Content uploaded by Piyush Priya
Author content
All content in this area was uploaded by Piyush Priya on Dec 27, 2023
Content may be subject to copyright.
RESOURCE
Stress combinations and their interactions in plants
database: a one-stop resource on combined stress responses
in plants
Piyush Priya
†
, Mahesh Patil
†
, Prachi Pandey, Anupriya Singh, Vishnu Sudha Babu and Muthappa Senthil-Kumar
*
National Institute of Plant Genome Research, Aruna Asaf Ali Marg, P.O. Box No. 10531, New Delhi 110067, India
Received 24 April 2023; revised 23 August 2023; accepted 28 September 2023.
*For correspondence (e-
mail skmuthappa@nipgr.ac.in).
†
Equal contribution
SUMMARY
We have developed a compendium and interactive platform, named Stress Combinations and their Interac-
tions in Plants Database (SCIPDb; http://www.nipgr.ac.in/scipdb.php), which offers information on morpho-
physio-biochemical (phenome) and molecular (transcriptome and metabolome) responses of plants to differ-
ent stress combinations. SCIPDb is a plant stress informatics hub for data mining on phenome, transcrip-
tome, trait-gene ontology, and data-driven research for advancing mechanistic understanding of combined
stress biology. We analyzed global phenome data from 939 studies to delineate the effects of various stress
combinations on yield in major crops and found that yield was substantially affected under abiotic–abiotic
stresses. Transcriptome datasets from 36 studies hosted in SCIPDb identified novel genes, whose roles have
not been earlier established in combined stress. Integretome analysis under combined drought–heat stress
pinpointed carbohydrate, amino acid, and energy metabolism pathways as the crucial metabolic, proteomic,
and transcriptional components in plant tolerance to combined stress. These examples illustrate the appli-
cation of SCIPDb in identifying novel genes and pathways involved in combined stress tolerance. Further,
we showed the application of this database in identifying novel candidate genes and pathways for com-
bined drought and pathogen stress tolerance. To our knowledge, SCIPDb is the only publicly available plat-
form offering combined stress-specific omics big data visualization tools, such as an interactive scrollbar,
stress matrix, radial tree, global distribution map, meta-phenome analysis, search, BLAST, transcript expres-
sion pattern table, Manhattan plot, and co-expression network. These tools facilitate a better understanding
of the mechanisms underlying plant responses to combined stresses.
Keywords: biocuration, climate resilience, data visualization, plant stress informatics hub, integretome,
pathway enrichment, phenome, transcriptome.
INTRODUCTION
Abiotic and biotic stresses are the major deterrents to the
achievement of global food security, necessitating
the urgency to develop better-adapted crops (IPCC, 2022;
Mittler & Blumwald, 2010). Plants are often exposed to
combinations of stresses during their life cycle, and
increasing evidence highlights that combined stress is a
more potent and realistic threat to plant growth and pro-
ductivity than individual stresses (Ahuja et al., 2010; Atkin-
son & Urwin, 2012; Desaint et al., 2021; Sinha et al., 2021;
Zandalinas, Fritschi, et al., 2021; Zandalinas, Sengupta,
et al., 2021). Considerable information on plant stress has
accumulated over the years, but current understanding of
the physiological and molecular responses of plants to
combined stress is inadequate (Mahalingam et al., 2021;
Pandey et al., 2017; Zandalinas, Fritschi, et al., 2020; Zanda-
linas & Mittler, 2022). Combined stress studies, although
under-represented compared to individual stress studies,
entail voluminous and highly complex information on
molecular mechanisms of plant defense (Cohen et al.,
2021; Zandalinas, Fichman, et al., 2020; Zandalinas,
Fritschi, et al., 2021).
A plant perceives combined stress as a new state of
stress, the physiological and molecular response of plant
to combined stress is often entirely different from the
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd. 1
The Plant Journal (2023) doi: 10.1111/tpj.16497
response under the individual stresses. Therefore, the
adaptation strategies to stress combinations are based on
the interaction between the physiological and molecular
responses simultaneously triggered by each stress entity
independently (Gupta et al., 2016; Lopez-Delacalle et al.,
2021; Pandey et al., 2017; Rizhsky et al., 2004). The out-
come of such interactions may be “positive” or “negative,”
wherein combined stress causes less or more
damage, respectively, than that of the individual stresses
(Figure S1). The outcome also depends on factors like
plant age, genotype, stress intensity, duration of the stres-
ses, order of stress perceived by the plant, and time of ini-
tiation of subsequent stress since the first stress, all of
which make combined stress more complex to understand
(Choudhary & Senthil-Kumar, 2022; Mittler, 2006; Pandey
et al., 2015; Sinha et al., 2022; Zandalinas, Sengupta, et al.,
2021). Further, reproductive tissues show more sensitivity
towards some stress combination like heat and drought
which leads to greater yield reduction (Sinha et al., 2022).
These responses are mediated by switching on specific
pathways and processes that are unique, specific, and
sometimes even contrasting from the individual stress
responses. Plants also exhibit shared responses common
to individual and combined stresses (Suzuki et al., 2014;
Zhang & Sonnewald, 2017).
Thus, to better comprehend the complexities of com-
bined stress responses and fill existing gaps, there is a
pressing need for a pertinent database. There is no
database or platform solely dedicated to combined stress.
TOMRES (https://www.tomres.eu/) and Stress Combina-
tion: A New Field in Molecular Stress Research by the Uni-
versity of North Texas (http://biology.unt.edu/
stresscombination/) are two web resources on combined
stress in specific plants, besides individual stress data-
bases such as STIFDB2, QlicRice, and the Arabidopsis
stress-responsive gene database (Borkotoky et al., 2013;
Naika et al., 2013; Smita et al., 2011). Although these
resources are immensely useful in their own way, they do
not contain updated information on stress combinations.
The Stress Combinations and their Interactions in Plants
database (SCIPDb; http://www.nipgr.ac.in/scipdb.php)
developed by us is a user-friendly platform providing
options to browse, search, analyze, and download data for
various stress combinations studied to date. SCIPDb pro-
vides researchers easy access to combined stress-related
information and tools for extracting need-based, specific
information.
RESULTS AND DISCUSSION
SCIPDb and its key features
SCIPDb is a comprehensive collection of morphological,
physiological, biochemical, and transcriptomic data on
combined stresses published to date, systematically ana-
lyzed and presented in an easy-to-use interactive database
and web server (Figure 1a,b; Figure S2). Currently, SCIPDb
Figure 1. Outline of the Stress Combinations and their Interactions in Plants Database (SCIPDb), indicating its key features and applications.
(a) The upper panel shows the steps involved in data mining, curation, analysis, and integration of phenome and transcriptome data into SCIPDb.
(b) The lower panel shows the key features and applications offered to the users in the phenome and transcriptome sections. Orange boxes indicate the two
major data sets hosted in SCIPDb.
(c) The interactive stress matrix shows the net impact of the interaction between the stresses. The net impact of combined stress was determined by analyzing
the percent reduction in plant growth, yield, and physiological traits. Three possible interactions, namely, positive (less damage under combined stress), nega-
tive (greater damage under combined stress), and others (equal damage under combined and individual stress), are depicted in green, red, and blue boxes,
respectively. A stress combination is classified as positive, negative, or others based on the maximum number of studies in a particular interaction. The size of
the box indicates the number of studies showing a particular interaction, i.e., a bigger size represents a greater number of studies.
(d) The radial tree depicts the effect of individual and combined stresses on various traits in different plant species arranged in hierarchical order (starting from
most to least damage). The parameters considered for developing radial trees were growth, yield, physiological, and pathogen-
associated parameters. To nor-
malize the data, percent change over control or percent change over individual stresses (in the case of pathogen-associated traits) is calculated for each trait
and presented.
(e) The interactive global map provides information on the global distribution of combined stresses and their effect on morpho-
physiological, biochemical, and
pathogen-associated traits. The map was generated using the geographic coordinates of the locations where the studies were conducted.
(f) An interactive heatmap enables users to visualize the gene expression profile of the top 20 differentially expressed genes (DEGs) for a particular
transcriptome.
(g) A co-
functional network depicts the correlation of the top 20 DEGs in the form of an interactive network. Nodes represent correlates (pink) of top 20 DEGs
(green). Edges represent a log-likelihood score. The co-functional network allows the user to interact with the graph, and it includes all the required gestures,
including pinch-to-zoom, box selection, panning, etc., to access other metadata for each node and edge.
(h) An interactive Manhattan plot depicts functional profiling of DEGs using various kinds of biological evidence, including Gene Ontology terms, biological
pathways, and regulatory DNA elements. The X-
axis represents functional terms grouped and color-coded by data sources, while the Y-axis shows the adjusted
enrichment P-values in negative log10 scales.
(i) BLAST server to find potential homologs and orthologs in SCIPDb.
(j) SCIPDb hosts science outreach materials like posters, slides, videos, and podcasts related to combined stress, which will be useful for students, researchers,
and scientists working in the area of combined stress.
(k) The unique keyword-
based search option helps to access all combined stress-related data with a single click. Searches can be performed using keywords like
plant name, pathogen, insect, name of combined stress, gene ID, and gene name.
(l) The download section provides processed phenome and transcriptome data and a reference list of combined stress articles hosted in SCIPDb. It also provides
a link to mutant transcriptome studies and diverse resources related to combined stress. Numbers within the cylinder indicate the total number of articles
curated and presented in the phenome and the total number of stress combinations covered under transcriptome. AA, abiotic–
abiotic stress; AB, abiotic–biotic
stress; BB, biotic–biotic stress; SC, stress combinations; PS, plant species. The figure was created with BioRender.com.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
2Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
hosts phenome data curated from 939 studies, covering
123 stress combinations, 118 plant species, 283 pathogenic
agents (including bacteria, fungi, oomycetes, nematodes,
viruses, mycoplasmas, viroids, and insects), and 7 weed
species (Figure 1). From the analysis of the phenome data,
107 agronomic traits affected by various stress combina-
tions were identified. Of these, 45 traits were mapped to
the identified Trait Ontology (TO) terms. Twenty traits
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 3
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
among the 45 were related to plant morphology and yield;
11 were related to plant physiology; and 15 were biochemi-
cal traits corresponding to changes in enzymes, metabo-
lites, and hormone levels. These traits can be targeted for
trait-based breeding programs to develop combined
stress-tolerant crops. Further, gene-to-TO relationships
(Pan et al., 2019) can be derived to decipher the genome-
to-phenome relationships under combined stress. The
transcriptome data hosted in SCIPDb is from 36 studies
available in the public domain thus far, representing 58
stress combinations and 16 plant species (Figure S3).
