Preprint:

Please note that this article has not completed peer review.

Computational Identification of miRNAs and

Nann Miky Moh Moh

Peijing Zhang
Zhejiang University Life Science Institute

Yujie Chen
Inner Mongolia University for the Nationalities

Ming Chen
Zhejiang University

[email protected] Author
ORCiD: https://orcid.org/0000-0002-9677-1699

10.21203/rs.3.rs-22698/v1
SUBJECT AREAS
Plant Physiology and Morphology Plant Molecular Biology and Genetics
KEYWORDS
mango (Mangifera indica), miRNA, lncRNA, stress response, target genes,
computational study

1
Abstract
Background
Mango is a major tropical fruit in the world and is known as the king of fruits because of its flavour,

aroma, taste, and nutritional values. Moreover, various parts of mango trees have been used for

medical purposes. Although various regulatory roles of miRNAs and lncRNAs have been investigated

in many plants, there is yet an absence of study in mango. This is the first study to provide

information on ncRNAs of mango with the aim of identifying miRNAs and lncRNAs of mango and

discovering of their potential functions by the interaction prediction of the miRNAs, lncRNAs and their

target genes.

Results
In this analysis, 104 miRNAs and 7,610 temperature responsive lncRNAs were identified and the

target genes of these ncRNAs were characterized. By analysing the interaction of miRNAs and their

target genes, it was observed that miRNAs are mainly involved in growth, development, and stress

responses of mango. For the lncRNAs, cold responsive lncRNAs bound to low temperature responsive

proteins expressed at low temperature stress. GO enrichment analysis of heat and cold responsive

lncRNAs revealed that they involved in all three basic processes; biological process, cellular

component, and molecular function. Moreover, mango lncRNAs can target miRNAs to reduce the

stability of lncRNAs and can function as molecular decoys or sponges of miRNAs.

Conclusion
This paper would provide the new information about miRNAs and lncRNAs of mango and would help

for the further investigation of mango ncRNAs.

Background
Non-coding RNAs (ncRNAs) are RNA molecules that have no or little protein coding potential and are

not translated into proteins although they are transcribed from DNA. Small ncRNAs such as microRNA

(miRNA), small interfering RNA (siRNA), small nucleolar RNA (snoRNA) and piwi-interacting RNA

(piRNA) are shorter than 200 nucleotides (nt) in length and piwi-interacting RNA (lncRNAs) are longer

than 200 nt.

The miRNAs are small (18–24 nt), endogenous and regulatory RNA molecules derived from their long

self-complementary precursor sequences which can fold into hairpin secondary structures [1]. In

2
plants, these long primary precursor miRNAs are transcribed by RNA polymerase II or RNA polymerase

III and then processed by dicer-like 1 enzyme (DCL1) into miRNA/miRNA* duplex [1, 2]. Finally the

mature miRNAs are incorporated into an RNA-induced silencing complex (RISC) [3]. The binding of

miRNAs to their targeted mRNAs in a perfect or nearly perfect complementarity suggests a method

for identifying their targets by BLAST analysis or other related publicly available software [4]. Many

experimental researches have proved that miRNAs involve in many important biological and

metabolic processes. In plants, miRNAs play a fundamental role in almost all biological and metabolic

processes including plant growth, development, signal transduction and various stress response by

binding to their target genes [5].

LncRNAs are a family of regulatory RNAs having a minimal length of 200 nt and unable to encode

proteins. Most lncRNAs are transcribed by RNA polymerase II although some are transcribed by RNA

polymerase III [6–8]. LncRNAs can interact with ncRNAs such as miRNAs [9]. LncRNAs not only can

target to miRNAs to reduce the stability of lncRNAs but also can function as molecular decoys or

sponges of miRNAs [10]. Moreover, lncRNAs can compete with miRNAs to bind to their target mRNAs

and are the precursors for the generation of miRNAs to silence target mRNAs [11]. Many evidence

showed that plant lncRNAs play an important role in fundamental biological processes including the

growth and development and abiotic stress responses [12, 13]. But, the molecular basis of how

lncRNAs function and mediate gene regulation is still poorly understood [14].

