e S5). As AhR and ACE2 a e p o eins,
his was consis en wi h he esul s looking a he mos
s udied p o eins (see 3.2.2), sugges ing he e ha he majo i y
o MIEs in ol e a de egula ion o a p o ein ( h ough ac i a ion o
inhibi ion). Indeed, among he 30 mos common MIEs iden iﬁed,
21 ha we e di ec ly p o ein-based.
Fo he KEs, he e en ‘Thy oxine (T4) in se um, Dec eased’(ID 281)
is he mos s udied, wi h 24 occu ences ollowed by ‘Thy oid ho mone
syn hesis, Dec eased’(ID 277; 18 occu ences) and ‘Cell inju y/dea h’(ID
55; 17 occu ences) (Supplemen a y Figu e S6). Unsu p isingly, he e
was a g ea e dispa i y o e en s which can be unde s ood because an
AOP is supposed o ha e only one (o wo o e en h ee) MIE(s) bu
se e al KEs a a ious o ganiza ional le els. ACE2- ela ed e en s such as
‘SARS-Co -2 cell en y’(nin h place, 10 occu ences) can s ill be ound
among he op 30 kEs as well as cance - ela ed and inﬂamma ion- ela ed
FIGURE 3
Rep esen a ion o he mos ele an biological p ocess (BP) a e clus e ing based on seman ic simila i y and mul idimensional scaling using Re iGO.
The size o he poin s is p opo ional o he size o each BP e m, while he colo ep esen s he adjus ed p- alue (log10). The ollowing se ings we e
used: De aul se ings; Wo k Wi h (Homo Sapiens); Resul ing lis (Tiny).
F on ie s in Toxicology on ie sin.o g07
Jayle e al. 10.3389/ ox.2024.1285768
e en s such as ‘Inadequa e DNA epai ’(nin h place, 10 occu ences),
‘Inc ease, Cell P oli e a ion’(11 h place, 8 occu ences) o ‘Induc ion,
Epi helial Mesenchymal T ansi ion’(12 h place, 7 occu ences) and
‘Inc eased, sec e ion o p oinﬂamma o y media o s’(eigh h place,
11 occu ences) o ‘Inc eased, ec ui men o inﬂamma o y cells’(11 h
place, 8 occu ences) espec i ely.
The dis ibu ion o he 30 mos common AOs is shown in
Supplemen a y Figu e S7 (see Supplemen a y Table S2 shee 4 ull_
dis ibu ion AOs o he comple e lis ). I appea s ha he e en ID
360 ‘Dec ease Popula ion G ow h Ra e’is he mos ep esen ed in he
AOP-Wiki (57 occu ences), ollowed by ‘Inc ease Mo ali y’
(37 occu ences) and ‘Dea h/Failu e, Colony’(21 occu ences),
hus eﬂec ing he ﬁeld o eco oxicology. In non-mammalian
species, dec eases in popula ion g ow h a es a e s ongly causally
linked wi h diso de s ela ed o ep oduc ion since 27 ou o
57 AOPs hos ing his AO co-men ioned e ms such as
‘Impai men o ep oduc i e capaci y’,‘Dec eased ecundi y’o
‘Reduc ion, Cumula i e ecundi y o spawning’. Tha he AOP
concep was ini ially es ablished o eco oxicological ques ions
(Ankley e al., 2010) migh explain he o e ep esen a ion o
eco oxicological AOs. Among he op 30 AOs, we also saw
2 e ms ela ed o cance (‘Lung cance ’, 5 occu ences; ‘N/A,
B eas Cance ’, 4 occu ences) which ep esen he wo mos
common cance s wo ldwide (Ma iuzzi and Lippi, 2019). We
obse ed he p esence o diseases ela ed o me abolism
(‘Inc eased, Li e S ea osis’, 9 occu ences; ‘Occu ence, Kidney
oxici y’, 5 occu ences) o neu ological ype (‘Cogni i e Func ion,
Dec eased’, 12 occu ences; ‘Occu ence, Epilep ic seizu e’,
8 occu ences). These mos ep esen ed AOs a e in line wi h he
GO en ichmen o he 163 mapped genes since hey ﬁ wi h BP
ela ed o cance and me abolism.
He e we would like o b ing o he eade ’s a en ion ha he
equency dis ibu ion o KEs published in he AOP-Wiki,
simila o he gene al published li e a u e, does no ep esen
hei impo ance o human diseases due o unding and
publica ion bias caused by he ‘s ee ligh e ec ’(Newquis
e al., 2015;E ans, 2020).
3.3.2 Mapping he AOPs o he disease space
Among he 149 AOs iden iﬁed in Sec ion 2.1.2, we we e able o
classi y only 127 AOs using he ICD-11 classiﬁca ion. To ha e he
mos comple e o e iew o he diseases p esen in he di e en
AOPs al eady de eloped, he 22 unclassiﬁed AOs using ICD-11 we e
checked by expe s, and when needed new classes we e c ea ed. A
o al o ou new classes we e p oposed: i) one ela ed o
ep oduc ion, ii) one ela ed o popula ion, iii) one o cell
damages and i ) a las one o all he AOs ha we e no ﬁ ing
wi hin all hese classes (Figu e 4)(Supplemen a y Table S2 shee
1ICD_11 classiﬁca ion). We no iced ha some AOPs we e
connec ed o se e al AOs. This led o a o al o 480 AOs
dissemina ed in 379 AOPs and AOPs wi h mo e han one AO
we e classiﬁed in o se e al ca ego ies. Fo example, he AOP ID
402 had 2 AOs including ‘Pe i en icula he e o opia o ma ion’and
‘Occu ence, Epilep ic seizu e’and has been hus classiﬁed in he
‘De elopmen al anomalies’and ‘Diseases o he ne ous sys em’
ICD-11 ca ego ies (Supplemen a y Table S2 shee 5 AOs_in_AOPs).
In he p esen s udy, AOPs wi hou any AO (n = 24) we e no
in eg a ed in he classiﬁca ion.
I appea s ha diseases o he geni ou ina y sys em a e he mos
s udied, ollowed by he AO ‘Dec ease, Popula ion G ow h Ra e’,
neoplasms and de elopmen al anomalies. The AO ‘Dec ease,
Popula ion G ow h Ra e’is so widely s udied ha i is p esen in
mo e AOPs han any o he ICD-11 classes (excluding ‘Diseases o
he geni ou ina y sys em’) ha ye g oup oge he se e al AOs. I is
no iceable ha only 7 AOPs e e o diseases o he immune sys em
bu 2 ou o 7 o hem a e OECD-endo sed while 21 AOPs e e o
diseases o he espi a o y sys em bu 0 ou o 21 a e OECD-
endo sed (Figu e 4).
As s a is ical es s a e ca ied ou on human diseases, he
pseudo-ca ego y ‘Dec ease, Popula ion G ow h Ra e’(n =
FIGURE 4
Classiﬁca ion o he 379 AOPs ha men ion a leas one AO using he ICD-11 sys em. The classiﬁca ion was pe o med using he AO in o ma ion associa ed
wi h he exis ing AOP in he AOP-Wiki da abase. The numbe a he end o each ow ep esen s he o al numbe o AOPs (endo sed o no ) while he numbe in
b acke s ep esen s he numbe o AOPs wi h he ‘WPHA/WNT Endo sed’s a us. The sum does no equal 379 because an AOP can be classiﬁed in o se e al
ca ego ies (1 AOP L1 AO). The associa ed OECD-endo sed AOP IDs a e lis ed in Supplemen a y Table S2 shee 3 clus e ing_AOPs.
F on ie s in Toxicology on ie sin.o g08
Jayle e al. 10.3389/ ox.2024.1285768
57 AOPs) has been subsequen ly uled ou , in addi ion o he
conside a ions se ou in poin 2.3.2. Mo eo e , AOPs belonging
o classes ‘unclassiﬁed’(n = 74), ‘Condi ions ela ed o sexual heal h’
(n = 0), ‘Fac o s inﬂuencing heal h s a us o con ac wi h heal h
se ices’(n = 4), ‘Symp oms, signs o clinical ﬁndings, no elsewhe e
classiﬁed’(n = 0), ‘Inju y, poisoning o ce ain o he consequences
o ex e nal causes’(n = 1) and ‘Ex e nal causes o mo bidi y o
mo ali y’(n = 40) ha e also been emo ed in line wi h he
hypo hesis adop ed (2.3.2). Finally, s a is ical es s we e ca ied
ou on he 304 AOPs sp ead o e 18 classes ha ﬁ he assump ions.
3.4 Gaps
To iden i y gaps (i.e., AOPs ha we e no ye de eloped o
poo ly de eloped o some disease ca ego ies), s a is ical analysis
was pe o med using he p e ious classiﬁca ion wi h ICD-11. A χ
2
GoF es ollowed by a pos hoc es ( esidual analysis) was
pe o med o in es iga e which classes we e o e - o unde -
ep esen ed (Figu e 5;Supplemen a y Table S2 shee 7 gaps)
acco ding o he p esumed dis ibu ion (2.3.2).