Phenomics
Systematically analyzed phenomes are presented as data
pages based on plant species (Figure S4). Data pages are
arranged and presented based on plant species under each
stress combination. The whole data pages can be divided
into two parts, first part explains the overall impact of com-
bined stress in comparison with individual stresses and
second part provides information on interaction between
the stresses at plant interface under combined stress
(Figure S4). For a holistic view of trends in the analyzed
phenome data, interactive visualizations such as combined
stress matrices, radial trees, and global combined stress
distribution maps have been provided (Figure 1). The inter-
active stress matrices in SCIPDb show the overall impact
of different stress combinations on various plant species
(Figure 1c;http://223.31.159.3/plant_complete/matrix.html).
These visualizations will aid in deciphering and distilling
comprehensive overviews of stress combinations across
plant species, unlike from individual stresses alone.
Among the 123 stress combinations, 69 combinations
showed a negative impact on plant growth and productiv-
ity (Figure 2; Figure S5). Twenty stress combinations
showed a positive impact on plants, many of them belong-
ing to the abiotic–biotic stress category. No combination
among the 41 abiotic-abiotic stress combinations studied
was observed to exhibit a net positive interaction,
highlighting abiotic stress combinations as significant
threats to plant yield (Figure S5,http://223.31.159.3/plant_
complete/matrix.html). This trend was also predicted in
earlier studies (Zandalinas et al., 2022; Zandalinas, Sen-
gupta, et al., 2021). In 12 stress combinations, an equal
number of studies reported both positive and negative
impacts of combined stress on plants, with the majority
belonging to the abiotic–biotic stress category (10 combi-
nations) (Figure S5). To decipher the impact of stress com-
binations on growth, yield, and physiological and
pathogen-associated traits in various plant species, data
were visualized in the form of a radial tree (Figure 1d). Our
analysis of the different abiotic and biotic stress combina-
tions reveals many pathogen infections that are aggra-
vated under several concurrent abiotic stresses. It also
reflects a number of pests–pathogen complexes that can
pose a challenge to agricultural productivity. A global com-
bined stress distribution map, another feature of SCIPDb,
shows the prevalence of particular stress combinations in
a locality with their impact on crop growth (Figure 1e).
Knowledge of the occurrence of important stress combina-
tions based on this interactive geographical distribution
map can assist researchers in identifying agronomically
relevant stress combinations. Our analysis of studies pub-
lished from 1952 to 2021 showed an increasing trend in
combined stress research, most of which were from the
Americas, Asia, and Europe, particularly from the last
decade (Figure S6a,b). Thus, the increase in the occurrence
of combined stresses in these areas is deepening crop
losses.
Transcriptomics
Transcriptome data were analyzed and presented as inter-
active bootstrap tables, which enlist differentially
expressed genes (DEGs) and their associated metadata in
the form of KEGG pathways and genes (Figure S7). Cross-
references to important resources are provided to enable
users to acquire more information directly. To further visu-
alize the high-dimensional transcriptome data, DEG tables
for common (shared genes between combined and individ-
ual stresses) category has been linked to interactive heat-
maps and Venn diagrams while for unique (genes uniquely
expressed under particular combined stress) category we
have added co-functional networks and Manhattan plots
(Figure 1f–h). SCIPDb hosts co-functional networks for the
top differentially expressed unique genes under multiple
combined stresses. It provides speculative evidence about
the genes that are co-regulated and thus might share a
similar biological function or act together to control a spe-
cific phenotype. The functional annotation of DEGs acts as
a key resource to elucidate the biological processes, func-
tions, and pathways controlling various combined stresses
in plants. Gene Ontology annotations provided in the form
of Manhattan plots can be used to visualize enriched bio-
logical processes, molecular functions, and cellular compo-
nents and pathways.
Metabolomics
Metabolome datasets integrated in SCIPDb is from 6 differ-
ent studies available in the public domain thus far, from 4
plant species. Out of this, drought and heat stress combi-
nation is the most studied stress combination represented
in 5 studies. The presented metabolomic data encom-
passes a comprehensive compilation of metabolites,
accompanied by their respective log fold change and abso-
lute fold change in comparison to the control group. These
values were derived through meticulous in-house analysis
of the amalgamated stress metabolome, readily accessible
within the public domain. Within this section, users are
afforded the opportunity to effortlessly navigate through
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
4Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
the metabolomics data. This is facilitated by a user-friendly
interface, which utilizes dropdown menus for the selection
of specific plants and their corresponding stress combina-
tions. Moreover, to cater to the diverse needs of users, the
provision for downloading the complete dataset resulting
from user selections is also made available, streamlining
access to the entirety of the data for further exploration
and analysis.
Additional features of SCIPDb
A large fraction of genes in non-model plant species
remains uncharacterized, which means that they lack func-
tional annotation. Prioritizing candidate genes without any
functional evidence in such species is challenging. The
standalone BLAST server integrated with SCIPDb provides
an option to query the database with batch nucleotide or
protein sequences and will help users identify genes
related to combined stress in the genomes of the ever-
increasing repertoire of newly sequenced crop species
(Figure 1i). SCIPDb hosts several videos, slides, podcasts,
and protocols related to combined stress, making it a
potential outreach portal to promote scientific communica-
tion and education (https://www.youtube.com/channel/
UCamQH8rGtNMBvjuyNiSVyCw) (Figure 1j). A unique
keyword-based search option allows a user to mine
desired information from both phenome and transcriptome
Figure 2. Stress matrix depicting the various stress combinations that affect plants.
Out of the 157 potential stress combinations identified through literature mining, SCIPDB presents curated phenomic datasets from 123 stress combinations.
The colors in the matrix, represented here, depict the impact of the stress combinations based on findings of available studies as analyzed in the database. Dark
maroon box-
Negative impact; majority of studies indicate a negative impact of stress combination, i.e. plants under these combined stresses are affected to a
greater extent as compared to individual stresses. Green box- Positive impact; majority of the studies indicate a positive impact of the stress combination imply-
ing plants under these combined stresses are less or equally affected as compared to one or both the individual stresses. Orange box- Positive or negative
impact; equal number of studies with positive or negative impact of stress combination. Purple box- Others/reported/uncurated; indicating the stress combina-
tions that are known to affect plants but are not curated in the database. Gray box- Not reported; depicting stress combinations that are not yet reported or are
less likely to exists. White box represents stress combinations of same stresses that are practically not feasible. The X and Y axes indicate different abiotic
(drought, cold highlight, flooding, heat, UV, ozone, heavy metal, nutrient [deficiency or toxicity] and salt) and biotic (bacteria, fungi, oomycetes, virus, mites,
nematodes, insects and weeds) stresses. The details of rule between the combination’s relationships are detailed in the materials and methods and also in the
database (under methodology).
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 5
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
datasets (Figure 1k). SCIPDb datasets are hosted on a local
FTP server, allowing users to download all curated phe-
nomes, genotypes, transcriptomes, and references locally
with just a few clicks. A user-defined download can also be
done through specific sections of the database. These
datasets can be further used for other downstream ana-
lyses to clearly grasp plant responses to combined stresses
(Figure 1l). SCIPDb also encourages users to submit their
data to the web portal to promote two-way communication
and ultimately contribute to making the database a
dynamic, robust, and single-stop platform for disseminat-
ing novel findings on combined stresses.
The interactive network developed by global com-
bined stress transcriptome profiling and pathway enrich-
ment analysis in Arabidopsis depicts the common and
unique pathways between major combined stress catego-
ries hosted under the transcriptome visualization section.
The “Applications” section hosts several case studies,
which can help users understand how to use the diverse
datasets hosted in SCIPDb to answer various biological
questions about combined stress. The “References and
links” section provides access to complete references of
the research articles used in developing the data page,
along with other related articles (review, thesis, and report
categories). A meta-phenome presents a combined trend
of the net impact of stress combinations on plant perfor-
mance after analyzing all the studies reported for a specific
crop for a particular stress combination. Overall, these
important features and tools in SCIPDb provide compre-
hensive information on each stress combination and can
help identify the most prominent stress combination in a
specific crop affecting polygenic traits like growth and
yield.
Key findings from the database
Impact of combined stress on yield and yield-attributing
traits in major crops
Among the 123 reported stress combinations, 58, 41, and
24 were from the abiotic–biotic, abiotic–abiotic, and biotic–
biotic stress categories, respectively (Figure 3a;
Figures S8–S10). Of the 58 stress combinations reported in
the abiotic–biotic stress category, 87 studies, covering 26
plant species, were of the nematode–fungus stress combi-
nation, indicating it as one of the most evident stress com-
binations (Figure S10, Table S1). Global analysis of yield
and yield-attributing traits belonging to plant performance,
plant physiological response, and plant pathogenesis
response showed greater reductions in yield under the
abiotic–abiotic stress category, followed by the biotic–
biotic stress category (Figure 3b). Evidently, combined
drought and heat stress cause substantial yield losses
(Cohen et al., 2021). Several upcoming studies have indi-
cated the role of combined drought, heat, and high light in
affecting plant development and metabolism (Zandalinas,
Fichman, et al., 2020; Zandalinas, Fritschi, et al., 2020,
2021). Apart from the drought–heat stress combination,
fungus–waterlogging and salinity–weeds stress combina-
tions substantially affected the yields of important mono-
cots like wheat and barley, respectively (Figure S11).