The genus Mangifera belongs to the family Anacardiaceae and contains about 69 different species.

Mangifera indica, L (mango) is the most common species among them [15, 16]. Mango is one of the

main tropical fruits over the world and is believed to be originated from Asia [17]. The well-known

countries for mango cultivation are China, India, Thailand, Pakistan, Mexico, Philippine and Myanmar.

The annual production of mango is approximately 42 million tons which is second after banana

production [18]. Mango is called as the king of fruits because of its special characteristic flavour,

pleasant aroma, taste, and nutritional values. Both ripe and raw fruits can be used as the food

products such as pickles, juice, jam, powder, sauce, cereal flakes and so on [19]. Moreover, various

parts of mango trees have been used for medical purposes since long times ago, mostly in Southeast

3
Asian and African countries [20]. In vitro and in vivo studies have been indicated the various

pharmacological potentials of M.indica such as anticancer, anti-inflammatory, antidiabetic,

antioxidant, antifungal, antibacterial, anthelmintic, gastroprotective, hepatoprotective,

immunomodulatory, antiplasmodial and antihyperlipemic effects [21].

Although mango is a popular plant with many important usages, its ncRNA data is still limited. Over

10,000 miRNA data of several plants can be accessed in miRNA database, miRBase, but mango

miRNAs and their functions have not yet been identified. The regulatory roles of lncRNAs and the

molecular basis of lncRNA-mediated gene regulation are also still poorly understood in plants

including mango. So, the aim of this research work is to identify and study about the miRNAs and

lncRNAs of mango and to examine their potential functions by the interaction prediction of the

miRNAs, lncRNAs and their target genes.

Results
Identification and characterization of mango miRNAs

From the mango unigene sequences, we have identified 104 miRNAs by following the identification

workflow explained in Figure 1. The length of the resulting mature miRNAs is in the range of 18-22 nt.

Among them, nearly 40% (41 miRNAs) of mango mature miRNAs are in the length of 18 nt and 6

miRNAs have the length of 22 nt. 32 miRNAs, 17 miRNAs and 8miRNAs are 19 nt, 20 nt and 21 nt of

length respectively (Figure 2A). However, the precursor length of mango miRNAs (MmiRs) was varied

significantly from 67 to 144 nt with an average length of 94 nt. The secondary structure of precursor

sequences was predicted by Zuker folding algorithm in MFOLD. The hairpin structures of five miRNAs

are shown in Figure 2B. Average MFE of pre-miRNAs is 29.92. The MFEI values were also calculated

and were in the range of 0.7 to 1.45 with the average MFEI of 0.84 (Table S1).

Target genes analysis of miRNAs

According to the result of target gene prediction by psRNATarget server, all the newly identified

mango miRNAs could bind to their targets and a total of 2,347 target genes were predicted for 104

mango miRNAs. The predicted target genes were annotated and assigned to GO term by BLAST2GO.

According to the result of GO analysis, the predicted target genes of mango miRNAs involved in all

4
three broad categories; biological processes, cellular components and molecular functions. Among

2347 target genes, 2081 target genes were enriched in 925 GO terms, and 502 in biological process,

127 in cellular component and 296 in molecular function. Highly enriched GO terms of miRNA target

genes were visualized in Figure 2C. From KEGG pathway analysis, 310 targets were involved in the

103 different KEGG pathways. Purine metabolism was the pathway with the highest 136 target genes.

The predicted miRNAs, their target genes, target descriptions, target GO terms and target KEGG

pathways are shown in Table S2.

Identification and characterization of mango lncRNAs

For the identification of lncRNAs, a total of 277,071 RNA transcripts from Zill, Shelly and Keitt mango

cultivars were used. First, the sequences less than 200 nt were removed because lncRNAs were

always longer than 200 nt. Then, the coding transcripts were removed by their protein-coding

potential, homology with known proteins, and potential ORFs. Finally, the house keeping RNAs and

precursor of miRNAs were removed. After a series of filtering steps, a total of 31,226 candidate

lncRNAs were predicted.