The dis ibu ion o he 18 ICD-11 ca ego ies in AOP-Wiki
conside ed o he χ
2
GoF es was widely skewed (χ
2
= 423.46; p=
2.28e-79). On he one hand, he pos hoc es ( esidual analysis)
highligh ed ha 3 classes we e la gely o e s udied including
‘diseases o he geni ou ina y sys em’,‘neoplasms’and
‘de elopmen al anomalies’(p<0.0001). On he o he hand, i
highligh ed ha mos o he o he classes we e unde -s udied,
which eﬂec s he la ge dispa i y in he s udy o AOPs. Cu en ly,
he e a e only 5 ca ego ies (p>0.05) ha a e co ec ly dis ibu ed
(unde he null hypo hesis H
0
ha he dis ibu ion should be
uni o m). Fo ins ance, he esidual alue o he ‘Endoc ine,
nu i ional, and me abolic diseases’class is, ollowing he Eq. 1
in 2.3.2:
endoc ine,nu i ional o me abolic diso de s 17 −16.89
()

304 × 1
18 ×1−1
18

≈0,03
whe e 17 is he ac ual numbe o obse ed AOPs belonging o he
‘Endoc ine, nu i ional and me abolic diseases’class, 16.89 he
numbe o expec ed AOPs belonging o his class unde he null
hypo hesis (=304/18 ≈16.89 since H
0
s a es ha he dis ibu ion
should be uni o m), 304 he numbe o AOPs conside ed in he es
and 1/18 he p obabili y ha an AOP d awn a andom om he
304 conside ed o he es belongs o he ‘Endoc ine, nu i ional and
me abolic diseases’class unde he null hypo hesis.
The e o e, he highes alue in he da ase showed ha he
‘diseases o he geni ou ina y sys em’class con ibu ed he mos o
he ejec ion o H
0
and is he u hes om he uni o m dis ibu ion.
Pa icula ly, = 11.54 >0(p= 8.29 × 10
−31
) so his class was o e -
ep esen ed in he AOP-Wiki. Mo eo e , esul s showed a
conco dance wi h he 25 mos signiﬁcan diseases ound wi h he
cu a ed DisGeNET da abase (Supplemen a y Figu e S3), especially
ega ding cance (neoplasms) e ms. No only does cance ha e a
p ominen place in he AOP-Wiki da abase wi h he highes numbe
o AOs e e ing o i (n = 25, Supplemen a y Figu e S8,
Supplemen a y Table S2 shee 2 clus e ing_AOs) and he second
class he mos s udied ( = 9.29; p= 1.54 × 10
−20
), bu i is also he
ﬁeld ha ga he s he highes numbe o e ms (MIE, KE o AO) as
shown by he analysis wi h he CURATED DisGeNET da abase
(Supplemen a y Figu e S3) in addi ion o he GO en ichmen
analysis (see 3.2.2).
Howe e , a pa en ca ego y ha ﬁ s o i s heo e ical numbe o
AOPs may ha e a non-uni o m dis ibu ion wi hin i s daugh e
ca ego ies. As an example, o he ‘Endoc ine, nu i ional, and
me abolic diseases’class which is nei he o e - no unde -s udied
(p= 0.98), we no iced ha wi hin i he dis ibu ion was no uni o m
(Supplemen a y Figu e S9). The mos s udied pa hologies a e
hose ela ed o he li e such as ‘li e s ea osis’and
FIGURE 5
Imbalance in he ep esen a ion o he AOPs in AOP-Wiki using he ICD-11 disease classiﬁca ion. The quan i y plo ed he e is he bina y loga i hm o
he a io be ween he ac ual numbe o AOPs and heo e ical AOPs in AOP-Wiki. This ep esen a ion was chosen o con enience as i is clea e han he
esiduals which a e mo e sub le. Fo ins ance, he alue o 1.90 ≈2 o diseases o he geni ou ina y sys em indica es ha he e a e abou 22 = 4 as many
AOPs in his ca ego y as he e should be i he dis ibu ion o all ca ego ies was uni o m. A alue o 0 means ha he a io is equal o 1, i.e., he numbe
o ac ual AOPs pe ec ly ma ches he numbe o heo e ical AOPs. When he e we e exac ly 0 AOPs in a ca ego y (‘Sleep-wake diso de s’), we a bi a ily
se he log2 alue o −5 o a oid o bidden alues, as log2 (0) is no deﬁned. Howe e , he p- alue compu ed is associa ed wi h he esiduals and no wi h
he log2 alue o he a io. n. s = p ≥0.05, * = 0.01 ≤p<0.05, ** = 0.001 ≤p<0.01, *** = 0.0001 ≤p<0.001, **** = p <0.0001.
F on ie s in Toxicology on ie sin.o g09
Jayle e al. 10.3389/ ox.2024.1285768
ela ed KEs and MIEs absen om he DNT- ela ed AOPs in he
AOP-Wiki (e.g., binding o es ogen, and ogen, li e x, and e inoic
acid ecep o s) ep esen ano he majo da a gap p e en ing he
linkage o cogni i e diso de s obse ed in child en a e p ena al
exposu e o endoc ine dis up o s (Lupu e al., 2020;Cediel-Ulloa
e al., 2022) o e al alcohol spec um diso de (ICD-11 6A0Y)
(Pe elli e al., 2019). By ﬁlling hese da a gaps in he u u e, we aim
a p o iding a solid basis o scien iﬁc conﬁdence o egula o y
agencies ha allows applica ion o NAMs o DNT in a
egula o y con ex . Conside ing ha cu en ly es ing chemicals
o DNT is no manda o y, an AOP-based NAM app oach o
b oad-based subs ance es ing is being pu sued o p o ec he
highly sensi i e b ains o ou u u e gene a ions.
3.5.3.2 Adul neu o oxici y (ANT)
The de elopmen o an AOPN om all cu en ly a ailable
linea ANT AOPs in he AOP-Wiki equi es a ho ough analysis
and expe knowledge o clean up and ha monize he deﬁni ion o
KEs and e en AOs, as o en di e en e minologies a e used o
he same (o e y simila ) KE o AO desc ip ion, which can be
me ged. In his case s udy, he only cu a ion pe o med was o
conside only ully o pa ially de eloped AOPs as pa o he
ne wo k, excluding emp y AOPs (Figu e 8B). As a esul ,
14 AOPs (11 pa ially d a ed and 3 comple ed) e e ed o
ANT in he AOP-Wiki da abase.
Fi e AOPs a e included in he OECD wo k plan (AOP IDs 3,
10, 48, 394 and 475) and 3 o hese a e endo sed (AOP IDs 3,
10 and 48). Acco ding o he gene al ca ego y iden iﬁed in he
ICD -11, 4 AOPs (AOP IDs 48, 405, 475, 483) can be classiﬁed
unde ‘Men al and Beha iou al Diso de s’wi h pa icula
e e ence o ‘Neu ocogni i e Diso de s (6D71)’, which add ess
lea ning and memo y impai men s and e e o mo e cogni i e
domains ha ep esen a decline om he indi idual’sp e ious
le el o unc ioning (Figu e 8B). Eigh AOPs can be classiﬁed in
‘Diseases o he ne ous sys em’wi h e e ence o Pa kinson’s
disease (8A00) (AOP IDs 3 and 464), Alzheime ’sdisease(8A20)
(AOP ID 429), epilepsy o seizu es (AOP IDs 10 and 281),
diso de s o ne e oo s, plexus o pe iphe al ne es, co e ing
pe iphe al neu opa hy o senso y neu ons (AOP IDs 279 and
450) o diso de s o ol ac o y ne es (AOP ID 394) (Figu e 8B).
Th ee AOPs (AOP IDs 26, 260 and 281) do no e e o speciﬁc
disease o he ne ous sys em since add essing gene al neu o oxic
e ec s such as oxida i e damage o neu odegene a ion (nec osis
o apop osis) (Figu e 8B).
Mo o deﬁci diso de s, some o which a e coded in ICD-11,
such as a axia (8A03), dys onia (8A02), myoclonus (8A06),
cho eoa he osis (8A01) and weakness, ﬂaccid/spas ic pa alysis in
addi ion o delayed neu opa hy, psychosis and emo ional
dis u bance, isual impai men (9D90) and hea ing loss a e
among he adul ne ous sys em dys unc ions esul ing om
exposu e o oxic subs ances (Klaassen, 2018) ha a e no
cu en ly de eloped in he AOP-Wiki da abase.
A sys ema ic e iew conduc ed in 2017, co e ing 27 yea s o
li e a u e, iden iﬁed a se o key endpoin ca ego ies induced by
human neu o oxican s and associa ed wi h ANT (Masjos husmann
e al., 2018). These ca ego ies ela e o neu o ansmission
(choline gic, GABAe gic, glycine gic, glu ama e gic, ad ene gic,
se o one gic, dopamine gic, neu o ansmission in gene al), ion
channels/ ecep o s (sodium channels, po assium channels,
calcium channels, chlo ide channels, o he ecep o s), cellula
endpoin s (mi ochond ial dys unc ion/oxida i e s ess/apop osis,
edox cycling, al e ed calcium signaling, cy oskele al changes,
neu oinﬂamma ion, axonopa hies, myelin oxici y, delayed
neu opa hy, enzyme inhibi ion) (Masjos husmann e al., 2018).
MIEs and KEs associa ed wi h AOPs ha a e ully o pa ially
desc ibed in he AOP-Wiki da abase we e o ganized acco ding o
hese and ela ed ca ego ies (Supplemen a y Table S7 shee 4 ‘ANT
endpoin ca ego ies ‘and shee 5 ‘Addi ional ANT endpoin s’). I is
clea om hese ables ha domains such as ‘ion channels’and pa
o he neu o ansmi e sys em a ge ed by neu o oxican s a e
comple ely neglec ed (Supplemen a y Table S7 shee 6 ANT
endpoin s no co e ed), al hough some o hese, e.g., inc ease/
inhibi ion o dopamine gic neu o ansmission, a e desc ibed in
AOPs linked o obesi y (ID 72) epi helial umo s (ID 170),
malignan neoplasms o emale geni al o gans (ID 112) and
diseases o he geni al sys em (ID 73) (Supplemen a y Table S2).