Wheat yield in particular was greatly affected under
drought–heat, drought–cold, boron deficiency–cold, and
Fusarium poae–waterlogging stress combinations
(Figure 3c). However, in the case of nematode–fungus and
fungus–fungus stress combinations, the wheat yield
response varied with the type of pathogen species
involved in the interaction and the order of stress per-
ceived by the plant, as shown in Figure 3c. The database
also highlights several other important stress combina-
tions significantly affecting plant yields. For example, in
pulses and oilseeds such as peanut, cowpea, soybean, and
common bean, yields were more affected under
nematode–fungus, ozone–UV, fungus–insects, and
drought–weeds stress combinations (Figure S11,
Table S1). Among solanaceous crops, fungus (Verticillium
dahlia) in association with nematodes (Heterodera rosto-
chiensis,Globodera rostochiensis, and Pratylenchus
neglectus) showed marked reduction in potato yield
(Figure S11). We have also done a comprehensive analysis
Figure 3. Phenome data analysis to assess the effect of stress combinations on agronomic traits.
(a) The bubble diagrams depict the total stress combinations covered in SCIPDb under abiotic–
abiotic, abiotic–biotic, and biotic–biotic stress categories. The size
of the bubble is directly proportional to the number of studies under the respective stress combinations. For crop-wise stress combinations, bubble diagrams
are presented in Supplemental Figures S8–S10.
(b) Schematic representation of the phenome application page and some of the key features offered on various plant traits. Phenome section can be successfully
utilized as knowledge resource for data mining and to address pertinent questions in plant science research. We show an example of two questions that can be
answered by the database. (i) The database can be used to understand the effect of abiotic–
abiotic stress combination on physiological traits, for example, pho-
tosynthetic efficiency. The orange dots indicate the ten different stress combinations analyzed in the database. We show an example of heat and salt stress com-
bination, where the bar graphs indicate the percentage reductions in photosynthetic efficiency under the individual and combined stress. (ii) Similarly, SCIPDb
can be utilized to assess the impact of stress combinations on plant performance. The blue dots indicate the nine different abiotic-abiotic stress combinations
analyzed in the database. We show an example of heat and drought stress combination, where the bar graphs indicate the percentage reductions in biomass
under the individual and combined stress. The enlarged versions of radial trees are given in Supplementary Figures S11–S13. and interactive versions are pre-
sented in database. Different colored dots are used to signify that the abiotic–abiotic stress combinations analyzed for (i) and (ii) were not same.
(c) As an example for ‘
applications’ tab, bar graph shows the impact of various stress combinations on wheat yield. The percentage change in parameter values
was computed by comparing the percentage change observed under stress conditions to that of the control. The treatment denoted by a star (*) indicates the
application of simultaneous stress, while all other treatments involve sequential stress applications. For abbreviations and pathogen names refer to Dataset S1.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
6Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
of the effect of different stress combinations on various
physiological traits (like photosynthesis, stomatal conduc-
tance, photochemical efficiency, Fv/Fm, and chlorophyll
content) of various plant species (Figure S12). Further, in
view of understanding the aggravation of plant diseases
and emergence of new disease complexes, we found that
biotic factors were critical in exacerbating several patho-
gen infections (Figure S13). In the biotic–biotic stress
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 7
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
category, nematode (Meloidogyne incognita and Hetero-
dera indicus) and fungus (Fusarium udum,F. oxysporum,
F. moniliforme, and Macrophomina phaseolina) stress
combinations were highly detrimental to maize, pigeon
pea, cotton, and chickpea crops, causing higher disease
incidence and damage compared to the individual stresses
(Figure S13). These results indicate that combined biotic
stresses are more detrimental to crops than the stressors
individually. In contrast, abiotic stresses have shown a
positive effect in terms of reducing pathogen infection and
its progression, e.g., ozone–Phytophthora sojae in soy-
bean, ozone–Bean common mosaic virus in pinto bean,
salinity–weeds in sorghum, shade–Colletotrichum kahawae
in coffee, Mn toxicity–Uncinula necator in grapevine, and
Pythium myriotylum–R. solani in peanut showed signifi-
cant reductions in disease incidence under combined
stress (Figure S13). However, recent reviews on
combined stress have also indicated that elevated drought,
high temperature, and poor nutrient conditions make
plants more vulnerable to pest or pathogen infection
(Cohen & Leach, 2020; Desaint et al., 2021; Hamann et al.,
2020; Savary & Willocquet, 2020).
Global transcriptome changes in combined stressed plants
Transcriptome analysis from 58 combined stress transcrip-
tomes resulted in 45,169 unique DEGs from 16 plant spe-
cies (‘downloads’ section in the database). Functional
profiling of significantly enriched DEGs revealed the
involvement of genes encoding key proteins like heat-
shock proteins (HSPs), Ca
2+
signal transduction proteins,
phytohormone-related genes, defense-related genes, reac-
tive oxygen species (ROS), peroxidases, cell wall–modify-
ing genes, and cytochrome P450 superfamily proteins.
Transcription factor (TF) enrichment analysis revealed sig-
nificant enrichment of dehydration response element-
binding protein (DREB), ABA-responsive element-binding
protein (ARF), ethylene-responsive element-binding factor
(ERF), heat-shock transcription factor (HSF), NAC domain–
containing protein, MYB, LOB domain–containing protein,
GATA TFs, and WRKY DNA-binding protein families in the
DEGs. MYBs and NAC TFs have been reported to regulate
pathogen and phytohormone responses like ethylene, jas-
monate, and/or salicylate (Bian et al., 2020, Vemanna et al.,
2019). MYB TFs have also been reported to regulate the
production of secondary metabolites during the induction
of stress responses via the phenylpropanoid pathway and
cell wall biosynthesis, making them excellent candidates
for broad-spectrum stress tolerance improvement in plants
(Atkinson et al., 2013; Cao et al., 2020; Rasmussen et al.,
2013; Zandalinas, Fritschi, et al., 2020; Zandalinas, Fich-
man, et al., 2020).
Twenty distinct combined stress transcriptomes were
meticulously examined in Arabidopsis (Figure 4). This
comprehensive analysis unveiled 10 804 DEGs that
exhibited exclusive expression patterns specifically under
combined stress conditions. Further stratifying these DEGs
into the primary categories of combined stress, followed
by an intersection analysis, highlighted 3587, 3182, and
866 DEGs uniquely associated with abiotic-biotic, abiotic-
abiotic, and biotic-biotic stress intersections, respectively
(Figure 5a).
Conducting pathway enrichment analysis on these
DEGs, specific to the major combined stress categories,
unveiled a range of pivotal pathway clusters that consis-
tently underwent alteration across the three primary com-
bined stress categories. These encompass pathways
integral to amino acid, carbohydrate, energy, carbon, lipid,
secondary metabolite, cofactor, and vitamin metabolism
(Figure 5b). Furthermore, within the biotic-biotic combined
stress categories, pathways related to glycan biosynthesis
and key metabolic pathways like glycosphingolipid biosyn-
thesis, glycosaminoglycan degradation, and N-glycan bio-
synthesis stood out prominently. Unique enrichment was
also observed in ethylene and phytochrome signaling
pathways, exclusively linked to biotic-biotic combined
stress conditions. Intriguingly, under the abiotic-biotic
stress category, distinctive interactions were found, includ-
ing those between sugar and hormone signaling, inositol
phosphate metabolism, photosynthesis, and ABC trans-
porter pathways. The abiotic-abiotic stress category exhib-
ited unique pathways such as glycosylphosphatidylinositol
(GPI)-anchor biosynthesis, mRNA surveillance pathways,
ketone body synthesis and degradation, and glucose sens-
ing and signaling in Arabidopsis.
This comprehensive analysis sheds light on the intri-
cate molecular responses orchestrated by Arabidopsis
under diverse combined stress scenarios, unraveling
unique pathway crosstalk and potential key regulators in
each stress category.
Deciphering key genes and pathways under combined
drought and heat stress by integrative multi-omics
While multi-omics approaches like joint pathway analysis
have been limited, they are now being increasingly used in
plants (Bjornson et al., 2017; Crandall et al., 2020; Lopez-
Hidalgo et al., 2018), with the underlying hypothesis that
by combining multi-omics evidence, it will be possible to
concretely pinpoint the pathways involved in the underly-
ing biological processes. Carbohydrate metabolism and
the expression of related genes have been identified to
contribute to the superior heat and drought tolerance of
anthers in the rice cultivar N22 compared to the cultivar
Moroberekan (Li et al., 2015). The joint pathway analysis
approach, integrating changes in gene expression, prote-
ome, and metabolite concentrations in drought and heat
combined stress treatments, suggested significant enrich-
ment of four major classes of pathways enriched based on
the KEGG BRITE hierarchy. Amino acid metabolism,
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
8Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
energy metabolism, carbohydrate metabolism, and signal
transduction pathways were supported by all three omics
(Figure 6) (Zandalinas et al., 2022). Within the amino acid
metabolism class, significantly enriched pathways were
related to glutathione metabolism; alanine, aspartate, and
glutamate metabolism; glycine, serine, and threonine
metabolism; and cysteine and methionine metabolism.
Pentose phosphate pathway; glycolysis or gluconeogene-
sis; and pyruvate, fructose, mannose, ascorbate, aldarate,
amino sugar, and nucleotide sugar metabolism pathways
were found to be enriched within the energy metabolism
class, while carbon fixation in photosynthetic organisms,
nitrogen metabolism, and sulfur metabolism pathways
were enriched under the carbohydrate metabolism class.
Among the signal transduction pathways, phosphatidylino-
sitol signaling system pathways mapped to all the three
omics data analyzed. Thus, genes commonly associated
between these pathway classes may play a significant role
in combined stress tolerance in plants (Figure 6).
Applications of SCIPDb and pipeline for novel genes and
pathway identification
The applications section of the database provides insight
(through multiple case studies) on how to use the diverse
Figure 4. Upset plot depicting intersections of combined stress genes analyzed in multiple combined stress transcriptomes in Arabidopsis.
Intersections among combined stress DEGs in Arabidopsis across different stress combinations is presented. The numbers above green bars indicate the num-
ber of genes within each intersection. Horizontal blue bars depict set size and set names. Connected dots (black) represent common genes between the tran-
scriptomes, while unconnected dots represent unique genes. Non-
overlaps are depicted as gray dots. Data used for the figure is available in Dataset S2.