The temperature responsive lncRNAs were then defined by fold change value and FDR. Fold change

value of <-2 or >2 and FDR adjusted p-value 0.05 were used to filter out the significantly expressed

mango lncRNAs, and as the result, 24 lncRNAs were significantly expressed to heat stress (55°C hot

water brushing) and 7586 lncRNAs to cold stress (5°C, 8°C or 12°C) (Figure 3A). In heat responsive

lncRNAs, 18 lncRNAs were upregulated and 6 lncRNAs were downregulated. The length of heat

responsive lncRNAs was ranging from 213 to 1186 nt (Table S3). Among the 7619 cold responsive

lncRNAs, 4335 were upregulated and 3251 were downregulated. The length of cold responsive

lncRNAs was in the range of 201 to 2746 nt (Table S4).

Conservation analysis of lncRNAs

The mango lncRNAs were searched by using BLASTn against the plant lncRNA database, CANTATAdb,

with e-value cutoff 1e-20 to check their evolutionary conservation. As the result, no heat responsive

lncRNAs was conserved and 22 cold responsive lncRNAs were conserved with 12 different plant

species (Figure 3B).

5
Target gene prediction of lncRNAs

To analyse the interaction of newly identified lncRNAs of mango with protein-coding genes, the

lncRNA target prediction tool LncTar was used. A total of 1998 mango mRNAs downloaded from NCBI

were used for the target prediction of lncRNAs. From the resulting data, 6975 lncRNAs interacted with

1985 target mRNAs. To analyse the functional overview of identified lncRNAs, the targets of the

identified lncRNAs were predicted by BLAST2GO. Among 24 heat responsive lncRNAs, 8 lncRNAs had

115 target genes (SCL14, STP13, Hsp70, At4g39970, ACO1 and so on) involved in the plant

development and stress response. In cold responsive lncRNAs, 6951 lncRNAs interacted with 1985

target genes. The WRKY proteins are a large family of transcriptional regulators in higher plant and 64

cold responsive lncRNAs interact with WRKY gene family in this study (Figure 3C).

Moreover, functional prediction of the target genes of identified lncRNAs were performed by GO

enrichment analysis and the resulting data showed that 11 GO terms were enriched in biological

process, 8 GO terms in cellular component, and 3 GO terms in molecular function for heat responsive

lncRNAs. In the biological process, metabolic process, cellular process, cellular component biogenesis

and cellular metabolic process were highly enriched. In the cellular component analysis, GO terms

associated with membranes and intracellular were highly enriched. Catalytic activity and binding GO

terms were highly enriched in molecular function analysis (Figure 3D). For the target genes of the

cold responsive lncRNAs, 40 GO terms were highly enriched; GO terms of 20 were enriched in the

biological process, 7 in cellular component and 14 in molecular functions. Metabolic processes and

cellular processes were highly enriched for biological processes. In the cellular component analysis,

GOs related to membranes, intracellular and cytoplasm were highly enriched. For molecular function

analysis, most of the enriched GO terms were related to catalytic activity and binding (Figure 3E).

From the results of the KEGG pathway analysis, heat responsive lncRNAs had target genes involved in

17 KEGG pathways (Table S5). Among these different pathways, amino sugar and nucleotide sugar

metabolism was the most significant pathway and 8 target genes involved in this pathway. For cold

responsive lncRNAs, 209 target genes had mapped to 87 KEGG pathways (Table S6). JK513026_1,

alcohol dehydrogenase 1 (ADH1, EC:1.1.1.1) was the most enriched target gene and involved in 12

6
different pathways.

Prediction of lncRNAs as miRNAs targets

To analyse the direct interaction of miRNAs and lncRNAs of mango, the psRNATarget server was used

to predict the target lncRNAs of miRNAs. The resulting data showed that 3 heat responsive lncRNAs

interacted with 6 miRNAs (Table S7). For cold responsive lncRNAs, 763 lncRNAs had 1203 pairs of

interactions with 89 miRNAs (Table S8).

