Bioinformatics, 38(4), 2022,1173-1175
https://doi.org/10.1093/bioinformatics/btab750
Advance Access Publication Date: 30 October 2021
Applications Note
Data and text mining
AOP-helpFinder webserver: a tool for comprehensive
analysis of the literature to support adverse outcome
pathways development
Florence Jornod1
, Thomas Jaylet1
, Luděk Blaha2
, Denis Sarigiannis3
, Luc Tamisier4
and Karine Audouze )1
'*
'Universitě de Paris, T3S, Inserm UMR-S1124, Paris F-75006, France, 2
RECET0X, Faculty of Science, Masaryk University, Brno CZ62500,
Czech Republic, 3
HERACLES Research Center on the Exposome and Health, Aristotle University of Thessaloniki, Center for
Interdiciplinary Research and Innovation, Thessaloniki 57001, Greece and "Universitě de Paris, SPPIN CNRS UMR 8003,Paris F-75006,
France
*To whom correspondence should be addressed.
Associate Editor: Jonathan Wren
Received on July 21, 2021; revised on September 30, 2021; editorial decision on October 21, 2021; accepted on October 27,2021
A b s t r a c t
Motivation: A d v e r s e o u t c o m e p a t h w a y s (AOPs) are a c o n c e p t u a l f r a m e w o r k d e v e l o p e d to s u p p o r t the use of alternative
t o x i c o l o g y a p p r o a c h e s in the risk a s s e s s m e n t . A O P s are structured linear organizations of existing k n o w l e d g e
illustrating causal p a t h w a y s f r o m the initial m o l e c u l a r perturbation triggered by v a r i o u s stressors, t h r o u g h key
events (KEs) at different levels of biology, to the ultimate health or e c o t o x i c o l o g i c a l a d v e r s e o u t c o m e .
Results: Artificial intelligence c a n be u s e d to systematically explore available t o x i c o l o g i c a l data that can be parsed
in the scientific literature. Recently, a tool called A O P - h e l p F i n d e r w a s d e v e l o p e d to identify a s s o c i a t i o n s b e t w e e n
stressors a n d K E s s u p p o r t i n g t h u s d o c u m e n t a t i o n of A O P s . T o facilitate the utilization of this a d v a n c e d bioinformatics
tool by the scientific a n d the regulatory c o m m u n i t y , a w e b s e r v e r w a s created. T h e p r o p o s e d A O P - h e l p F i n d e r
w e b s e r v e r uses better p e r f o r m i n g v e r s i o n of the tool w h i c h reduces the n e e d for m a n u a l curation of the o b t a i n e d
results. A s an e x a m p l e , the server w a s s u c c e s s f u l l y a p p l i e d to explore relationships of a set of e n d o c r i n e disruptors
with metabolic-related events. T h e A O P - h e l p F i n d e r w e b s e r v e r assists in a rapid evaluation of existing k n o w l e d g e
stored in the P u b M e d database, a global resource of scientific information, to build A O P s a n d A d v e r s e O u t c o m e
N e t w o r k s s u p p o r t i n g the c h e m i c a l risk a s s e s s m e n t .
Availability and implementation: A O P - h e l p F i n d e r is available at http://aop-helpfinder.u-paris-sciences.fr/index.php
Contact: karine.audouze@u-paris.fr
Supplementary information: S u p p l e m e n t a r y data are available at Bioinformatics online.
1 Introduction
Structured organization of toxicological and ecotoxicological data is
now feasible using the adverse outcome pathways (AOP) framework
(Ankley et al., 2010). A n A O P is defined by a linear combination of
biological events, started from a molecular initiating event (MIE)
triggered by stressors (pollutants, ionizing radiations, nanomaterials
or climate stressors) connected through a series of key events (KEs)
occurring at various levels of the biological organization, to an adverse
outcome (AO). Biological events (MIE, K E and AO) are not
linked to a unique AOP, but can be shared, allowing the establishment
of Adverse Outcome Network (AON) that reflect better the
true complexity of the biology. Combined with new approach methodologies
(Parish et al., 2020), AOPs and A O N s are extremely
useful in establishing integrated approaches to testing and assessment
(IATA) for environmental and risk assessment, and they aid to
the development of novel nonanimal toxicity testing strategies
(Delrue eta/., 2016).