Figure 5. Global pathway and process enrichment analysis of differentially expressed combined stress genes in Arabidopsis.
(a) Circos plot representing the overlap between category-
wise DEGs lists in abiotic–abiotic, abiotic–biotic, and biotic–biotic categories. The inner circle repre-
sents gene lists, where hits are arranged in the form of an arc. Genes that hit multiple lists are colored in dark orange, and genes unique to a list are shown in
light orange. Purple curves link shared genes between the three categories, and blue curves link genes that belong to the same enriched ontology term.
(b) Network representation of unique and common pathway clusters among the major combined stress categories. Analysis showed the enrichment of six main
pathway clusters, namely, carbohydrate metabolism, amino acid metabolism, lipid metabolism, metabolism of cofactors and vitamins, secondary metabolism,
and energy metabolism. Ellipse-
shaped nodes depicted as donuts are key pathway clusters (names indicated). The pathway clusters were grouped into broader
categories based on KEGG pathway classification (for details on each node, refer to the “Transcriptome –Visualize Transcriptomics data” link in the database).
Nodes in the circle represent the genes mapped to those pathways. The color of the nodes indicates the different enriched pathways and their corresponding
genes in green. The network is visualized with Cytoscape (v3.8.2) with a “Group by attribute circle” layout. The network of enriched terms is represented as
donut charts, where donuts are color-coded based on the identities of gene lists (Dataset S3). The size of a donut is proportional to the total number of hits that
fall into that specific term.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 9
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
datasets hosted in SCIPDb for uncovering the complexities of
combined stress in plants. SCIPDb can be used as a knowl-
edge resource to address pertinent unanswered questions
(Harris et al., 2020) in combined stress research, and the
information so deciphered can be further used for transla-
tional research to discover novel genes and pathways.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
10 Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
SCIPDb as a knowledge resource
Knowledge discovery from databases (KDD) is defined as
deciphering valid, novel, potentially useful, and under-
standable patterns in data. SCIPDb acts as a powerful com-
putational tool for KDD in the area of combined stress
research in plants. It has been developed and designed
with the aim of extracting new insights from an avalanche
of unstructured combined stress data available to date.
SCIPDb aims to complement laboratory experimentation
by prioritizing candidate genes/pathways, which can signif-
icantly accelerate combined stress research (Figures 7
and 8; Table S2).
The extensive datasets presented in the ‘SCIPDb phe-
nome’ tab help us visualize the impact of different stress
combinations on plant growth, yield, and physiology, as
well as their interactions with pathogens. Voluminous
information has been succinctly depicted as radial trees
(Figure 3b; Figures S11–S13). Interestingly, our analysis
suggests that the combinatorial infection of nematode and
fungus is one of the most critical stress combinations,
apart from the well-known drought–heat and drought–
pathogen combinations (Figures S11 and S12). The infor-
mation compiled in SCIPDb also provides clarity on vari-
ous inter-microbial interactions affecting plant–pathogen
interactions.
From an analysis of 278 research articles, we found
that 22 fungi, 13 nematodes, 7 oomycetes, 6 bacteria, and
5 viral pathogens are involved in various biotic–biotic
stress combinations. Among them, the most commonly
found biotic stressors in disease complexes involve Meloi-
dogyne incognita,Rhizoctonia solani, and Fusarium spp.
Similarly, an analysis of 441 research articles on various
abiotic–biotic stress combinations revealed drought, salin-
ity, temperature extremes, heavy metals, and low light
intensity as the chief abiotic factors affecting plant–patho-
gen interactions.
Furthermore, SCIPDb hosts an interactive global map
illustrating the distribution of various combined stresses
Figure 6. Integrative multi-
omics analysis to decipher key omics features and pathways differentially altered during the drought-heat stress combination in
Arabidopsis.
The network representation of differentially regulated genes, proteins, and metabolites under the drought and heat stress combination in Arabidopsis was done
via joint pathway analysis and visualized in the “
y-Files Organic layout” layout in Cytoscape (v3.8.2). The network is presented as nodes indicating various path-
ways and their associated omics features connected by edges. Edges have been bundled for clarity. The size of a pathway node represents the pathway impact
in terms of evidence from omics features, wherein arrow-headed nodes signify pathways having evidence from all the three omics features [transcriptomics (T),
proteomics (P), and metabolomics (M)]. Node color corresponds to the class of the pathway or features as mentioned in the node shape and color code box.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 11
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
worldwide (Figure 1e). Mapping the geographical distribu-
tion of abiotic–biotic combined stresses provides clues
as to where environmental conditions exacerbate
essential plant diseases. The geographical distribution of
abiotic–abiotic stress combinations, on the other hand,
highlights the emerging complex environmental condi-
tions negatively impacting plant performance. Similarly,
the global distribution of biotic–biotic stress combinations
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
12 Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
provides insight into places where new disease complexes
are emerging or being exacerbated by local environmental
conditions.
For scenarios where the effect of stressors depended
on factors such as stress intensity and order of imposition,
interactive scroll bars were employed. These scroll bars
allow users to access specific data points by manipulating
the scroll bar arrow (Figure S4). A similar presentation of
information is also offered in tabular form within the data
pages. To synthesize findings from diverse literature and
derive overarching conclusions, the study combined data
from multiple sources. However, it is important to note
that variations in response were expected. This under-
standing is particularly relevant when interpreting the
“Phenomics Data” section, as it may differ from observa-
tions made on individual data pages.
SCIPDb as a source for data mining for biological knowl-
edge discovery
Transcriptome datasets hosted in SCIPDb identify large
numbers of potential candidate genes that are shared and
unique under multiple combined stresses, together with
their expression value, expression trend, and several meta-
data, including KEGG pathways and gene information. It is
often tedious and time consuming for the scientific com-
munity to review, curate, and analyze the heterogenous
transcriptome information available from the literature and
biological databases in various data formats with divergent
quality and coverage. SCIPDb provides ready-to-use, easily
interpretable, and downloadable combined stress tran-
scriptomics data for further downstream analysis by end
users. A generic pipeline for knowledge-based gene dis-
covery using SCIPDb has been outlined in Figure 7a and is
also hosted in the “Application” tab of the database. The
figure outlines four different ways to mine analyzed tran-
scriptome datasets hosted in SCIPDb to filter genes of
interest.
SCIPDb for biological knowledge discovery for gene
prioritization
Multiple visualizations and computational tools hosted in
SCIPDb can aid in the filtering of genes based on a user’s
prior knowledge and interest. This can help in prioritizing
candidate genes and pathways that when perturbed
through potential experimental interventions, can have a
desired biological outcome in plants. Interactive
visualizations such as heatmaps, Venn diagrams, and GO
enrichment results in the form of Manhattan plots and co-
functional networks have been provided to aid users to fur-
ther filter and identify genes of interest based on functional
categorization (Figure 7a). Further SCIPDb transcriptome
dataset and its associated data and visualizations can be
used to mine, filter and identify candidate genes involved
in defense against stress combinations using a generic
pipeline described in Figure 7b. This figure demonstrates
how CBP60g and SARD1, which are known regulators of
SA-mediated defense responses against bacterial patho-
gens in Arabidopsis, were identified (Choudhary & Senthil-
Kumar, 2022). Although transcriptomic analysis of various
combined stressed plants has been done (Cohen et al.,
2021; Prasch & Sonnewald, 2013; Zandalinas, Fichman,
et al., 2020; Zandalinas, Fritschi, et al., 2020), SCIPDb pre-
sents a comparison of transcriptional changes occurring
due to different stress combinations. Using the pipeline,
we identified putative candidate genes involved in defense
against drought and pathogen stress in Arabidopsis and
chickpea (Table S2). The pipeline can also be applied to
identify combined stress-related genes in other plant
Figure 7. Workflow for identifying key genes responsive to combined abiotic–
biotic stress combinations in Arabidopsis using SCIPDb.
(a) Pipeline for knowledge-
based gene discovery. The transcriptome datasets hosted in SCIPDb can be mined using four different approaches to filter genes of
interest. 1. The SCIPDb FTP server can be used to directly download the analyzed combined stress transcriptomes (unique and shared genes between individual
and combined stress treatments). 2. In the combined stress category, it is possible to identify the differentially expressed genes (DEGs) in the pre-computed
comparisons based on three-step dropdown-based selection of the plant, followed by the stress combination, and finally, the category of desired genes. It is
also possible to visualize the DEGs interactively on KEGG pathways and download several other metadata using the KEGG genes link. 3. SCIPDb hosts a standa-
lone BLAST server and integrated database of unique combined stress genes identified in eight plant species. Users can perform sequence-based searches
using the BLAST server in batch mode to identify potential homologs or orthologs. 4. Candidate genes can also be identified using a keyword-based search:
The transcriptome search section accepts input in seven categories, namely, gene name, gene ID, pathway, stress combination, plant, and user-defined key-
words. The final result is depicted in the interactive bootstrap table, which presents lists of DEGs, gene names and other annotations, expression values, KEGG
pathway, and gene information. Multiple interactive visualizations like heatmaps, Venn diagrams, GO enrichment results in the form of Manhattan plots, and
co-functional networks, have been provided to aid users in identifying a gene of interest based on functional categorization (marked as a star).
(b) Case studies: Two case studies are illustrated here. b1 depicts a generic pipeline to identify genes of interest in any plant using the SCIPDb FTP server. A
five-
step workflow is shown starting from downloading combined stress DEGs from the SCIPDb FTP server. This step has been further detailed and demon-
strated using screenshots from the SCIPDb. The SCIPDb FTP server was browsed to navigate to the transcriptome datasets section of the database (Path: home/
downloads/transcriptome datasets). The server provides a hierarchical organization of analyzed combined stress transcriptomes from multiple plants. The plant
of interest (Arabidopsis) and stress combinations were selected (abiotic–biotic). The dataset was selected and downloaded locally. A total of 8243 DEGs from all
the abiotic–biotic stress combinations were downloaded, followed by filtering overlapping genes. This step fetched 5855 genes, from which potential genes
were selected based on expression values and functional categorization (star indicates use of metadata and visualizations as in A part). b2 demonstrates how
the search section of SCIPDb transcriptome was used to identify genes of interest. CBP60g and SARD1 are known regulators of SA-mediated defense responses
against bacterial pathogens in Arabidopsis. To decipher their potential role in combined stresses, SCIPDb transcriptome datasets were mined using CBP60g and
SARD1. The results from SCIPDb (depicted in screenshots as red hexagon symbol) showed the potential roles of these genes in combined drought and Pseudo-
monas syringae infection, which was validated by Choudhary and Senthil-Kumar (2022).