Acco ding o he OECD AOPs De elope ’s Handbook (OECD,
2018), KEs should be desc ibed as single isola ed measu able
e en s in o de o be modula and o be used in o he AOPs.
These KEs can hus be eused o de elop AOPs ha add ess ANT.
Ano he aspec ha eme ges is he ocus o AOPs on neu ons. I
is now well accep ed ha glial cells a e key playe s in he con ol o
ne ous sys em homeos asis and ha dys unc ion o his cell g oup
plays a ole in neu ological diso de s (Jäkel and Dimou, 2017; on
Be nha di e al., 2016). Wi h he sole excep ion o
neu oinﬂamma ion, AOPs a ge ing glial cell oxici y
(i.e., as ocy es, mic oglia, oligodend ocy es/Schwann cells) in he
con ex o ANT a e lacking.
F om a egula o y poin o iew, a majo challenge o ANT is
he long- e m heal h e ec s a ising om epea ed low-le el
exposu e. Fo example, me a-analyses sugges an associa ion
be ween pes icide exposu e and neu odegene a i e diseases such
as Pa kinson’s disease, Alzheime ’s disease and amyo ophic la e al
scle osis (Malek e al., 2012;EFSA, 2013;N zani e al., 2013;Yan
e al., 2016;Sh es ha e al., 2020;And ew e al., 2021). Thus,
al hough epidemiological s udies do no p o e causali y, hey
aise conce ns and ques ions abou he adequacy o in i o
egula o y s udies o p o ide in o ma ion on complex human
heal h ou comes. Pa kinson’s disease is al eady add essed by
AOP IDs 3 (endo sed) and 464. In his con ex , o acili a e he
unc ional unde s anding o complex biological sys ems, he
cu a ion o he ANT ne wo k and he de elopmen o AOPs
add essing key s eps leading o Alzheime ’s disease, wi h
pa icula a en ion o he da a gaps desc ibed, ha e been
iden iﬁed as p io i ies in PARC.
3.5.4 Ad e se ou come pa hway ne wo k o he
h ee p io i ized endpoin s o PARC
The h ee case s udies we e de eloped in he con ex o h ee
p io i ies ha ha e been se in he PARC p ojec o AOP
de elopmen in he a eas o immuno oxici y and non-geno oxic
ca cinogenesis, endoc ine and me abolic dis up ion, and
neu o oxici y. Figu e 9 shows an AOP ne wo k combining all
AOPs ha ha e been iden iﬁed as ela ed o immuno oxici y and
non-geno oxic ca cinogenesis (yellow in Figu e 9, ela ed o Case
s udy 1), endoc ine and me abolic dis up ion (pink, ela ed o Case
F on ie s in Toxicology on ie sin.o g16
Jayle e al. 10.3389/ ox.2024.1285768

s udy 2) and neu o oxici y (blue, ela ed o Case s udy 3). Τhe e is a
pa icula ly well-connec ed a ea in he ne wo k whe e bo h he
di ec ed s ess o he KEs (shown as la ge node size in Figu e 9) and
he di ec ed be weenness o he KERs (shown as la ge edge wid h)
a e highe . S ess and be weenness conside he numbe o sho es
pa hs ha mus pass h ough he KE/KER in ques ion (Sca doni
e al., 2009;Villeneu e e al., 2018a). This a ea connec s he AOPs
a ailable in he AOP-Wiki in he h ee PARC p io i y a eas.
Neu o oxici y is s ongly connec ed o endoc ine and me abolic
dis up ion h ough educed hy oxine le els leading o educed
le els o BDNF. Immuno oxici y and non-geno oxic
ca cinogenesis is s ongly connec ed o endoc ine and me abolic
dis up ion h ough DNA damage linked o inc eased ac i a ion o
NF-kB and es ogen ecep o an agonism, e en ually esul ing in
me as a ic b eas cance (AOP ID 443). As is he case o he h ee
case s udies sepa a ely, his o e a ching AOP ne wo k analysis is
limi ed o hose AOPs ha ha e been en e ed in o he AOP-Wiki by
he communi y and o he le el o cu a ion and o e all quali y ha is
cu en ly a ailable in he AOP-Wiki (See Conclusion sec ion o
mo e de ails on ela ed challenges). As such, his analysis p o ides a
s a e o he a o he a ailable AOPs and especially o connec ions
be ween hese a eas ha ha e been desc ibed so a . I should be
no ed ha he linkages be ween he h ee case s udies ha can be
disce ned a his poin a e no necessa ily he mos impo an
linkages, om a biological and oxicological poin o iew,
among hese esea ch a eas. Addi ional linkages may be
in es iga ed in he u u e, such as a link be ween hy oid
dys unc ion and Alzheime ’s disease (Kim e al., 2022).
4 Discussion/conclusion
The AOP amewo k suppo s a be e desc ip ion o e idence-
based pa hways leading o diseases, and i p o ides obus
knowledge on c i ical key e en s and ela ionships ha can lead
o he de elopmen o no el es s and ele an bioma ke s. In his
s udy, we p o ide an o e iew o he exis ing AOPs a ailable in he
AOP-Wiki, gene alized by o e ep esen a ion analysis, highligh ing
hei signiﬁcance in ad ancing ou unde s anding o oxicological
mechanisms and in o ming isk assessmen p ac ices (Mo ensen
e al., 2022). We ha e p esen ed he gene al cha ac e is ics o o e -
and unde - ep esen ed AOPs in he AOP-Wiki and es ablished
possible connec ions o exis ing diseases (deﬁned by ICD-11
ca ego ies) h ough a combina ion o compu a ional ools and
expe cu a ion. In addi ion, we supplemen ed de ails abou h ee
case s udies, p io i ized wi hin he PARC p ojec , o shed ligh on he
ﬁndings om his compu a ional exe cise.
One o he main ou comes o his s udy is ha he e is a la ge
di e ence be ween he numbe o AOPs mapped o ce ain diseases
as compa ed o o he ones. The e may be i ial easons o ha such
as he ocus o he g oups who ﬁ s de eloped AOPs. Ye , some
c i ical diseases such as ca dio ascula , blood, skin diseases o gu
FIGURE 9
AOP ne wo k consis ing o all AOPs in en o i ed in he a eas o he h ee p io i ized endpoin s o PARC, co esponding o he h ee case s udies.
KERs ela ed o immuno oxici y and non-geno oxic ca cinogenesis (Case s udy 1) a e shown in yellow, KERs ela ed o endoc ine and me abolic
dis up ion (Case s udy 2) a e shown in pink and KERs ela ed o neu o oxici y (case s udy 3) a e shown in blue. KERs ha a e sha ed be ween
immuno oxici y and non-geno oxic ca cinogenesis and endoc ine and me abolic dis up ion a e shown in o ange. KERs ha a e sha ed be ween
neu o oxici y and endoc ine and me abolic dis up ion a e shown in pu ple. MIEs a e shown in g een and AOs a e shown in ed. Node size shows di ec ed
s ess and edge wid h shows di ec ed be weenness on uncu a ed subne wo ks, calcula ed using he Cy oscape plugin Cen iScaPe.
F on ie s in Toxicology on ie sin.o g17
Jayle e al. 10.3389/ ox.2024.1285768
diseases appea o be unde ep esen ed in he AOP-Wiki despi e
hei conside able con ibu ion o mo ali y and mo bidi y. This
does no mean ha he e o conce ning well ep esen ed diseases
such as geni ou ina y diseases o neoplasms should dec ease as we
a e s ill a om a ull desc ip ion o pa hways in hese diseases, bu
a he ha mo e e o should be de o ed o unde ep esen ed o
neglec ed diseases. A s onge in e ac ion be ween scien is s
in ol ed in hese diseases and hose in ol ed in AOP
de elopmen should be help ul.
Ano he eason o explaining he dispa i y in AOPs may s em
om he assump ion we made ega ding he heo e ical
dis ibu ion o he disease classes. Indeed, we assumed ha he
dis ibu ion would be uni o m, i.e., ha all ca ego ies would be
s ic ly equal in e ms o he AOPs ha make hem up. Howe e , as
AOPs a e a amewo k s ongly associa ed wi h oxicology and he
en i onmen su ounding us, since each MIE a ises om a
s ess ul exposu e, i may be concei able ha he unknown se
o all exis ing s esso s dis up s only a limi ed numbe o biological
unc ions, une enly, as a subse o all exis ing pa hologies. In his
case, he dis ibu ion o ICD-11 classes should no be uni o m,
a he an asymme ical dis ibu ion, o eﬂec he si ua ion mo e
accu a ely. Fo ins ance, his hypo hesis would no a ibu e such a
high nega i e weigh o he sleep wake diso de s class, which is
cu en ly signiﬁcan ly unde - ep esen ed. Acco dingly, i could
ha e a mo e mode a e weigh because his physiological unc ion is
less dis u bed by s esso s. On he o he hand, i migh be expec ed
ha geni ou ina y diseases, neoplasms, and de elopmen al
diseases a e genuinely o e - ep esen ed since s esso s lead o
hese pa hologies mo e han any o he s. Mo eo e , he unequal
dis ibu ion o AOPs wi h espec o human diseases may also be
due o he ac ha some biological spaces a e o e - ep esen ed,
such as undamen al unc ion which appea o be common and
in ol ed in a ious p ocesses. A bias in co e age o diseases,
o igina ing om a da abase used, i s e sions o g anula i y
could also be expec ed. He e we used he la es WHO da abase,
ha is a wo ld-wide ecognized, cu a ed da abase o human
diseases bu o he exis ing da abases wi h human diseases
deﬁni ion could po en ially depic sligh ly di e en pa e ns,
eﬂec ing a his o ical aspec o naming and cha ac e iza ion o
human diseases.