With advances in technologies, huge amounts of data have become
available, compiled in well-structured toxicological databases
(e.g. C T D , CompTox), in AOP-oriented webservers (AOP-wiki,
sAOP, AOP4EUpest) and scientific publications (Williams et al.,
2017). Innovative data mining tools are needed to identify sparse
but complementary data such as Abstract Sifter allowing to have a
view of the toxicological information landscape for a set of entities
as chemicals (Baker et al., 2017) or ComptoxAI (https://comptox.ai/
index.html). Artificial intelligence (Al) technology, that uses natural
language processing (NLP), is an interesting way to facilitate the
©Tbe Autbor(s) 2021. Published by Oxford University Press. 1173
This is an Open A c c e s s article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https://creativecommons.Org/licenses/by-nc/4.0/),
which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact
journals.permissions@oup.com
1174 F.Jornod et al.
identification of links between relevant information that can be used
to build novel AOPs (Song et al., 2020), and identify knowledge
gaps and research needs (Zgheib et al., 2021). Several tools use text
mining (TM), an A l method to transform unstructured into structured
text. For example, Limtox provides a biomedical search for
adverse hepatobiliary reactions (Canada et al., 2017). Recently, the
AOP-helpFinder tool, based on T M and graph theory was proposed
to identify stressor-KE relationships by examining large collections
of scientific abstracts, and was applied to bisphenol A substituents
and pesticides (Carvaillo et al., 2019; Jornod et al., 2020; Rugard
etal., 2020).
Here, we present the AOP-helpFinder webserver, which uses an
updated version of the tool, to provide an easy but effective resource
for identifying and compiling existing knowledge from the scientific
literature. The main optimized features are (i) the capability to
choose to search in full abstracts or without considering the introductory
parts, (ii) the possibility to perform a refined search using
machine learning and (iii) an automatic update of the PubMed database
before each search. A case study with endocrine disruptors
(ED) and metabolism-related events is provided, illustrating the capacity
of the tool to collate quickly an overview of the existing
information.
2 M a t e r i a l s a n d m e t h o d s
2.1 T h e A O P - h e l p F i n d e r w e b s e r v e r
The proposed webserver is easy to use, and requires only the user
email to access the upload page and to receive information when the
results are available for download. This simple procedure is in line
with digital sobriety that aims to reduce the environmental impact
by limiting computing use. Two input files are needed: one with the
stressors of interest and the second with biological events (i.e. MIE,
KE and/or A O ) . Before running the tool to identify if knowledge
connecting stressors and biological events exists, the user can choose
between two options: reduce search and refinement filter (see the
following section and Supplementary Material), as well as the output
format (date, title, PMID, etc.).
2.2 T h e A O P - h e l p F i n d e r tool
To increase the performance of the previously developed version,
several methods were tested using a set on ED and biological events
related to metabolism, and two were kept through the process (see
Supplementary Material, https://github.com/jornod/aophelpFinder):(i)
'reduced search': searches are performed in the full
abstracts or without considering the introductory part, which
appears to be covered usually by the first 20% of the abstracts. This
option allows avoiding too many false positives, as the introduction
often reflects a working hypothesis instead of the conclusions of the
publication and (ii) 'refinement filter': after the preprocessing that
uses a stemming process (Carvaillo et al., 2019), the tool can refine
the searches by combining a deletion of sentences containing context
words with a lemmatization process. Lemmatization is a machine
learning method for text normalization used in N L P that considers
the context and converts the word to its meaningful base form. This
option is very useful when terms have common stems (e.g. tests, testis
• test) leading to incorrect meanings and spelling errors. Further,
an automatic daily update of the PubMed database was newly
implemented using the NCBI API to screen the full existing
knowledge.
The current version of the A l tool mined the PubMed database,
that is a global source for scientific literature. Nevertheless, the
developed method screens text-based knowledge, and therefore the
AOP-helpFinder server could be improved for mining multiple sources
(databases, literature), including studies reporting negative findings,
to accelerate information gathering when data are limited and
present in diverse sources (Carvaillo etal., 2019).
The advantage of the proposed method, is its capacity to be
adapted for literature searches in general, independent of A O P development,
in order to identify interconnections between the query
Fig. 1. Example of E D comentioned in P u b M e d scientific abstracts w i t h biological
events related to metabolism, identified by the A O P - h e l p F i n d e r webserver. The
numbers correspond to the % of retrieved abstracts mentioning both the stressor
(column) and the event (line) among all identified abstracts (the colors are according
to the percentage for better visualization). For example, among all identified
abstracts that comentioned bisphenol S (BPS) and at least one event from the list,
1 3 % of the abstracts were comentioning BPS (fourth column) and obesity (the second
line from the bottom)
keywords, as it was successfully done to decipher nonvalidated test
methods for ED (Zgheib et al., 2021).