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 13
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
species as per the user’s interest. From our gene list con-
sisting of seven genes, we found AtPP2C49 in Arabidopsis
as an interesting novel candidate gene under drought and
pathogen stress (Figure 8). Protein phosphatase 2C (PP2Cs)
are class protein known to regulate ABA signaling pathway
through SnRK2 function (Miyazono et al., 2009). In total 19
PP2Cs have been identified in Arabidopsis and some are
involved in plant immunity against different pathogens
(Bhaskara et al., 2019; DeFalco et al., 2022; Mine et al.,
2017).
Omics feature and novel pathways identification
SCIPDb has been used to identify novel combined stress-
responsive pathways and omics features for abiotic–biotic
stress combinations in Arabidopsis (Figure 9). We inte-
grated the transcriptome datasets downloaded from
SCIPDb with the metabolome datasets using the integra-
tive joint pathway analysis approach. The major pathways
identified were categorized into amino acid metabolism,
carbohydrate metabolism, energy metabolism, lipid
metabolism, and signal transduction based on KEGG
BRITE hierarchy (Dataset S4). Finally, alanine, aspartate,
and glutamate metabolism were chosen from the amino
acid metabolism category (one of the top pathways
enriched), and omics features (transcripts and metabolites)
mapped onto it were back searched using the search but-
ton of SCIPDb to decipher their potential role under com-
bined stress. Based on the result obtained from SCIPDb,
we propose alanine, aspartate and glutamate metabolism
as a novel pathway [through asparagine synthetase 1
(AtASN1), AT5G65010 in Arabidopsis] with a potential role
in defense against drought and pathogen stress
combination.
FUTURE PERSPECTIVES
SCIPDb is a comprehensive database amenable to data
mining and data-driven research on combined stresses in
plants. Global phenome data analysis shows that abiotic–
abiotic stress combinations are major threats to crop pro-
ductivity. In the face of global climate change, the occur-
rence of these stress combinations is projected to increase
in coming years. Therefore, dedicated studies on this
aspect are essential to sustain crop yields in the future.
Key takeaways from yield analyses are that monocots will
be more affected under the abiotic–abiotic and abiotic–
biotic stress category, whereas pulses, oilseeds, and vege-
table crops will be more affected under the biotic–biotic
stress category.
Our unique combined stress integretome developed
using multi-omics data integration highlights sugar metab-
olism, energy metabolism, and amino acid metabolism as
the key pathways operating under combined stress condi-
tions. The addition of proteomics and metabolomics data
to the database with multi-omics analysis will further
Figure 8. SCIPDb as a tool for identifying candidate genes involved in com-
bined abiotic-
biotic stress resistance: Evidence from drought and pathogen
studies in Arabidopsis.
Out of several genes shortlisted from the SCIPDb transcriptome (as
described in Figure 7) seven were validated for their role in defense against
combined drought and Pseudomonas syringae pv. tomato DC3000 (Pto
DC3000) infection in Arabidopsis thaliana. Graph showing (a) in planta bac-
terial number and (b) electrolyte leakage in mutants under combined
drought and Pto DC3000 infection. Bars represent data collected from at
least six biological replicates. Error bars designate the standard error of
mean, and different letter indicates the significant difference at P
<0.05
(two-way-ANOVA followed by Duncan’s multiple range test). The experi-
ment was repeated thrice and found similar trend.
(c) Plant phenotype showing the extent of disease incidence in wild-
type
and atpp2c49 mutants under pathogen-only and combined stress treat-
ments. Numbers inside the pictures indicate the percent disease incidence
under respective treatments.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
14 Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
demystify combined stress responses of plants. Unraveling
the mechanisms by which the molecular signatures associ-
ated with these pathways impact plant responses to com-
bined stress can open new vistas for developing resilient
crop varieties with better adaption to changing climate and
global warming.
The pipelines presented in the database can be useful
to identify novel genes and pathways from other crop and
model plants for various stress combinations. Hence, this
database can be used to advance fundamental knowledge
of plant stress tolerance mechanisms. Further, we have
shown case studies in this manuscript for substantiating
the same along with the potential for applying the informa-
tion to improve plant performance under stress.
With the continual accumulation of available data in
the field of combined stress, we will update the database
annually by incorporating newer studies. There have been
recent advancements on the varying vulnerability of differ-
ent growth stages and plant tissues to combined stresses.
Considering this, based on data availability we will be seg-
regating the transcriptomic datasets based on growth
stages and plant tissues. We intend to add other omics
datasets related to combined stress research in the future
version of SCIPDb, together with novel features like predic-
tion modeling based on machine learning and meteorolog-
ical data integration with geographical distribution
information. Overall, SCIPDb is an informative and valu-
able resource for combined stress research in plants.
MATERIALS AND METHODS
Bio curation and integration of phenome datasets into
SCIPDb
Literature mining
To retrieve all available articles and to have greater than 90% liter-
ature coverage, several search engines were queried using suit-
able and carefully designed keywords (including several variants).
Bibliography from each article was also searched to achieve better
coverage. Table S3 details the search engines used for the phe-
nome literature mining. A detailed pipeline is presented under the
SCIPDb methodology tab (http://223.31.159.3/plant_complete/
methlogy_orangesunset.php).
Sorting of articles
After retrieving all the articles related to particular stress combi-
nation, they were sorted as ‘main research articles (Science Cita-
tion Index articles)’ and ‘ancillary articles’. This was done based
on the type of articles (e.g., research, review, reports, etc.) and
type of data (e.g., morpho-physiological and molecular data)
present in the articles. Main research articles with only morpho-
physiological data were considered for data extraction, whereas
reports, thesis, book chapters, abstracts, reviews, gene overex-
pression, and gene-silencing studies were listed under ancillary
articles and not considered for data extraction. All these types of
articles were integrated into the database under the ‘phenomics’
tab. Plant competition, in this database, implies intra-specific
competition and constituted studies related to the effect of plant
density involved with other stresses on plants. Cases of inter-
specific competitions, in most of the cases, have not been dis-
cussed under this category. The studies involving plant growth
promoting bacteria and fungi were excluded as both of them
independently do not act as pathogen to plants. In most cases
tree species were not included, however, in some places where
the economically important tree species have been considered.
Studies wherein the experiments involved young saplings were
included. The studies with gene overexpression or silencing
were not considered for the data extraction however they were
included as ancillary articles under the particular stress combi-
nation. Articles with chemical treatment like fungicide, herbicide,
and fertilizer amendments were also excluded since these are
anthropogenic stresses.
Listing out parameters and their classification
From the main research articles, parameters studied in each article
were listed out and classified into type A, B, and C parameters
based on their significance in reflecting the net impact of the stress.
Type A includes growth (plant height, biomass, leaf area, leaf num-
ber, root length, shoot weight, root weight, etc.) and yield (seed
weight, seed number, test weight, etc.), attributing parameters that
directly reflect the impact of stress, while Type B includes physio-
logical (photosynthesis, stomatal conductance, transpiration, chloro-
phyll content, etc.) and pathogenesis (disease index, pathogen load,
disease score, etc.) related parameters which indirectly reflect the
impact of stress. Finally, biochemical parameters such as proline
content, MDA content, nutrient composition, ROS content, etc.,
which also explains the impact of stress but to a lesser extent com-
pared to the other two classes of parameters were categorized as
Type C. For the complete list of parameters, refer “Traits” Tab in
the SCIPDb (http://223.31.159.3/plant_complete/traits.php).
Data extraction and depiction
Once parameters were listed out from each article, data values
were extracted into the Excel file. The values from the table were
directly copied into the Excel sheet, whereas values from graphs
were extracted using the ‘GetData Graph Digitizer’ (http://getdata-
graph-digitizer.com/) tool for better accuracy. To make the hetero-
geneous data uniform, normalization was done using the formula
mentioned below.
Change over control %ðÞ¼
ControlstressðÞ100
Control
Few parameters, such as electrolyte leakage, pathogenesis-
related parameters, etc., were not subjected to calculation. Using
both the calculated and un-calculated values, a table was pre-
pared, reflecting the net impact of stress and interaction between
the two stresses at the plant interface. This table was used for pre-
paring the ‘data page’ file for each study on a specific plant and
was finally represented in the database in tabular form. For easy
understanding, percent change values were shown along with
arrows in red and green color. A red-colored downward arrow
indicates that the parameter is affected under stress; the higher
the positive value greater the damage to the parameter under
stress. Green-colored upward arrow indicated parameters were
not affected under stress conditions as compared to control.
Data analysis and interpretation
Data presented in tabular form was analyzed by comparing the
individual and combined stress values of each parameter. When
the percent change values was greater in combined stress
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 15
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
compared to both the individual stresses, the outcome of com-
bined stress was depicted as ‘negative’. If percent change values
were less in combined stress compared to both the individual
stress, then the outcome was depicted as ‘positive’. Together with
the tabular part, a brief inference was also documented for each
article. Each of the data pages has a brief introduction of the
stress combination, with information about the number of studies
available in different crops for that particular stress combination.
Mapping of important traits hosted in SCIPDb was done using the
Plant Trait Ontology (http://www.ontobee.org/ontology/TO) and
Planteome (https://planteome.org/) databases wherever possible.
The impact of stress combinations was evaluated consider-
ing the developmental stage of the plants, the timing of stress
application, and the order in which stressors were introduced.
Detailed data pages were constructed to present information for
each study conducted. An illustrative example can be found on
the data pages where the top right corner showcases the factors
under consideration. Each study was treated independently, and
analyses were conducted on an individual basis, avoiding data
aggregation. Acknowledging potential variability in stress
response across different genotypes, distinct links were provided
on the data pages to download genotype-specific Excel sheets
containing comprehensive data.