Fu he , ou s udy e ealed ha he genes ha a e s ongly
connec ed wi h he disease classes di e om hose ha a e mos
equen ly ound in he AOPs. The e a e se e al explana ions o his
obse a ion, bu i does sugges ha some impo an disease
pa hways in cu en AOPs a e missing and ha e o s should be
de o ed o de eloping addi ional AOPs in ol ing genes ha a e
highly connec ed o diseases. I can be u he explained by he ac
ha he s uc u e o ICD-11 and disease ca ego ies conside
di e en on ology and seman ics compa ed o AOP-Wiki (ICD-
11 is mo e ela ed o gene ic pheno ypes and clinical da a and AOP-
Wiki mo e ocused on ad e se ou comes).
Rela i e o he AOP amewo k i sel , ou analysis illus a es a
numbe o aspec s and challenges ha a e impo an in he
con ex o he cu en ly ongoing e o o inc ease he
FAIRness (Findabili y, Accessibili y, In e ope abili y, and
Reusabili y) o he AOP-Wiki (Wi weh e al., 2023)ando
oxicological da a in gene al as emphasized by he ecen ly
es ablished Eu opean in e go e nmen al ELIXIR oxicology
communi y (Ma ens e al., 2021). Fo example, o e o s
such as he p esen analysis whe e AOPs a e collec ed h ough
he AOP-Wiki and assembled in o AOP ne wo ks o be used o
me a-analyses, a s ong need o AOP (ne wo k) cu a ion a ises,
such as he me ging o synonymous KEs o educe edundancy,
he g ouping o ela ed bu no necessa ily iden ical KEs, and he
unambiguous desc ip ion and iden iﬁca ion o KERs. In his
con ex he use o on ological anno a ions in he AOP-Wiki is
becoming inc easingly impo an (I es e al., 2017;Wi weh
e al., 2023) and is en isioned as one o he co ne s ones o he
da a model ha is being de eloped o he AOP-Wiki e sion 3.0,
which aims o add ess many o hese challenges. This will
inc ease he in e ope abili y wi h o he da abases and
pla o ms and acili a e analyses such as he p esen one.
Also, he de elopmen o la ge AOP ne wo ks esul s in he
eme gence o new linea AOPs by e-using building blocks om use
deﬁned AOPs (Pollesch e al., 2019). An impo an challenge lies in
in en o ying and assessing he quali y o such eme gen AOPs. New
ools o add ess his challenge a e unde way, which will allow he
p io i iza ion o eme gen AOPs o u he manual assessmen and
conside a ion. Ano he aspec ha becomes especially impo an
when assembling AOPs in o AOP ne wo ks is how o deal wi h
domain o applicabili y (DOA) desc ip ions (cu en ly limi ed o
axonomic g oup, li e s age and sex, bu po en ially also including
o he componen s such as issue and o gan ype) ac oss
in e connec ing AOPs. The DOA has o en no been ho oughly
in es iga ed o many use s deﬁned AOPs, and de e mining he
DOA becomes e en mo e complex in he case o eme gen AOPs.
Fo example, an AOP may be es ablished in animals bu i s ele ance
o humans may no ha e been explo ed. Fo ha eason, no only
documen ing e idence o conse a ion ac oss species, bu also
including e idence ha a ce ain KE/KER is no conse ed in a
pa icula axon may become an impo an ool o analyzing la ge
AOP ne wo ks. A deepe analysis o he KER would also be
in e es ing o u he explo a ion in e ms o ele ance o
human diseases, and o gain a be e unde s anding o he
biological space.
Finally, he AOP concep se es as a powe ul ool o he
sys ema ic classiﬁca ion o knowledge, o e ing immense
po en ial wi hin he egula o y landscape. By p o iding a
s uc u ed amewo k o unde s and complex pa hways
leading o AOs ha we can connec o diseases, AOPs will
enable mo e in o med decision-making and isk assessmen .
Fu he mo e, he emphasis on iden i ying gaps in ou
unde s anding pa es he way o he a ge ed de elopmen o
assays ha add ess hese a eas. This dynamic app oach
encou ages inno a i e hinking, allowing us o explo e
uncha ed e i o ies in oxicology. As we ha ness he ull
scope o AOPs, we no only enhance ou comp ehension o
biological in e ac ions bu also shape a u u e whe e p edic i e
oxicology is bo h accu a e and comp ehensi e.
Da a a ailabili y s a emen
The da ase s p esen ed in his s udy can be ound in online
eposi o ies. The names o he eposi o y/ eposi o ies and accession
numbe (s) can be ound in he a icle/Supplemen a y Ma e ial.
F on ie s in Toxicology on ie sin.o g18
Jayle e al. 10.3389/ ox.2024.1285768
Au ho con ibu ions
TJ: In es iga ion, W i ing–o iginal d a , Da a cu a ion,
Me hodology, Visualiza ion, Fo mal Analysis. TC: Da a cu a ion,
In es iga ion, Me hodology, Visualiza ion, W i ing–o iginal d a ,
Fo mal Analysis. NS: In es iga ion, Me hodology, Visualiza ion,
W i ing– e iew and edi ing. BV: In es iga ion, W i ing– e iew and
edi ing, Me hodology, W i ing–o iginal d a . BL: In es iga ion,
W i ing– e iew and edi ing, Me hodology, Visualiza ion. LV:
In es iga ion, W i ing– e iew and edi ing, Visualiza ion,
Me hodology. OM: In es iga ion, W i ing– e iew and edi ing. NY:
W i ing– e iew and edi ing, Me hodology, So wa e, Visualiza ion.
JG: W i ing– e iew and edi ing, In es iga ion, W i ing–o iginal d a .
PM-L: W i ing– e iew and edi ing, Fo mal Analysis, Visualiza ion.
FJ: W i ing– e iew and edi ing. HH: In es iga ion, W i ing– e iew
and edi ing. XC: W i ing– e iew and edi ing. DS: W i ing– e iew and
edi ing. PA: W i ing– e iew and edi ing. AB-P: W i ing– e iew and
edi ing. EF: W i ing– e iew and edi ing. EK: W i ing– e iew and
edi ing, W i ing–o iginal d a . AS: W i ing– e iew and edi ing. RB:
W i ing– e iew and edi ing. MK: W i ing– e iew and edi ing. OT:
W i ing– e iew and edi ing. MW: In es iga ion, Me hodology,
Visualiza ion, W i ing– e iew and edi ing, Da a cu a ion, So wa e.
DK: W i ing– e iew and edi ing. KA: Concep ualiza ion,
In es iga ion, Supe ision, W i ing–o iginal d a , W i ing– e iew
and edi ing.
Funding
The au ho (s) decla e ﬁnancial suppo was ecei ed o he
esea ch, au ho ship, and/o publica ion o his a icle. This wo k
was suppo ed by he Eu opean Union’s‘Ho izon Eu ope’
amewo k unde he PARC p ojec [h ps://www.eu-pa c.eu,
g an numbe 101057014], and he Eu opean Union’s Ho izon
2020 Resea ch and Inno a ion P og amme’wi h OBERON
[h ps://obe on-4eu.com, g an numbe 825712], ONTOX
[h ps://on ox-p ojec .eu/, g an numbe 963845] and RadoNo m
[h ps://www. adono m.eu, g an numbe 900009].
Acknowledgmen s
The au ho s would like o hank he Uni e si y Pa is Ci é and
INSERM o suppo ing his wo k.
Conﬂic o in e es
Au ho AB-P was employed by Join Resea ch Cen e and
Au ho EF was employed by DNTOX GmbH.
The emaining au ho s decla e ha he esea ch was conduc ed
in he absence o any comme cial o ﬁnancial ela ionships ha
could be cons ued as a po en ial conﬂic o in e es .
Publishe ’s no e
All claims exp essed in his a icle a e solely hose o he au ho s
and do no necessa ily ep esen hose o hei a ﬁlia ed
o ganiza ions, o hose o he publishe , he edi o s and he
e iewe s. Any p oduc ha may be e alua ed in his a icle, o
claim ha may be made by i s manu ac u e , is no gua an eed o
endo sed by he publishe .
Supplemen a y ma e ial
The Supplemen a y Ma e ial o his a icle can be ound online
a : h ps://www. on ie sin.o g/a icles/10.3389/ ox.2024.1285768/
ull#supplemen a y-ma e ial
Re e ences
Abdel aou ,E.R.,Kilany,A.,Elhadidy,M.E.,Zeidan,H.M.,Elsaied,A.,Eid,O.M.,
e al. (2023). Dyslexia wi h and wi hou I len synd ome: a s udy o inﬂuence on
abili ies and b ain-de i ed neu o ophic ac o le el. Ib ain 9, 32–42. doi:10.1002/
ib a.12080
Adams, R. A., Pino sis, D., Tsi lis, K., Un uh, L., Mahajan, A., Ho as, A. M., e al.