The miRNA target mimicry search was also performed by using TAPIR. No heat responsive lncRNA act

as the target mimic of miRNAs. But 20 cold responsive lncRNAs were predicted as the target mimics

of 20 miRNAs (Table S9). CRlnc31221 was the target mimic of MmiR5408 which targeted to 8 cold

responsive lncRNAs and 47 target genes (Figure 4B). Base-pairing interaction between MmiR5408 and

its target mimic cold responsive lncRNA, CRlnc31221 was shown in Figure 4C.

The interaction network of mango ncRNAs (miRNAs, lncRNAs and mimic) and their target genes was

visualized by using Cytoscape contained a total of 5388 pairs of interaction among miRNAs, lncRNAs

and their targets (Figure 4A). These interactions were 4155 pairs of 104 MmiRNAs and 2347 mRNAs,

1203 pairs of 89 MmiRNAs and 763 CRlncRNAs, 6 pairs of 6 MmiRNAs and 4 HRlncRNAs, and 24 pairs

of 20 miRNAs and their 20 target mimics.

Discussion
Identification, characterization and target gene prediction of miRNAs

Most of the plant miRNAs are evolutionarily conserved from species to species [22, 23] and this

indicates the powerful strategy for the identification of new miRNAs by using the already known

miRNAs [24]. Many conserved miRNAs have been identified from the expressed sequence tag (EST)

[25, 26] and genome survey sequence (GSS) [27] by using this homology search approach. For

mango, there is no GSS data and the available EST data for mango is only 1709 and it was not

sufficient for identification of miRNA. Hence, unigene sequences (107,744) were used for the

identification of miRNAs in this study. Unigene is a unique transcript that is transcribed from a

genome and many miRNAs have been identified from the unigenes of many plant species such as

Artemisia annua [28], coconut [29], Litchi fruit [30] and black pepper [31].

7
The potential 104 pre-miRNAs of mango were predicted based on the parameters of Zhang[25] and

the MFEI values were also calculated as the MFEI gave the best prediction of miRNAs [32]. Although

the length of the predicted mature miRNAs was in the range of 18-22 nt, the length of precursor

miRNAs varied significantly from 67 to144 nt with an average length of 94 nt. The predicted 104

mango miRNAs belong to 86 different families. Among them, over 70% of the miRNA families have

only one family member. The highest 5 family members were found in the miR2673 family followed

by miR159, with 4 family members. The remaining miRNAs have the family member of 2 or 3.

Therefore, we can see that the mango miRNA distribution across various families is highly

heterogeneous.

The previous studies have already proved that the plant miRNAs bind to their targets in a perfect or

nearly perfect complementarity and thus the psRNATarget server was used to search the target gene

of mango miRNAs in this study. Both mRNAs collected from NCBI and mRNA identified in this study

were used as the target candidates of miRNAs due to the absence of Mangifera indica target

candidates in psRNATarget server. Some previous studies indicated that miR156 was a master

regulator of the juvenile phase in plants and it targeted the squamosa promoter binding protein-Like

(SPL) gene family to regulate the transition from vegetative phase to floral phase in Arabidopsis,

maize and rice [33-39]. In mango, MmiR105772, a family of miR156, also bound to its target SPL6

and thus the predicted targets of mango miRNAs were in the agreement with the previously published

papers in other plants. The resulting data from psRNATarget also showed that only one miRNA

(MmiR1653) had the single target gene which was the member of miR482 family and bound to

monodehydroascorbate reductase 4 enzyme, the important gene related to the nutritional quality of

mango fruit [33]. All other miRNAs could target to multiple genes and some miRNAs had over two

hundred target genes. For example, MmiR73030 had 230 target genes and these target genes

involved in 16 KEGG pathways such as biosynthesis of antibiotics, purine metabolism, sulfur

metabolism, glycerophospholipid metabolism, T cell receptor signalling pathway, steroid degradation

and so on.