The methodology employed for this study included a com-
prehensive consideration of influencing factors, individualized
study analyses, inclusion of genotype-specific responses, interac-
tive data exploration tools, and recognition of potential variations
in the aggregated data analysis.
Phenomics datasets integration into the SCIPDb
The frontend user interface was implemented using HTML5,
CSS and PHP (version: 7.0.12). The dataset was finally inte-
grated and depicted as an HTML page presented to the users
based on the three-level-dropdown selection designed in Java-
Script, specific to each plant species. The back-end schema was
designed using MySQL, an open-source relational database
management system, and stored in MySQL tables (Version:
5.7.17). To provide an interactive interface and enhanced user
experience, Bootstrap 4 and jQuery were used. For data page
format, please visit the ‘phenomics’ section of database (http://
223.31.159.3/plant_complete/data.php).
Visualization of phenomics dataset
We employed Tableau Public Desktop (Version 2020.4, available at
https://public.tableau.com/en-us/s/) to visualize intricate phe-
nomics data with high dimensions. This visualization took the
form of an interactive Treemap/stress matrix. The stress matrix
was constructed based on the frequency of studies demonstrating
positive or negative outcomes within various stress combinations.
In this context, a positive interaction denoted that the combined
stress resulted in less harm to the plants compared to individual
stresses. Conversely, a negative interaction indicated instances
where the plants experienced greater damage under combined
stress conditions. To illustrate, if a particular stress combination
was the subject of 12 research articles and the majority of these
articles reported a favorable impact, the combination was catego-
rized as “positive.” Equally distributed instances of positive and
negative outcomes led to a classification of “positive/negative.”
For the development of radial trees, we utilized Flourish Stu-
dio (accessible at https://app.flourish.studio/). Additionally, we
harnessed various interactive visualization tools offered by Flour-
ish Studio, including Chord diagrams and Sankey Diagrams, to
effectively represent the analyzed information. Furthermore, an
interactive geographical map was generated using Google My
Maps, which can be accessed at https://www.google.com/maps/
about/mymaps/.4.2.
Bio curation and integration of transcriptome datasets
into SCIPDb
Data mining
The relevant transcriptome datasets for combined stress in plants
were compiled and curated using two major public databanks for
microarray data, including Gene Expression Omnibus (GEO)
(https://www.ncbi.nlm.nih.gov/geo/) and Array Express (http://
www.ebi.ac.uk/arrayexpress/). The NCBI GEO and ArrayExpress
functional genomics repository were queried using keywords,
“combined stress” AND “Plants” [organism]. For the compilation
of RNA-seq transcriptomics data NCBI, Sequence Read Archive
(SRA) (https://www.ncbi.nlm.nih.gov/sra) database was used.
Data curation
The transcriptomic datasets mined from public databases were
manually curated, to determine whether they were from actual
combined stress studies. Also, duplicate studies were filtered out.
For the final analysis, studies that showed availability of
completely raw and processed datasets and also at least 4 sam-
ples (2 combined stress treatments and 2 respective controls,
comparable case and control samples) were included.
Analysis of transcriptomics dataset
Microarray data mined from relevant studies were analyzed in the
R environment using the GEOquery (http://www.bioconductor.org/
Figure 9. Workflow for unveiling novel pathways and omics features in response to abiotic and biotic stress combinations in Arabidopsis.
1. The transcriptome datasets hosted in SCIPDb were mined using the in-
house FTP server to download and filtered all the DEGs from abiotic–biotic stress com-
binations. The four-step workflow (Figure 7) was used to download 8243 genes. 2. Similarly, the analyzed metabolome dataset was downloaded for the drought
and pathogen stress combination. 3. Integrative joint pathway analysis was done using MetaboAnalystR to identify novel pathways and omics features (tran-
scripts and metabolites) differentially altered under drought and pathogen stress combinations. 4. The pathways identified were further classified using KEGG
BRITE hierarchy into major pathway categories. The top 50 pathways (sorted based on Q value) plotted in the form of a bar graph are shown in (a). The Y-axis
depicts the pathway names, while the X-axis shows the gene count (number of genes or metabolites mapped onto these pathways). Bar color is based on the -
log10 (P) value. The visualization has been faceted based on major KEGG pathway categories, namely, AM: amino acid metabolism, CM: carbohydrate metabo-
lism, LM: lipid metabolism, EM: energy metabolism, ST: signal transduction, PS: protein synthesis, EA: environmental adaptation, TC: transport and catabolism,
SS: secondary metabolite synthesis, MP: metabolism of terpenoids and polyketides, RR: replication and repair, MV: metabolism of cofactors and vitamins,
DNAR: DNA replication and repair, and CC: cellular transport and catabolism. AM, CM, EM, LM, and ST were the main pathway categories identified. 5. Network
analysis was done to identify major gene clusters or hub genes. The top 50 pathways (magenta diamond-shaped nodes) and their associated omics features
(transcript: green square-shaped nodes, metabolites: yellow octagon-shaped nodes) are shown in (b). 6. Alanine, aspartate, and glutamate metabolism were
chosen from the amino acid metabolism category [highlighted in maroon in (b)] and omics features were mapped onto it (shown in magnified view). The
mapped omics features are represented in (c) in green. 7. To further gain insights into the functional role of all the mapped omics features, they were searched
using the keywords (gene names) in SCIPDb. 8. Asparagine synthetase 1 (ASN1, AT5G65010) showed a potential role under combined stress.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
16 Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
packages/2.8/bioc/html/GEOquery.html) and limma (Linear Models
for Microarray Analysis) (http://www.bioconductor.org/packages/
release/bioc/html/limma.html) R packages from the Bioconductor
project. The GEOquery R package performs the initial parsing of
the GEO data into R data structures. This data is used by the
limma R package for identifying differentially expressed genes
(DEGs) in the input dataset. Following this probe ID conversion
was done for DEGs obtained from individual microarray datasets.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 17
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Based on the platform used in the individual studies probe ID
match was done to convert the DEGs into a common ID format.
Affymetrix (https://www.affymetrix.com/analysis/netaffx/
xmlquery_ex.affx?netaffx=wtgene_transcript) platform was used
to do the probe ID conversion to fetch their corresponding tran-
script IDs. For Agilent and other platforms, the bioDBnet (biologi-
cal DataBase network) platform was used (https://biodbnet-abcc.
ncifcrf.gov/db/db2db.php). In cases where multiple probes
matched the same locus, the probe ID with the highest fold
change was considered. For RNA-seq transcriptome datasets anal-
ysis, ENA (European Nucleotide Archive) (https://www.ebi.ac.uk/
ena/browser/home) and SRA (Sequence Read Archive) toolkit
(Version 2.10.8) were used to download and split the SRA raw
read data from NCBI (National Center for Biotechnology Informa-
tion) into FASTQ files (https://www.ncbi.nlm.nih.gov/). Raw
sequence reads in the format FASTQ files, thus downloaded were
subjected to quality check by FastQC (https://www.bioinformatics.
babraham.ac.uk/projects/fastqc/) followed by removal of adapter
sequences and subsequent trimming of the reads using Trimmo-
matic (http://www.usadellab.org/cms/?page=trimmomatic)or
Cutadapt (https://cutadapt.readthedocs.io/en/stable/) based on the
FASTQC report. This was followed by mapping the reads to
the reference genome using HISAT2 (Hierarchical Indexing for
Spliced Alignment of Transcripts 2, version 2.0.1) (http://www.ccb.
jhu.edu/software/hisat/index.shtml) which is a fast and sensitive
spliced alignment program for mapping RNA-seq reads. SAM
(Sequence Alignment/Map format) files to BAM (Binary Align-
ment/Map format) files conversion was done using SAMtools
(http://samtools.sourceforge.net, version 0.1.19). BAM files were
then further sorted. Subsequent to this counting of reads mapped
to individual genes or transcripts was done using bedtools (https://
bedtools.readthedocs.io/en/latest/content/installation.html). In the
final step differentially expressed genes between different combined
stress and individual stress treatments were identified using DESeq2
in R (https://bioconductor.org/packages/release/bioc/html/DESeq2.
html). For plants like Solanum dulcamara where the reference
genome wasn’t available de novo RNA Seq analysis was performed.
After quality check and subsequent trimming of reads de novo
assembly was generated using Trinity (https://github.com/
trinityrnaseq/trinityrnaseq.github.io/blob/master/deprecated/index.
asciidoc) followed by abundance calculation using RSEM (https://
github.com/deweylab/RSEM) or Salmon (https://salmon.
readthedocs.io/en/latest/salmon.html). Differential expression anal-
ysis was done using Trinity-Edge R (https://bioconductor.org/
packages/release/bioc/html/edgeR.html). Assembled reads were
annotated using the NR/NT NCBI blast database (https://blast.ncbi.
nlm.nih.gov/Blast.cgi). For transcriptome studies where analyzed
data was already provided by the authors, DEGs were directly
downloaded and used for further analysis pipelines.
Biological interpretation of transcriptome data
The DEGs identified under each stress combination were associ-
ated with several meta-data, like, gene names and mapped to
‘KEGG (Kyoto Encyclopedia of Genes and Genomes) Pathway’. In
order to aid users to navigate through several other databases
and cross-references outside KEGG and fetch a deluge of
meta-data corresponding to a particular entry, KEGG genes links
were provided in cases where pathway information was not avail-
able. KEGG API (Application Programming Interface) (https://
www.kegg.jp/kegg/rest/keggapi.html) was used to fetch both of
these links for respective DEGs. In order to further gain insights
into the biological functions of DEGs identified by transcriptome
analysis, functional and pathway enrichment analysis was per-
formed using the gProfileR package (https://cran.r-project.org/
web/packages/gProfileR/). Benjamini–Hochberg adjustment for
multiple hypothesis testing was used to correct the P-values. The
over-represented GO (Gene Ontology) and KO (KEGG Ontology)
terms measured by the adjusted P-values depicted gene function
and biological pathway associations.