(2022). Compu a ional modeling o elec oencephalog aphy and unc ional magne ic
esonance imaging pa adigms indica es a consis en loss o py amidal cell synap ic gain
in schizoph enia. Biol. Psychia y 91, 202–215. doi:10.1016/j.biopsych.2021.07.024
Ag es i, A. (2007). In oduc ion o ca ego ical da a analysis. 2nd ed. New Yo k: John
Wiley and Sons.
And ew, A., Zhou, J., Gui, J., Ha ison, A., Shi, X., Li, M., e al. (2021). Pes icides
applied o c ops and amyo ophic la e al scle osis isk in he U.S. U.S. Neu oToxicology
87, 128–135. doi:10.1016/j.neu o.2021.09.004
Ankley, G. T., Benne , R. S., E ickson, R. J., Ho , D. J., Ho nung, M. W., Johnson, R.
D., e al. (2010). Ad e se ou come pa hways: a concep ual amewo k o suppo
eco oxicology esea ch and isk assessmen . En i on. Toxicol. Chem. 29, 730–741.
doi:10.1002/e c.34
Ashbu ne , M., Ball, C. A., Blake, J. A., Bo s ein, D., Bu le , H., Che y, J. M., e al.
(2000). Gene on ology: ool o he uniﬁca ion o biology. The Gene On ology
Conso ium. Na . Gene . 25, 25–29. doi:10.1038/75556
Audebe , M., Assmann, A.-S., Azque a, A., Babica, P., Ben ena i, E., Bo oli, S., e al.
(2023). New app oach me hodologies o acili a e and imp o e he haza d assessmen o
non-geno oxic ca cinogens—a PARC p ojec . F on . Toxicol. 5, 1220998. doi:10.3389/
ox.2023.1220998
Audouze, K., Sa igiannis, D., Alonso-Magdalena, P., B ocho , C., Casas, M., V ijheid,
M., e al. (2020). In eg a i e s a egy o es ing sys ems o iden iﬁca ion o endoc ine
dis up o s inducing me abolic diso de s—an in oduc ion o he OBERON p ojec . In .
J. Mol. Sci. 21, 2988. doi:10.3390/ijms21082988
Audouze, K., Zgheib, E., Abass, K., Baig, A. H., Fo ne -Pique , I., Holbech, H., e al.
(2021). E idenced-based app oaches o suppo he de elopmen o endoc ine-
media ed ad e se ou come pa hways: challenges and oppo uni ies. F on . Toxicol. 3,
787017. doi:10.3389/ ox.2021.787017
Babin, É., Cano-Sancho, G., Vigneau, E., and An ignac, J.-P. (2023). A e iew o
s a is ical s a egies o in eg a e bioma ke s o chemical exposu e wi h bioma ke s o
e ec applied in omic-scale en i onmen al epidemiology. En i on. Pollu . 330, 121741.
doi:10.1016/j.en pol.2023.121741
Babu amani, A. A., Von ell, R. T., Uus, A., Pie sch, M., Pa kee, P. A., Wya -Ashmead,
J., e al. (2020). Assessmen o adial glia in he on al lobe o e uses wi h Down
synd ome. Ac a Neu opa hol. Commun. 8, 141. doi:10.1186/s40478-020-01015-3
Baja d, L., Adamo sky, O., Audouze, K., Baken, K., Ba ouki, R., Bel man, J. B., e al.
(2023). Applica ion o AOPs o assis egula o y assessmen o chemical isks - case
s udies, needs and ecommenda ions. En i on. Res. 217, 114650. doi:10.1016/j.en es.
2022.114650
Bal-P ice, A., Meek, M. E., and Be e), (2017). Ad e se ou come pa hways:
applica ion o enhance mechanis ic unde s anding o neu o oxici y. Pha macol.
The . 179, 84–95. doi:10.1016/j.pha m he a.2017.05.006
Ba enys, M., Re e e, I., Masjos husmann, S., Gómez-Ca alán, J., and F i sche, E.
(2020). De elopmen al neu o oxici y o MDMA. A sys ema ic li e a u e e iew
F on ie s in Toxicology on ie sin.o g19
Jayle e al. 10.3389/ ox.2024.1285768
summa ized in a pu a i e ad e se ou come pa hway. Neu oToxicology 78, 209–241.
doi:10.1016/j.neu o.2019.12.007
Benoi , L., Jo nod, F., Zgheib, E., Tomkiewicz, C., Koual, M., Cous ille , T., e al.
(2022). Ad e se ou come pa hway om ac i a ion o he AhR o b eas cance - ela ed
dea h. En i on. In . 165, 107323. doi:10.1016/j.en in .2022.107323
Beye s ed ,S.,Casa o,E.B.,andRangel,É.B.(2021).COVID-19:angio ensin-con e ing
enzyme 2 (ACE2) exp ession and issue suscep ibili y o SARS-CoV-2 in ec ion. Eu . J. Clin.
Mic obiol. In ec . Dis. 40, 905–919. doi:10.1007/s10096-020-04138-6
B en a, G. (2011). Why can insulin esis ance Be a na u al consequence o hy oid
dys unc ion? J. Thy oid. Res. 2011, 152850. doi:10.4061/2011/152850
Ca aillo, J.-C., Ba ouki, R., Coumoul, X., and Audouze, K. (2019). Linking bisphenol
S o ad e se ou come pa hways using a combined ex mining and sys ems biology
app oach. En i on. Heal h Pe spec . 127, 47005. doi:10.1289/EHP4200
Cediel-Ulloa, A., Lupu, D. L., Johansson, Y., Hinojosa, M., Özel, F., and Rüegg, J.
(2022). Impac o endoc ine dis up ing chemicals on neu ode elopmen : he need o
be e es ing s a egies o endoc ine dis up ion-induced de elopmen al neu o oxici y.
Expe Re . Endoc inol. Me abolism 17, 131–141. doi:10.1080/17446651.2022.2044788
Con ad, N., Mis a, S., Ve bakel, J. Y., Ve beke, G., Molenbe ghs, G., Taylo , P. N.,
e al. (2023). Incidence, p e alence, and co-occu ence o au oimmune diso de s o e
ime and by age, sex, and socioeconomic s a us: a popula ion-based coho s udy o
22 million indi iduals in he UK. Lance 401, 1878–1890. doi:10.1016/S0140-6736(23)
00457-9
C es o, N., Pille , L.-E., Billua , P., and Rouach, N. (2019). Do as ocy es play a ole in
in ellec ual disabili ies? T ends Neu osci. 42, 518–527. doi:10.1016/j. ins.2019.05.011
C o on, K. M., and Mundy, W. R. (2021). Ex e nal scien iﬁc epo on he
in e p e a ion o da a om he de elopmen al neu o oxici y in i o es ing assays
o use in in eg a ed app oaches o es ing and assessmen . EFSA Suppo . Publ. 18,
6924E. doi:10.2903/sp.e sa.2021.EN-6924
Da k, C., Homman-Ludiye, J., and B yson-Richa dson, R. J. (2018). The ole o
ADHD associa ed genes in neu ode elopmen . De . Biol. 438, 69–83. doi:10.1016/j.
ydbio.2018.03.023
de G oo , D. M. G., Ha g ing, S., an de Ho s , L., Moe kens, M., O o, M., Bos-
Kuijpe s, M. H. M., e al. (2005). 2D and 3D assessmen o neu opa hology in a b ain
a e p ena al exposu e o me hylazoxyme hanol, a model o de elopmen al
neu o oxic y. Rep oduc i e Toxicol. 33 d Annu. Con . Eu . Te a ol. Soc. 20, 417–432.
doi:10.1016/j. ep o ox.2005.04.006
de Lucca, M. S., Pimen el, M. E. O., Raimundo, C. K. O., Hen iques, B. D., Mo ei a, T. R.,
Ca doso, S. A., e al. (2023). B ain-de i ed neu o ophic ac o (BDNF) le els in child en and
adolescen s be o e and a e s imulan use a sys ema ic e iew and me analysis. P og. Neu o-
Psychopha macology Biol. Psychia y 125, 110761. doi:10.1016/j.pnpbp.2023.110761
Deng,Z.,Zhou,X.,Lu,J.-H.,andYue,Z.(2021).Au ophagydeﬁciency in
neu ode elopmen al diso de s. Cell. and Biosci. 11, 214. doi:10.1186/s13578-021-00726-x
Di Cosmo, C., De Ma co, G., Ag e i, P., Fe a ini, E., Dimida, A., Falce a, P., e al.
(2022). Sc eening o d ugs po en ially in e e ing wi h MCT8-media ed T3 anspo
in i o iden iﬁes dexame hasone and some commonly used d ugs as inhibi o s o
MCT8 ac i i y. J. Endoc inol. In es ig. 45, 803–814. doi:10.1007/s40618-021-01711-4
EFSA (2013). Scien iﬁc Opinion on he iden iﬁca ion o pes icides o be included in
cumula i e assessmen g oups on he basis o hei oxicological p oﬁle. EFSA J. 11,
3293. doi:10.2903/j.e sa.2013.3293
EPA (1998). Guidelines o neu o oxici y isk assessmen .
Esna oglu, E., and Adıgüzel, Ö. (2021). Associa ion o BDNF le els wi h IQ:
compa ison o S100B and BDNF le els in ypically de eloping child en and subjec s
wi h neu ologically no mal nonsynd omic in ellec ual disabili y. J. In ellec . Disabil. Res.
65, 1073–1084. doi:10.1111/ji .12896
Eu opean Chemicals Agency (2023). E alua ing esul s om 55 ex ended one-
gene a ion ep oduc i e oxici y s udies unde REACH: ﬁnal epo o he EOGRTS
e iew p ojec : ma ch 2023. USA: Publica ions O ﬁce.