Sivankalyani, Sela et al. published that the mango stress-response pathways were activated by cyclic

8
nucleotide-gated channel (CNGC) and leucine-rich repeat receptor (Lrr) [40]. In this study, we found

that MmiR90392 targeted to CNGC1, and MmiR68471 and MmiR68478 targeted to Lrr2. Moreover,

MmiR10167 and MmiR15558 bound to the stress WRKY transcription factor 44 which play a major role

in plant defence to biotic and abiotic stresses. MmiR78769 and MmiR101928 also bound to their

target genes of phospholipase A and phospholipase D which were key factors in plant responses to

biotic and abiotic stresses [41]. The ethylene response could improve the tolerance of mango fruit to

chilling stress [42] and ten mango miRNAs identified in this study had six ethylene responsive target

genes such as ethylene-insensitive protein and ethylene-responsive transcription factor. So, these

newly identified mango miRNAs have potential roles in chilling stress responsive process of mango.

Two mango miRNAs (MmiR23777 and MmiR36814) also targeted to the auxin efflux carrier which had

the potential role in mango plant organ development [43]. A total of 17 miRNAs interacted with auxin-

related genes. MmiR51876 was a miRNA that targeted to auxin responsive protein. The pentose and

glucoronate interconversions pathway, phenlypropanoid biosynthesis pathway and alpha-linolenic

acid metabolism pathway were KEGG pathway involved in the adventitious root formation of mango

cotyledon segments [44]. In this study, 9 miRNAs bound to 8 target genes involved in these three

pathways for mango root formation. MmiR10167 bound to target genes that involved in

phenlypropanoid biosynthesis pathway and MmiR7519 bound to target genes involved in alpha-

linolenic acid metabolism pathway. From these findings, it could be observed that these five mango

miRNAs (MmiR23777, MmiR36814, MmiR51876, MmiR10167 and MmiR7519) involved in the

developmental process of mango (Figure 2B).

Identification, characterization and target gene prediction of lncRNAs

As the genome sequence of mango is not available till now, the de novo assembled transcriptome

sequences were used for the identification of lncRNAs in this study. A total of 277,071 RNA transcripts

from Zill, Shelly and Keitt mango cultivars studied by the former researchers were used and a total of

31,226 candidate lncRNAs were predicted in this study. Among them, 24 lncRNAs were significantly

expressed to heat stress and 7586 lncRNAs to cold stress. The most significantly expressed down-

regulated heat responsive lncRNA was HRlnc25944 with the fold change value of -6.22. HRlnc11351

9
and HRlnc27371 were the mostly expressed up-regulated lncRNAs with fold change value greater

than 7. For the cold responsive lncRNAs, CRlnc10871 was the mostly expressed down-regulated

lncRNA (FC value -11.19), and CRlnc26299, CRlnc30496 and CRlnc36473 were the most significantly

expressed lncRNAs with fold change value greater than 11.

No heat responsive lncRNAs was conserved but 0.29% of cold responsive lncRNAs were conserved

with 12 different plant species. Among them, the highest conserved lncRNAs were CRlnc32663 and

CRlnc47883. Each of which was conserved with 4 different lncRNAs of other plants. CRlnc32663

conserved with 4 different lncRNAs of 3 three different plant species such as Manihot esculenta, Malus

domestica and Populus trichocarpa. CRlnc42883 also conserved with 4 lncRNAs of Oryza rufipogon,

Oryza barthii and Solanum lycopersicum (Figure 3B).

For heat responsive lncRNAs, 8 bound to 115 target genes involved in the plant development and

stress response. HRlnc11351 was the most significantly expressed up-regulated lncRNAs with fold

change value of 7.55 and bound to six heat shock proteins. In cold responsive lncRNAs, CRlnc26299

was one of the most significantly expressed up-regulated lncRNAs and bound to RC12B (JK513200_1)

which is the low temperature and salt responsive protein found in Arabidopsis thaliana [45]. The

WRKY proteins are a large family of transcriptional regulators in higher plant and are exhibited the

variable expression patterns in response to chilling stress in cucumber, mango and rice [40, 46, 47].

In this study, 64 cold responsive lncRNAs have interaction with WRKY gene family. So, we can observe

that the cold responsive lncRNAs of mango have the interaction with the target genes that are

expressed at the low temperature stress.