Data integration into the SCIPDb
The frontend user interface was implemented using HTML5, CSS,
Jquery and PHP (version: 7.0.12). HTML pages were designed to
accept queries based on a three-level dropdown-based selection
using JavaScript, specific to each plant species. The back-end
schema was designed using MySQL, an open-source relational
database management system, and data was stored in mysql
tables (version: 5.7.17). To provide an interactive interface and
enhanced user experience, we used Bootstrap 4, to present final
results. For further details please visit the ‘transcriptomics’ section
of database (http://223.31.159.3/plant_complete/trancriptome.php).
Visualizations
To visualize the high-dimensional transcriptomics data, several in-
house scripts were generated using shell scripts and R. The heat
map was made interactive using the “Heatmaply” (https://github.
com/talgalili/heatmaply) package from R. “VennDiagram” (https://
cran.r-project.org/web/packages/VennDiagram/) package was used
and scripts were customized to generate color-coded Venn dia-
grams representing specific categories. Gene ontology enrichment
analysis and interactive visualization of the results in the form of
Manhattan plots were done using “gprofilerR” package (https://
cran.r-project.org/web/packages/gProfileR/index.html). Correlates
data for the top twenty genes in each category have been col-
lected from ATTED-II (http://atted.jp/) which is a plant co-
expression database and for Arabidopsis from AraNetv2 (https://
www.inetbio.org/aranet/). Functional annotation and pathway
mapping for the correlates was done using KEGG and TAIR. Final
visualization of the network together with the other meta-data was
done using Cytoscape (https://cytoscape.org/), an open-source
software platform, widely used for visualizing complex networks.
Transcription factor enrichment analysis was done using EatupTF
tool (http://chromatindynamics.snu.ac.kr:8080/EatupTF).
In the process of creating co-functional gene networks for A.
thaliana and other plants, we utilized AraNet v2, accessible at
(http://www.inetbio.org/aranet). Correlated genes associated with
the top 20 uniquely differentially expressed genes under com-
bined stress were retrieved from the AraNet server, each interac-
tion being assigned a log-likelihood score (LLS) reflecting the
probability of genuine functional linkage between genes. For visu-
alization purposes, the downloaded data underwent conversion
into a sif format (Simple Interaction Format) using the Cytosca-
pe.js library. The resultant visualization focused on the top 200
correlated genes, while the full correlation dataset remains avail-
able for reference. A Node Attributes (.noa) file was compiled by
merging various datasets, offering users comprehensive informa-
tion accessible through interactions with individual nodes. This
includes details such as Gene Ontology Biological Process (GO-
BP), Gene Ontology Molecular Function (GO-MF), Gene Ontology
Cell Compartmentalization (GO-CC), log-likelihood scores (LLS),
rank among the top 200 correlated genes, evidence codes, paralog
information, linked queries, candidate genes, and gene symbols.
Moreover, edge information, attained by clicking on network
edges, provides insights into connected nodes. The visualization
process was empowered by the Cytoscape.js library, an
open-source JavaScript-based graph theory and network library,
allowing interactive graph manipulation and analysis. A range of
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
18 Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
interactive features, such as pinch-to-zoom, box selection, and
panning, were supported, along with customizable styling, lay-
outs, and background options. Ultimately, this approach enriches
the exploration of DEGs by unveiling their functional relationships
amid distinct stress conditions.
Arabidopsis combined stress transcriptome
Upset plot was generated using UpSetR package, while circos plot
was generated using Metascape, a gene annotation & analysis
resource. (https://metascape.org/gp/index.html#/main/step1). Path-
way enrichment analysis was done using major pathway data-
bases like KEGG (https://www.kegg.jp/kegg/rest/keggapi.html),
Aracyc (https://plantcyc.org/typeofpublication/aracyc) and Wiki-
pathways (https://www.wikipathways.org/index.php/WikiPathways).
Final visualization and network analysis were done using Cytoscape,
v3.8.2 (https://cytoscape.org/).
Integretome analysis
MetaboAnalystR package was used to perform joint pathway anal-
ysis of transcriptome, proteome, and metabolome profiles. For
enrichment analysis (ORA) hypergeometric analysis was used,
while for topology measure, degree centrality was used. Combin-
ing P-values at pathway level was used for integration of three
omics datasets. Final visualization and network analysis was done
using Cytoscape, v3.8.2 (https://cytoscape.org/). A detailed list of
enriched pathways is given in dataset S3.
FUNDING INFORMATION
This work was supported by funding to M.S-K. majorly
from the National Institute of Plant Genome Research core
funding and partly by a project from Science and Engineer-
ing Research Board (SERB; CRG/2019/005659). Pi.P. and
M.P were supported by fellowships from CSIR (No.13
(9106-A)/2020-Pool) and No.13 (9064-A)/2019-Pool)),
respectively.
AUTHOR CONTRIBUTIONS
MS-K conceived the idea, designed the database, outlined
the manuscript, and provided all resources. Pi.P developed
the webtool, performed data integration and data visualiza-
tion. MP and Pr.P contributed to the phenomics part. Pi.P,
MP, and Pr.P performed overall data analysis. AS contrib-
uted to the data collection for the phenomics part of the
manuscript. Pi.P and VSB contributed to the data collection
and analysis part of transcriptomics. MS-K, Pi.P, MP, and
Pr.P drafted the manuscript. MS-K edited and finalized
the manuscript and SCIP database. All authors agreed to
the submitted version of the manuscript.
ACKNOWLEDGEMENTS
The authors are thankful to the Department of Biotechnology
(DBT) eLibrary Consortium, India, and the NIPGR library for pro-
viding access to e-resources. Computational facilities provided by
Genome Analysis Facility and the DBT-DISC facility at NIPGR are
also duly acknowledged for sharing resources. We acknowledge
Dr. Jyoti Singh for her contribution in collecting data for the
phenome part.
CONFLICT OF INTEREST
The authors declare no conflicts of interest.
DATA AVAILABILITY STATEMENT
The curated phenomics, genotypes, transcriptomics, and
references are hosted on a local FTP server and can be
downloaded by the users using the link http://223.31.159.3/
plant_complete/downloads.php.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online ver-
sion of this article.
Dataset S1. Raw data for phenome data analysis to assess the
effect of stress combinations on agronomic traits.
Dataset S2. DEGs under various stress combinations in Arabidop-
sis thaliana used to generate Upset Plot
Dataset S3. GO and Pathway enrichment analysis of DEGs under
various stress combination in Arabidopsis
Dataset S4. Integrative multi-omics analysis to decipher pathways
jointly supported by all omics features (transcriptome, proteome
and metabolome) during drought and heat stress combination in
Arabidopsis.
Figure S1. Illustration indicating the various effects of abiotic
stresses on phytopathogens and the plant defense
Figure S2. Content and construction of SCIPDb.
Figure S3. Combined stress transcriptome analyzed and integrated
into SCIPDb.
Figure S4. A typical data page entry for phenome in SCIPDb.
Figure S5. Stress matrix representing the impact of stress combi-
nations curated and integrated in SCIPDb.
Figure S6. Literature analysis of combined stress articles pub-
lished from 1950 to 2021.
Figure S7. A typical data page entry for transcriptome in SCIPDb
and its associated visualizations.
Figure S8. Literature analysis of combined stress articles pub-
lished from 1950 to 2021 under the abiotic–abiotic stress category.
Figure S9. Literature analysis of combined stress articles pub-
lished from 1950 to 2021 under the abiotic–biotic stress category.
Figure S10. Literature analysis of combined stress articles pub-
lished from 1950 to 2021 under the biotic–biotic stress category.
Figure S11. Literature analysis of combined stress growth and
yield data for various plant species.
Figure S12. Literature analysis of combined stress physiological
data for various plant species.
Figure S13. Literature analysis of disease incidence data under
combined stress for various plant species.
Table S1. List of references of important stress combinations dis-
cussed in results section of manuscript.
Table S2. List of putative candidate genes identified from SCIPDb
with potential role under combined drought and pathogen stress.
Table S3. List of search engines used for literature mining for Phe-
nomics dataset.
REFERENCES
Ahuja, I.,de Vos, R.C.,Bones, A.M. &
Hall, R.D. (2010) Plant molecular stress
responses face climate change. Trends in Plant Science,15, 664–674.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 19
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Atkinson, N.J.,Lilley, C.J. &
Urwin, P.E. (2013) Identification of genes
involved in the response of Arabidopsis to simultaneous biotic and abi-
otic stresses. Plant Physiology,162, 2028–2041.
Atkinson, N.J. &
Urwin, P.E. (2012) The interaction of plant biotic and abi-
otic stresses: from genes to the field. Journal of Experimental Botany,
63, 3523–3543.
Bhaskara, G.B.,Wong, M.M. &
Verslues, P.E. (2019) The flip side of
phospho-signalling: regulation of protein dephosphorylation and the
protein phosphatase 2Cs. Plant, cell & environment,42, 2913–2930.
Bian, Z.,Gao, H. &
Wang, C. (2020) NAC transcription factors as positive or
negative regulators during ongoing battle between pathogens and our
food crops. International Journal of Molecular Sciences,22, 81.
Bjornson, M.,Balcke, G.U.,Xiao, Y.,de Souza, A.,Wang, J.Z.,Zhabinskaya,
D. et al. (2017) Integrated omics analyses of retrograde signaling
mutant delineate interrelated stress-
response strata. Plant Journal,91,
70–84.
Borkotoky, S.,Saravanan, V.,Jaiswal, A.,Das, B.,Selvaraj, S.,Murali, A.
et al. (2013) The Arabidopsis stress responsive gene database. Interna-
tional Journal of Plant Genomics,2013, 949564.
Cao, Y.,Li, K.,Li, Y.,Zhao, X. &
Wang, L. (2020) MYB transcription factors
as regulators of secondary metabolism in plants. Biology,9, 61.
Choudhary, A. &
Senthil-Kumar, M. (2022) Drought attenuates plant defence
against bacterial pathogens by suppressing the expression of CBP60g/
SARD1 during combined stress. Plant, Cell and Environment,45, 1127–
1145.