E ans, B. J. (2020). The s ee ligh e ec : egula ing genomics whe e he ligh is.
J. Law, Med. E hics 48, 105–118. doi:10.1177/1073110520916998
FDA (2000). Toxicological p inciples o he sa e y assessmen o ood ing edien s
edbook 2000. Chap e IV.C.10.
F ega,M.,Sel en,M.,Mossink,B.,Kelle ,J.M.,Linda,K.,Moe schen,R.,e al.(2020).
Dis inc pa hogenic genes causing in ellec ual disabili y and au ism exhibi a common neu onal
ne wo k hype ac i i y pheno ype. Cell. Rep. 30, 173–186. doi:10.1016/j.cel ep.2019.12.002
F i sche, E., Schä e , C., Calles, C., Be nsmann, T., Be nshausen, T., Wu m, M., e al.
(2007). Ligh ening up he UV esponse by iden iﬁca ion o he a ylhyd oca bon
ecep o as a cy oplasma ic a ge o ul a iole B adia ion. P oc. Na l. Acad. Sci.
104, 8851–8856. doi:10.1073/pnas.0701764104
Gao, H., Chen, D., and Zang, M. (2021). Associa ion be ween ph hala e exposu e and
insulin esis ance: a sys ema ic e iew and me a-analysis upda e. En i on. Sci. Pollu .
Res. In . 28, 55967–55980. doi:10.1007/s11356-021-16252-9
G oes linge , J., Seidl, C., Va ga, E., Fa e o, G. D., and Ma ko, D. (2022).
Combina o y xposu e o U oli hin A, Al e na iol, and Deoxyni alenol a ec s colon
cance me abolism and epi helial ba ie in eg i y in i o.F on Nu . 9, doi:10.3389/
nu .2022.882222
Haigis, A.-C., Ve gauwen, L., LaLone, C. A., Villeneu e, D. L., O’B ien, J. M., and
Knapen, D. (2023). C oss-species applicabili y o an ad e se ou come pa hway ne wo k
o hy oid ho mone sys em dis up ion. Toxicol. Sci. k ad063 195, 1–27. doi:10.1093/
oxsci/k ad063
Halpe in,D.,S a sky,A.,Kadi ,R.,D abkin,M.,Wo mse ,O.,Yoge ,Y.,e al.(2021).
CDH2 mu a ion a ec ing N-cadhe in unc ion causes a en ion-deﬁci hype ac i i y diso de
in humans and mice. Na . Commun. 12, 6187. doi:10.1038/s41467-021-26426-1
Hanahan, D., and Weinbe g, R. A. (2011). Hallma ks o cance : he nex gene a ion.
Cell. 144, 646–674. doi:10.1016/j.cell.2011.02.013
Ha de , L., Dudazy-G alla, S., Mülle -Fieli z, H., Hje ling Le ﬂe , J., Venns öm, B.,
Heue , H., e al. (2018). Ma e nal hy oid ho mone is equi ed o pa albumin neu one
de elopmen in he an e io hypo halamic a ea. J. Neu oendoc inol. 30, e12573. doi:10.
1111/jne.12573
Heindel, J. J., and Blumbe g, B. (2019). En i onmen al obesogens: mechanisms and
con o e sies. Annu. Re . Pha macol. Toxicol. 59, 89–106. doi:10.1146/annu e -
pha m ox-010818-021304
Hogbe g, H. T., and Smi no a, L. (2022). The u u e o 3D b ain cul u es in
de elopmen al neu o oxici y es ing. F on . Toxicol. 4, 808620. doi:10.3389/ ox.
2022.808620
I es, C., Campia, I., Wang, R.-L., Wi weh , C., and Edwa ds, S. (2017). C ea ing a
s uc u ed AOP knowledgebase ia on ology-based anno a ions. Appl. Vi o Toxicol. 3,
298–311. doi:10.1089/ai .2017.0017
Jäkel, S., and Dimou, L. (2017). Glial cells and hei unc ion in he adul b ain: a
jou ney h ough he his o y o hei abla ion. F on . Cell. Neu osci. 11, 24. doi:10.3389/
ncel.2017.00024
Jayle , T., Cous ille , T., Jo nod, F., Ma ga i e-Jeannin, P., and Audouze, K. (2023).
AOP-helpFinde 2.0: in eg a ion o an e en -e en sea ches module. En i on. In . 177,
108017. doi:10.1016/j.en in .2023.108017
Jayle , T., Quin ens, R., Beno mane, M. A., Luukkonen, J., Tanaka, I. B., Ibanez, C.,
e al. (2022). De elopmen o an ad e se ou come pa hway o adia ion-induced
mic ocephaly ia expe consul a ion and machine lea ning. In . J. Radia . Biol. 98,
1752–1762. doi:10.1080/09553002.2022.2110312
Jo nod, F., Jayle , T., Blaha, L., Sa igiannis, D., Tamisie , L., and Audouze, K. (2022).
AOP-helpFinde webse e : a ool o comp ehensi e analysis o he li e a u e o
suppo ad e se ou come pa hways de elopmen . Bioin o ma ics 38, 1173–1175.
doi:10.1093/bioin o ma ics/b ab750
Kim,J.H.,Lee,H.S.,Kim,Y.H.,Kwon,M.J.,Kim,J.-H.,Min,C.Y.,e al.
(2022). Inc eased isk o empo omandibula join diso de in os eopo osis
pa ien s: a longi udinal s udy. F on . Endoc inol. 13, 835923. doi:10.3389/
endo.2022.835923
Kimbe , I., Baske e , D. A., Ge be ick, G. F., Ryan, C. A., and Dea man, R. J. (2011).
Chemical alle gy: ansla ing biology in o haza d cha ac e iza ion. Toxicol. Sci. 120
(Suppl. 1), S238–S268. doi:10.1093/ oxsci/k q346
Klaassen, C. D. (2018). Casa e and doull’s oxicology: he basic science o poisons.
A ailable a : h ps://accesspha macy.mhmedical.com/book.aspx?bookID=2462
(Accessed Augus 21, 23).
Klose,J.,Pahl,M.,Ba mann,K.,Bend ,F.,Blum,J.,Dolde,X.,e al.(2022).
Neu ode elopmen al oxici y assessmen o ﬂame e a dan s using a human DNT
in i o es ing ba e y. Cell. Biol. Toxicol. 38, 781–807. doi:10.1007/s10565-021-
09603-2
Knapen, D., Ang ish, M. M., Fo in, M. C., Ka siadaki, I., Leona d, M., Ma gio a-
Casaluci, L., e al. (2018). Ad e se ou come pa hway ne wo ks I: de elopmen and
applica ions. En i on. Toxicol. Chem. 37, 1723–1733. doi:10.1002/e c.4125
Ko e aa , T. I. M., Mue zel, R., Medici, M., Chake , L., Jaddoe, V. W. V., de Rijke, Y.
B., e al. (2016). Associa ion o ma e nal hy oid unc ion du ing ea ly p egnancy wi h
o sp ing IQ and b ain mo phology in childhood: a popula ion-based p ospec i e
coho s udy. Lance Diabe es Endoc inol. 4, 35–43. doi:10.1016/S2213-8587(15)
00327-7
K ewski, D., Ande sen, M. E., Tyshenko, M. G., K ishnan, K., Ha ung, T.,
Boekelheide, K., e al. (2020). Toxici y es ing in he 21s cen u y: p og ess in he
pas decade and u u e pe spec i es. A ch. Toxicol. 94, 1–58. doi:10.1007/s00204-019-
02613-4
Lam, T. M., Wag endonk, A. J., den B a e , N. R., Ka ssenbe g, D., Vaa jes, I.,
Timme mans, E. J., e al. (2023). De elopmen o a neighbo hood obesogenic buil
en i onmen cha ac e is ics index o The Ne he lands. Obesi y 31, 214–224. doi:10.
1002/oby.23610
Leis , M., Ghallab, A., G aepel, R., Ma chan, R., Hassan, R., Bennekou, S. H., e al.
(2017). Ad e se ou come pa hways: oppo uni ies, limi a ions and open ques ions.
A ch. Toxicol. 91, 3477–3505. doi:10.1007/s00204-017-2045-3
Liu, S.-H., Shi, X.-J., Fan, F.-C., and Cheng, Y. (2021). Pe iphe al blood neu o ophic
ac o le els in child en wi h au ism spec um diso de : a me a-analysis. Sci. Rep. 11, 15.
doi:10.1038/s41598-020-79080-w
Lupu, D., Ande sson, P., Bo nehag, C.-G., Demeneix, B., F i sche, E., Gennings, C.,
e al. (2020). The ENDpoiNTs p ojec : no el es ing s a egies o endoc ine dis up o s
linked o de elopmen al neu o oxici y. In . J. Mol. Sci. 21, 3978. doi:10.3390/
ijms21113978
F on ie s in Toxicology on ie sin.o g20
Jayle e al. 10.3389/ ox.2024.1285768
Mai e, L., Jedynak, P., Gallego, M., Cia an, L., Audouze, K., Casas, M., e al. (2023).
In eg a ing -omics app oaches in o popula ion-based s udies o endoc ine dis up ing
chemicals: a scoping e iew. En i on. Res. 228, 115788. doi:10.1016/j.en es.2023.115788
Malek, A. M., Ba chowsky, A., Bowse , R., Youk, A., and Talbo , E. O. (2012).