2.3 C a s e study on e n d o c r i n e disruptors a n d m e t a b o l i s m
The AOP-helpFinder webserver was used for a case study aiming at
automatically identifying existing relationships and knowledge gaps
between 10 ED (Supplementary Table SI) and 294 biological events
related to metabolism (Supplementary Table S2). The webserver
was launched using 'reduced search' (omitting searches in the first
20% of the abstracts) and 'refinement filter'. Among the 83 970
abstracts retrieved in the PubMed database as of May 10, 2021
related to at least one ED (Supplementary Table SI), a total of 4622
were retained (comentioning ED and event). Among the 294 events,
108 were identified as comentioned with at least one ED
(Supplementary Table S2). Figure 1 illustrates the large disparity of
knowledge for the 10 selected ED in the area of metabolism (see
Supplementary Fig. SI for all results). For example, cadmium,
bisphenol A and di(2-ethylhexyl) phthalate (DEHP) are well studied
chemicals as the webserver retrieved scientific articles for almost all
biological events of interest. Other chemicals (bisphenol F, bisphenol
S, butyl-paraben) appear to be less studied (Supplementary
Table SI), and the information were essentially identified for extensively
studied biological events such as oxidative stress or obesity.
3 C o n c l u s i o n
The AOP-helpFinder webserver uses an automatic A l screening to
rapidly retrieve existing knowledge on links between stressors and
biological events to build AOPs and A O N s . This webserver allows
highly effective searches in PubMed as it considerably reduces the
time of finding relevant scientific articles. The comprehensive AIbased
analyses of existing literature support various needs of the risk
assessment such as establishment of causality between chemicals
and AOs through AOPs and A O N s , identification of gaps or prioritization
and design of future experimental and epidemiological
studies.
A c k n o w l e d g e m e n t s
The authors would like to acknowledge Inserm and the Universitě de Paris for
supporting the work.
AOP-helpFinder webserver 1175
F u n d i n g
This work was supported by the European Union's Horizon 2020 Research
and Innovation Programme O B E R O N [https://oberon-4eu.com, Grant
8257121 and H B M 4 E U [https://www.hbm4eu.eu/, Grant 733032].
Conflict of Interest: none declared.
References
Ankley,G.T. et al. (2010) Adverse outcome pathways: a conceptual framework
to support ecotoxicology research and risk assessment. Environ.
Toxicol. Chem.,29, 730-741.
Baker,N. et al. (2017) Abstract Sifter: a comprehensive front-end system to
PubMed. FlOOOResearch, 6,2164.
Canada,A. et al. (2017) LimTox: a web tool for applied text mining of adverse
event and toxicity associations of compounds, drugs and genes. Nucleic
Acids Res., 45, W484-W489.
Carvaillo,J.-C. et al. (2019) Linking bisphenol S to adverse outcome pathways
using a combined text mining and systems biology approach. Environ.
Health Perspect., 127,47005.
Delrue,N. et al. (2016) The adverse outcome pathway concept: a basis for
developing regulatory decision-making tools. Altern. Lab. Anim., 44,
417-429.
Jornod,F. et al. (2020) AOP4EUpest: mapping of pesticides in Adverse
Outcome Pathways using a text mining tool. Bioinformatics, 36,
4379-4381.
Parish,S.T. et al. (2020) A n evaluation framework for new approach methodologies
(NAMs) for human health safety assessment. Regul. Toxicol.
Pharmacol., 112,104592.
Rugard,M. et al. (2020) Deciphering adverse outcome pathway network
linked to bisphenol F using text mining and systems toxicology approaches.
Toxicol. Sci., 173, 32-40.
SongJ. et al. (2020) Upregulation of angiotensin converting enzyme 2 by shear
stress reduced inflammation and proliferation in vascular endothelial cells.
Biochem. Biophys. Res. Commun., 525, 812-818.
Williams,A.J. et al. (2017) The CompTox chemistry dashboard: a community
data resource for environmental chemistry./. Cheminform., 9, 61.
Zgheib,E. et al. (2021) Identification of non-validated endocrine disrupting
chemical characterization methods by screening of the literature
using artificial intelligence and by database exploration. Environ. Int.,
154,106574.