Cohen, I.,Zandalinas, S.I.,Huck, C.,Fritschi, F.B. &
Mittler, R. (2021) Meta-
analysis of drought and heat stress combination impact on crop yield
and yield components. Physiologia Plantarum,171,66–76.
Cohen, S.P. &
Leach, J.E. (2020) High temperature-induced plant disease
susceptibility: more than the sum of its parts. Current Opinion in Plant
Biology,56, 235–241.
Crandall, S.G.,Gold, K.M.,Jim´
e
nez-Gasco, M.D.M.,Filgueiras, C.C. &Will-
ett, D.S. (2020) A multi-omics approach to solving problems in plant dis-
ease ecology. PLoS One,15, e0237975.
DeFalco, T.A.,Anne, P.,James, S.R.,Willoughby, A.C.,Schwanke, F.,
Johanndrees, O. et al. (2022) A conserved module regulates receptor
kinase signalling in immunity and development. Nature Plants,8, 356–
365.
Desaint, H.,Aoun, N.,Deslandes, L.,Vailleau, F.,Roux, F. &
Berthom´
e, R.
(2021) Fight hard or die trying: when plants face pathogens under heat
stress. New Phytologist,229, 712–734.
Gupta, A.,Sarkar, A.K. &
Senthil-Kumar, M. (2016) Global transcriptional
analysis reveals unique and shared responses in Arabidopsis thaliana
exposed to combined drought and pathogen stress. Frontiers in Plant
Science,7, 686.
Hamann, E.,Blevins, C.,Franks, S.J.,Jameel, M.I. &
Anderson, J.T. (2020)
Climate change alters plant–herbivore interactions. New Phytologist,
229, 1894–1910.
Harris, J.M.,Balint-
Kurti, P.,Bede, J.C.,Day, B.,Gold, S.,Goss, E.M. et al.
(2020) What are the top 10 unanswered questions in molecular plant-
microbe interactions? Molecular Plant-Microbe Interactions,33, 1354–
1365.
IPCC. (2022). Climate change 2022: Impacts, adaptation, and vulnerability.
In: P ¨
o
rtner H-O, Roberts DC, Tignor M, Poloczanska ES, Mintenbeck K,
Alegría A, Craig M, Langsdorf S, Löschke S, M¨
oller V, Okem A, & Rama
B. (Eds.), Contribution of Working Group II to the Sixth Assessment
Report of the Intergovernmental Panel on Climate Change (2022). Cam-
bridge and New York: Cambridge University Press.
Li, X.,Lawas, L.M.,Malo, R.,Glaubitz, U.,Erban, A.,Mauleon, R. et al.
(2015) Metabolic and transcriptomic signatures of rice floral organs
reveal sugar starvation as a factor in reproductive failure under heat and
drought stress. Plant, Cell and Environment,38, 2171–
2192.
Lopez-
Delacalle, M.,Silva, C.J.,Mestre, T.C.,Martinez, V.,Blanco-Ulate, B.
&Rivero, R.M. (2021) Synchronization of proline, ascorbate and oxidative
stress pathways under the combination of salinity and heat in tomato
plants. Environmental and Experimental Botany,183, 104351.
L´
o
pez-Hidalgo, C.,Guerrero-S´
anchez, V.M.,G´
omez-G´
alvez, I.,S´
anchez-
Lucas, R.,Castillejo-S´
anchez, M.A.,Maldonado-Alconada, A.M. et al.
(2018) A multi-omics analysis pipeline for the metabolic pathway recon-
struction in the orphan species Quercus ilex. Frontiers in Plant Science,
9, 935.
Mahalingam, R.,Pandey, P. &
Senthil-Kumar, M. (2021) Progress and pros-
pects of concurrent or combined stress studies in plants. Annual Plant
Reviews,15, 813–868.
Mine, A.,Berens, M.L.,Nobori, T.,Anver, S.,Fukumoto, K.,Winkelm ¨
u
ller,
T.M. et al. (2017) Pathogen exploitation of an abscisic acid- and
jasmonate-inducible MAPK phosphatase and its interception by Arabi-
dopsis immunity. Proceedings of the National Academy of Sciences
U.S.A,114, 7456–7461.
Mittler, R. (2006) Abiotic stress, the field environment and stress combina-
tion. Trends in Plant Science,11,15–
19.
Mittler, R. &
Blumwald, E. (2010) Genetic engineering for modern agricul-
ture: challenges and perspectives. Annual Review of Plant Biology,61,
443–462.
Miyazono, K., II,Miyakawa, T.,Sawano, Y.,Kubota, K.,Kang, H.J.,
Asano, A. et al. (2009) Structural basis of abscisic acid signalling.
Nature,462, 609–
614.
Naika, M.,Shameer, K.,Mathew, O.K.,Gowda, R. &
Sowdhamini, R. (2013)
STIFDB2: an updated version of plant stress-responsive transcription fac-
tor database with additional stress signals, stress-responsive transcrip-
tion factor binding sites and stress-responsive genes in Arabidopsis and
rice. Plant & Cell Physiology,54, e8.
Pan, Q.,Wei, J.,Guo, F.,Huang, S.,Gong, Y.,Liu, H. et al. (2019) Trait ontol-
ogy analysis based on association mapping studies bridges the gap
between crop genomics and Phenomics. BMC Genomics,20,1–
13.
Pandey, P.,Irulappan, V.,Bagavathiannan, M.V. &
Senthil-Kumar, M. (2017)
Impact of combined abiotic and biotic stresses on plant growth and ave-
nues for crop improvement by exploiting physio-morphological traits.
Frontiers in Plant Science,8, 537.
Pandey, P.,Ramegowda, V. &
Senthil-Kumar, M. (2015) Shared and unique
responses of plants to multiple individual stresses and stress combina-
tions: physiological and molecular mechanisms. Frontiers in Plant Sci-
ence,6, 723.
Prasch, C.M. &
Sonnewald, U. (2013) Simultaneous application of heat,
drought, and virus to Arabidopsis plants reveals significant shifts in sig-
naling networks. Plant Physiology,162, 1849–1866.
Rasmussen, S.,Barah, P.,Suarez-
Rodriguez, M.C.,Bressendorff, S.,Friis, P.,
Costantino, P. et al. (2013) Transcriptome responses to combinations of
stresses in Arabidopsis. Plant Physiology,161, 1783–1794.
Rizhsky, L.,Liang, H.,Shuman, J.,Shulaev, V.,Davletova, S. &
Mittler, R.
(2004) When defense pathways collide. The response of Arabidopsis to a
combination of drought and heat stress. Plant Physiology,134(4), 1683–
1696.
Savary, S. &
Willocquet, L. (2020) Modeling the impact of crop diseases on
global food security. Annual Review of Phytopathology,58, 313–341.
Sinha, R.,Irulappan, V.,Patil, B.S.,Reddy, P.C.O.,Ramegowda, V.,Mohan-
Raju, B. et al. (2021) Low soil moisture predisposes field-grown chickpea
plants to dry root rot disease: evidence from simulation modeling and
correlation analysis. Scientific Reports,11,1–12.
Sinha, R.,Zandalinas, S.I.,Fichman, Y.,Sen, S.,Zeng, S.,G´
o
mez-Cadenas,
A. et al. (2022) Differential regulation of flower transpiration during abi-
otic stress in annual plants. New Phytologist,235(2), 611–629.
Smita, S.,Lenka, S.K.,Katiyar, A.,Jaiswal, P.,Preece, J. &
Bansal, K.C. (2011)
QlicRice: a web interface for abiotic stress responsive QTL and loci interac-
tion channels in rice. Database: the Journal of Biological Databases and
Curation,2011,bar037.https://doi.org/10.1093/database/bar037
Suzuki, N.,Rivero, R.M.,Shulaev, V.,Blumwald, E. &
Mittler, R. (2014) Abi-
otic and biotic stress combinations. New Phytologist,203,32–43.
Vemanna, R.S.,Bakade, R.,Bharti, P.,Kumar, M.K.,Sreeman, S.M.,Senthil-
Kumar, M. et al. (2019) Cross-talk signaling in rice during combined
drought and bacterial blight stress. Frontiers in Plant Science.,10, 193.
Zandalinas, S.I.,Balfag ´
o
n, D.,G´
omez-Cadenas, A. &Mittler, R. (2022)
Responses of plants to climate change: metabolic changes during abiotic
stress combination in plants. Journal of Experimental Botany,73, 3339–
3354. Available from: https://doi.org/10.1093/jxb/erac073
Zandalinas, S.I.,Fichman, Y.,Devireddy, A.R.,Sengupta, S.,Azad, R.K. &
Mittler, R. (2020) Systemic signaling during abiotic stress combination in
plants. Proceedings of the National Academy of Sciences USA,117,
13810–13820.
Zandalinas, S.I.,Fritschi, F.B. &
Mittler, R. (2020) Signal transduction net-
works during stress combination. Journal of Experimental Botany,71,
1734–1741.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
20 Piyush Priya et al.
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Zandalinas, S.I.,Fritschi, F.B. &
Mittler, R. (2021) Global warming, climate
change, and environmental pollution: recipe for a multifactorial stress
combination disaster. Trends in Plant Science,26, 588–599.
Zandalinas, S.I. &
Mittler, R. (2022) Plant responses to multifactorial stress
combination. New Phytologist,234, 1161–1167. Available from: https://
doi.org/10.1111/nph.18087
Zandalinas, S.I.,Sengupta, S.,Fritschi, F.B.,Azad, R.K.,Nechushtai, R. &
Mittler, R. (2021) The impact of multifactorial stress combination on
plant growth and survival. New Phytologist,230, 1034–1048.
Zhang, H. &
Sonnewald, U. (2017) Differences and commonalities of
plant responses to single and combined stresses. Plant Journal,90, 839–
855.
Ó2023 Society for Experimental Biology and John Wiley & Sons Ltd.,
The Plant Journal, (2023), doi: 10.1111/tpj.16497
A tool for navigating combined stress responses 21
1365313x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/tpj.16497 by Librarian/Info Scientist Natl Inst Of Plant Genome Res., Wiley Online Library on [12/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License