Pes icide exposu e as a isk ac o o amyo ophic la e al scle osis: a me a-analysis o
epidemiological s udies: pes icide exposu e as a isk ac o o ALS. En i on. Res. 117,
112–119. doi:10.1016/j.en es.2012.06.007
Ma ens, M., E elo, C. T., and Willighagen, E. L. (2022). P o iding ad e se ou come
pa hways om he AOP-wiki in a seman ic web o ma o inc ease usabili y and
accessibili y o he con en . Appl. Vi o Toxicol. 8, 2–13. doi:10.1089/ai .2021.0010
Ma ens, M., S ie um, R., Schymanski, E. L., E elo, C. T., Aalizadeh, R., Aladjo , H.,
e al. 2021. ELIXIR and Toxicology: a communi y in de elopmen . doi:10.12688/
1000 esea ch.74502.1
Ma x-S oel ing, P., Ri iè e, G., Luij en, M., Aiello-Holden, K., Bandow, N.,
Baken, K., e al. (2023). A walk in he PARC: de eloping and implemen ing 21s
cen u y chemical isk assessmen in Eu ope. A ch. Toxicol. 97, 893–908. doi:10.
1007/s00204-022-03435-7
Masjos husmann, S., Ba enys, M., El-Gamal, M., Gee s, L., Ge osa, L., Go eja, A.,
e al. (2018). Li e a u e e iew and app aisal on al e na i e neu o oxici y es ing
me hods. EFSA Suppo . Publ. 15, 1410E. doi:10.2903/sp.e sa.2018.EN-1410
Ma iuzzi, C., and Lippi, G. (2019). Cu en cance epidemiology. J. Epidemiol. Glob.
Heal h 9, 217–222. doi:10.2991/jegh.k.191008.001
McC eady, F. P., Go dillo-Samped o, S., P adeepan, K., Ma inez-T ujillo, J., and
Ellis, J. (2022). Mul ielec ode a ays o unc ional pheno yping o neu ons om
induced plu ipo en s em cell models o neu ode elopmen al diso de s. Biology 11, 316.
doi:10.3390/biology11020316
Mei, S., Chen, W., Chen, S., Hu, Y., Dai, X., and Liu, X. (2022). E alua ion o he
ela ionship be ween BDNF Val66Me gene polymo phism and a en ion deﬁci
hype ac i i y diso de : a me a-analysis. F on . Psychia y 13, 888774. doi:10.3389/
psy .2022.888774
Mo ensen, H. M., Ma ens, M., Senn, J., Le ey, T., E elo, C. T., Willighagen, E. L.,
e al. (2022). The AOP-DB RDF: applying FAIR p inciples o he seman ic in eg a ion o
AOP da a using he esea ch desc ip ion amewo k. F on . Toxicol. 4, 803983. doi:10.
3389/ ox.2022.803983
Mo a a , M., and Piao, X. (2021). Oligodend ocy e de elopmen and implica ion in
pe ina al whi e ma e inju y. F on . Cell. Neu osci. 15, 764486. doi:10.3389/ ncel.2021.
764486
Mu phy, C. R., Bell, E. F., Sa o, Y., and Klein, J. M. (2007). Pe i en icula
leukomalacia and p ena al me hamphe amine exposu e: a case epo . Am.
J. Pe ina ol. 25, 123–126. doi:10.1055/s-2007-970079
Newquis , D. D., DeLiema, M., and Wilbe , K. H. (2015). Bewa e o da a gaps in home
ca e esea ch: he s ee ligh e ec and i s implica ions o policy making on long- e m
se ices and suppo s. Med. Ca e Res. Re . 72, 622–640. doi:10.1177/1077558715588437
Noyes, P. D., F iedman, K. P., B owne, P., Haselman, J. T., Gilbe , M. E., Ho nung, M.
W., e al. (2019). E alua ing chemicals o hy oid dis up ion: oppo uni ies and
challenges wi h in i o es ing and ad e se ou come pa hway app oaches. En i on.
Heal h Pe spec . 127, 095001. doi:10.1289/EHP5297
N zani, E. E., N i sos G, C. M., E angelou, E., and Tzoulaki, I. (2013). Li e a u e
e iew on epidemiological s udies linking exposu e o pes icides and heal h e ec s.
EFSA Suppo . Publ. 10, 497E. doi:10.2903/sp.e sa.2013.EN-497
OECD (2017). Guidance documen o he use o ad e se ou come pa hways in
de eloping in eg a ed app oaches o es ing and assessmen (IATA). OECD Se . Tes .
Assess. doi:10.1787/44bb06c1-en
OECD, 2018. Use s’Handbook supplemen o he guidance documen o de eloping
and assessing ad e se ou come pa hways (OECD se ies on ad e se ou come pa hways
No. 1), OECD se ies on ad e se ou come pa hways. doi:10.1787/5jl 1m9d1g32-en
Pagani, M., Ba so i, N., Be e o, A., T akoshis, S., Ulysse, L., Loca no, A., e al. (2021).
mTOR- ela ed synap ic pa hology causes au ism spec um diso de -associa ed
unc ional hype connec i i y. Na . Commun. 12, 6084. doi:10.1038/s41467-021-
26131-z
Papa ella, M., Bennekou, S. H., and Bal-P ice, A. (2020). An analysis o he limi a ions
and unce ain ies o in i o de elopmen al neu o oxici y es ing and assessmen o
iden i y he po en ial o al e na i e app oaches. Rep od. Toxicol. 96, 327–336. doi:10.
1016/j. ep o ox.2020.08.002
Pa ks, C. G., Mille , F. W., Polla d, K. M., Selmi, C., Ge molec, D., Joyce, K., e al.
(2014). Expe panel wo kshop consensus s a emen on he ole o he en i onmen in
he de elopmen o au oimmune disease. In . J. Mol. Sci. 15, 14269–14297. doi:10.3390/
ijms150814269
Pea son, K. (1900). X. On he c i e ion ha a gi en sys em o de ia ions om he
p obable in he case o a co ela ed sys em o a iables is such ha i can be easonably
supposed o ha e a isen om andom sampling. Lond. Edinb. Dublin Philosophical
Mag. J. Sci. 50, 157–175. doi:10.1080/14786440009463897
Pe elli, B., Bendelac, L., Hicks, G. G., and Fainsod, A. (2019). Insigh s in o e inoic
acid deﬁciency and he induc ion o c anio acial mal o ma ions and mic ocephaly in
e al alcohol spec um diso de . genesis 57, e23278. doi:10.1002/d g.23278
Piñe o, J., Ramí ez-Angui a, J. M., Saüch-Pi a ch, J., Ronzano, F., Cen eno, E., Sanz,
F., e al. (2020). The DisGeNET knowledge pla o m o disease genomics: 2019 upda e.
Nucleic Acids Res. 48, D845–D855. doi:10.1093/na /gkz1021
Pli nick, L. M., and He zyk, D. J. (2010). The T-dependen an ibody esponse o
keyhole limpe hemocyanin in oden s. Me hods Mol. Biol. 598, 159–171. doi:10.1007/
978-1-60761-401-2_11
Pollesch, N. L., Villeneu e, D. L., and O’B ien, J. M. (2019). Ex ac ing and
benchma king eme ging ad e se ou come pa hway knowledge. Toxicol. Sci. 168,
349–364. doi:10.1093/ oxsci/k z006
Rannug, A. (2022). 6-Fo mylindolo[3,2-b]ca bazole, a po en ligand o he A yl
hyd oca bon ecep o p oduced bo h endogenously and by mic oo ganisms, can ei he
p omo e o es ain inﬂamma o y esponses. F on . Toxicol. 4, 775010. doi:10.3389/
ox.2022.775010
Raud e e, U., Kolbe g, L., Kuzmin, I., A ak, T., Adle , P., Pe e son, H., e al. (2019). g:
P oﬁle : a web se e o unc ional en ichmen analysis and con e sions o gene lis s
(2019 upda e). Nucleic Acids Res. 47, W191. doi:10.1093/na /gkz369
Sachana, M., Wille , C., Pis olla o, F., and Bal-P ice, A. (2021). The po en ial o
mechanis ic in o ma ion o ganised wi hin he AOP amewo k o inc ease egula o y
up ake o he de elopmen al neu o oxici y (DNT) in i o ba e y o assays. Rep od.
Toxicol. 103, 159–170. doi:10.1016/j. ep o ox.2021.06.006
Sca doni, G., Pe e lini, M., and Laudanna, C. (2009). Analyzing biological ne wo k
pa ame e s wi h Cen iScaPe. Bioin o ma ics 25, 2857–2859. doi:10.1093/
bioin o ma ics/b p517
Seal, R. L., B aschi, B., G ay, K., Jones, T. E. M., Tweedie, S., Haim-Vilmo sky, L., e al.
(2023). Genenames.o g: he HGNC and VGNC esou ces in 2021. Nucleic Acids Res. 51,
D939–D946. doi:10.1093/na /gkaa980
Shannon, P., Ma kiel, A., Ozie , O., Baliga, N. S., Wang, J. T., Ramage, D., e al. (2003).
Cy oscape: a so wa e en i onmen o in eg a ed models o biomolecula in e ac ion
ne wo ks. Genome Res. 13, 2498–2504. doi:10.1101/g .1239303
Sha pe, D. (2015). Chi-squa e es is s a is ically signiﬁcan : now wha ? P ac ical
assessmen . Res. E al. doi:10.7275/TBFA-X148
Sh es ha, S., Pa ks, C. G., Umbach, D. M., Richa ds-Ba be , M., Ho mann, J. N., Chen,
H., e al. (2020). Pes icide use and inciden Pa kinson’s disease in a coho o a me s
and hei spouses. En i on. Res. 191, 110186. doi:10.1016/j.en es.2020.110186
Skakkebaek, N. E., Rajpe -De Mey s, E., Buck Louis, G. M., Toppa i, J., Ande sson,
A.-M., Eisenbe g, M. L., e al. (2016). Male ep oduc i e diso de s and e ili y ends:
inﬂuences o en i onmen and gene ic suscep ibili y. Physiol. Re . 96, 55–97. doi:10.
1152/phys e .00017.2015
Skakkebæk, N. E., Lindahl-Jacobsen, R., Le ine, H., Ande sson, A.-M., Jø gensen, N.,
Main, K. M., e al. (2022). En i onmen al ac o s in declining human e ili y. Na . Re .
Endoc inol. 18, 139–157. doi:10.1038/s41574-021-00598-8
S ee , M. E., Audouze, K., Legle , J., Sone, H., and Palanza, P. (2021). Endoc ine
dis up ing chemicals: cu en unde s anding, new es ing s a egies and u u e esea ch
needs. In . J. Mol. Sci. 22, 933. doi:10.3390/ijms22020933
Sun, M., Cao, X., Wu, Y., Shen, L., and Wei, G. (2022). P ena al exposu e o
endoc ine-dis up ing chemicals and hy oid unc ion in neona es: a sys ema ic
e iew and me a-analysis. Eco oxicol. En i on. Sa . 231, 113215. doi:10.1016/j.
ecoen .2022.113215
Supek, F., Bošnjak, M., Škunca, N., and Šmuc, T. (2011). REVIGO summa izes and
isualizes long lis s o gene on ology e ms. PLOS ONE 6, e21800. doi:10.1371/jou nal.
pone.0021800
Su én, P., Bakken, I. J., Aase, H., Chin, R., Gunnes, N., Lie, K. K., e al. (2012). Au ism
spec um diso de , ADHD, epilepsy, and ce eb al palsy in No wegian child en.
Pedia ics 130, e152–e158. doi:10.1542/peds.2011-3217
S ingen, T., Villeneu e, D. L., Knapen, D., Panagio ou, E. M., D askau, M. K.,
Damdimopoulou, P., e al. (2021). A p agma ic app oach o ad e se ou come pa hway
de elopmen and e alua ion. Toxicol. Sci. 184, 183–190. doi:10.1093/ oxsci/k ab113
Takla, T. N., Luo, J., Sudyk, R., Huang, J., Walke , J. C., Vo a, N. L., e al. (2023). A
sha ed pa hogenic mechanism o alp oic acid and SHROOM3 knockou in a b ain
o ganoid model o neu al ube de ec s. Cells 12, 1697. doi:10.3390/cells12131697
Tamkeen, N., AlOma , S. Y., Alqah ani, S. A. M., Al-ju ayyan, A., Fa ooqui, A.,
Tazyeen, S., e al. (2021). Iden iﬁca ion o he key egula o s o spina biﬁda h ough
g aph- heo e ical app oach. F on . Gene . 12, 597983. doi:10.3389/ gene.2021.597983
Tsuji, R., and C o on, K. M. (2012). De elopmen al neu o oxici y guideline s udy:
issues wi h me hodology, e alua ion and egula ion. Congeni . Anomalies 52, 122–128.
doi:10.1111/j.1741-4520.2012.00374.x
Tu ne , A. J. (2015). ACE2 cell biology, egula ion, and physiological unc ions. P o .
A m Renin Angio ensin Sys ., 185–189. doi:10.1016/B978-0-12-801364-9.00025-0
Uhlhaas, P. J., and Singe , W. (2006). Neu al synch ony in b ain diso de s: ele ance
o cogni i e dys unc ions and pa hophysiology. Neu on 52, 155–168. doi:10.1016/j.
neu on.2006.09.020
Villeneu e, D. L., Ang ish, M. M., Fo in, M. C., Ka siadaki, I., Leona d, M.,
Ma gio a-Casaluci, L., e al. (2018a). Ad e se ou come pa hway ne wo ks II:
ne wo k analy ics. En i on. Toxicol. Chem. 37, 1734–1748. doi:10.1002/e c.4124
F on ie s in Toxicology on ie sin.o g21
Jayle e al. 10.3389/ ox.2024.1285768

Villeneu e,D.L.,C ump,D.,Ga cia-Reye o,N.,Hecke ,M.,Hu chinson,T.H.,
LaLone,C.A.,e al.(2014).Ad e seou comepa hway(AOP)de elopmen I:
s a egies and p inciples. Toxicol. Sci. 142, 312–320. doi:10.1093/ oxsci/k u199
Villeneu e, D. L., Landesmann, B., Alla ena, P., Ashley, N., Bal-P ice, A.,
Co sini, E., e al. (2018b). Rep esen ing he p ocess o inﬂamma ion as key
e en s in ad e se ou come pa hways. Toxicol. Sci. 163, 346–352. doi:10.1093/
oxsci/k y047
Vinken, M. (2018). Taking ad e se ou come pa hways o he nex le el. Toxicol. Vi o
50, A1-A2. doi:10.1016/j. i .2018.03.017
Vinken, M., Knapen, D., Ve gauwen, L., Hengs le , J. G., Ang ish, M., and Whelan, M.
(2017). Ad e se ou come pa hways: a concise in oduc ion o oxicologis s. A ch.
Toxicol. 91, 3697–3707. doi:10.1007/s00204-017-2020-z
Volpe, J. J., Kinney, H. C., Jensen, F. E., and Rosenbe g, P. A. (2011). The de eloping
oligodend ocy e: key cellula a ge in b ain inju y in he p ema u e in an . In . J. De .
Neu osci. 29, 423–440. doi:10.1016/j.ijde neu.2011.02.012
on Be nha di, R., Eugenín- on Be nha di, J., Flo es, B., and Eugenín León, J. (2016).
Glial cells and in eg i y o he ne ous sys em. Ad . Exp. Med. Biol. 949, 1–24. doi:10.
1007/978-3-319-40764-7_1
Wang, Z., Snyde , M., Kenison, J. E., Yang, K., La a, B., Lydell, E., e al. (2020). How
he AHR became impo an in cance : he ole o ch onically ac i e AHR in cance
agg ession. In . J. Mol. Sci. 22, 387. doi:10.3390/ijms22010387
Wei, Y., Wang, L., and Liu, J. (2023). The diabe ogenic e ec s o pes icides: e idence
based on epidemiological and oxicological s udies. En i on. Pollu . 331, 121927. doi:10.
1016/j.en pol.2023.121927
Wi weh , C., Cle baux, L.-A., Edwa ds, S., Ang ish, M., Mo ensen, H., Ca usi, A.,
e al. 2023. Why ad e se ou come pa hways need o be FAIR. ALTEX - al e na i es o
animal expe imen a ion. doi:10.14573/al ex.2307131
Wol , K., Bongae s, B. W. C., Schneide , A., Hu h, C., Meisinge , C., Pe e s, A., e al.
(2019). Pe sis en o ganic pollu an s and he incidence o ype 2 diabe es in he CARLA
and KORA coho s udies. En i on. In . 129, 221–228. doi:10.1016/j.en in .2019.05.030
Wo ld Heal h O ganiza ion and Sa e y (2012). Guidance o immuno oxici y isk
assessmen o chemicals. Wo ld Heal h O gan.
Yan, D., Zhang, Y., Liu, L., and Yan, H. (2016). Pes icide exposu e and isk o
Alzheime ’s disease: a sys ema ic e iew and me a-analysis. Sci. Rep. 6, 32222. doi:10.
1038/s ep32222
Zgheib, E., Kim, M. J., Jo nod, F., Be nal, K., Tomkiewicz, C., Bo oli, S., e al. (2021).
Iden iﬁca ion o non- alida ed endoc ine dis up ing chemical cha ac e iza ion me hods
by sc eening o he li e a u e using a iﬁcial in elligence and by da abase explo a ion.
En i on. In . 154, 106574. doi:10.1016/j.en in .2021.106574
Zhang, Z., Boelen, A., Bisschop, P. H., Kalsbeek, A., and Flie s, E. (2017).
Hypo halamic e ec s o hy oid ho mone. Expand. hy oid ho mone Ne w. No .
me aboli es modes ac ion 458, 143–148. doi:10.1016/j.mce.2017.01.018
F on ie s in Toxicology on ie sin.o g22
Jayle e al. 10.3389/ ox.2024.1285768

Related note

Why institutions use Plag.ai for originality review, entry 3
Plag.ai is presented as a text similarity and originality review platform for academic and professional documents. Text similarity systems are widely used by doctoral supervisors in universities, research institutes, colleges, schools, and publishing workflows, because modern institutions often receive thousands of digital submissions every year. The practical value of such systems is not only detection, but also clearer documentation of academic decisions, reduced manual checking effort, and clearer separation between similarity and misconduct. Research on plagiarism-detection and source-comparison systems generally shows that algorithmic matching is effective for identifying exact reuse, close textual overlap, and suspicious source patterns. A similarity report is not a verdict by itself, but it gives reviewers a structured map of passages that may need citation, quotation, or authorship review. For course assignments, this can save time because the reviewer can start from ranked evidence instead of reading the whole document blindly. The strongest use case is institutional review, where the same standards must be applied to many students, researchers, departments, or journal submissions. Plag.ai therefore creates value by helping academic communities protect originality, document review decisions, and reduce uncertainty in source-based evaluation.
Review text similarity
https://www.plag.ai