GO enrichment analysis and KEGG pathway analysis were performed for the better understanding of

the target genes of newly identified lncRNAs. From GO enrichment analysis result, we could see that

both types of heat responsive lncRNAs and cold responsive lncRNAs had interaction with the target

genes involved in all three broad categories such as biological process, cellular component and

molecular function. For the biological processes, both heat responsive and cold responsive lncRNAs

were highly enriched in metabolic processes and cellular processes. In the cellular component

analysis, GOs related to membranes, intracellular and cytoplasm were highly enriched for both type of

10
lncRNAs. Also for molecular function analysis, most of the enriched GO terms in both type of lncRNAs

were related to catalytic activity and binding. Therefore, we could see that, the GO terms highly

enriched in both heat responsive and cold responsive lncRNAs were not quite different.

Among 17 KEGG pathways of the target genes of the heat responsive lncRNAs, amino sugar and

nucleotide sugar metabolism was the most significant pathway and 8 target genes involved in this

pathway. As mentioned above, HRlnc11351 was the most significantly expressed up-regulated

lncRNAs and its target gene, JK513625_1 is 3-ketoacyl-CoA thiolase 2 (KAT2, EC:2.3.1.16) which could

be mapped to 9 different pathways such as benzoate degradation, fatty acid elongation, biosynthesis

of unsaturated fatty acids, alpha-linolenic acid metabolism, fatty acid degradation, valine, leucine and

isoleucine degradation, biosynthesis of antibiotics, geraniol degradation and ethylbenzene

degradation according to the result of KEGG pathway analysis. In Arabidopsis, KAT2 is an enzyme that

catalyses the β-oxidation of fatty acid and involves in abscisic acid (ABA) signal transduction [48]. The

phytohormone ABA plays an important role in plant development and adaptation to diverse

environmental stresses. Therefore, HRlnc11351 may involve in the important role of mango

development and stress response by targeting to KAT2. For cold responsive lncRNAs, 209 target

genes had mapped to 86 KEGG pathways. JK513026_1, alcohol dehydrogenase 1 (ADH1, EC:1.1.1.1)

was the most enriched target gene and involved in 12 different pathways including

glycolysis/gluconeogenesis, metabolism of xenobiotics by cytochrome P450, glycine, serine and

threonine metabolism, methane metabolism, fatty acid degradation and so on. In plants, ADH genes

are involved in mediating stress responses and developments. In mango, ADH1 has important role in

the fruit ripening [49] and thus, cold responsive lncRNAs that target to ADH1 gene may play

important role in mango fruit ripening process. According to the KEGG pathway analysis results,

purine metabolism and biosynthesis of antibiotics were the highly enriched pathways among 86

pathways and more than 50 target genes were enriched in each pathway.

Interaction between lncRNAs and miRNAs

The interaction between the miRNAs and lncRNAs showed that the most of the miRNAs had targeted

to more than one lncRNAs and only 8 miRNAs had single target lncRNAs. The number of lncRNAs

11
targeted by a single miRNA was in the range of 1 to 90. A total of 90 target lncRNAs were found for

MmiR73030 which also targeted 230 mRNAs. This miRNA had the highest target numbers in both

lncRNAs and mRNAs.

LncRNAs not only can target to miRNAs to reduce the stability of lncRNAs but also can function as

molecular decoys or sponges of miRNAs [[10, 50]]. So the miRNA target mimicry search was

performed by using TAPIR, which is a web server for the prediction of plant miRNA targets including

target mimics. Although no heat responsive lncRNA act as the target mimic of miRNAs, 20 cold

responsive lncRNAs were predicted as the target mimics of 20 miRNAs. CRlnc31221 was the target

mimic of MmiR5408 which targeted to 8 cold responsive lncRNAs and 47 target genes. These target

genes were involved in starch and sucrose metabolism, inositol phosphate metabolism and

phenylpropanoid biosynthesis pathways which pathways are important for plant growth and

development, and plant response towards biotic and abiotic stress. During target mimicry, the

interactions between miRNAs and their authentic targets were blocked by binding of decoy RNA to

miRNAs via partially complementary sequences [51]. So, the target mimicry of CRlnc31221 had the

potential regulation effect to the interaction between the target genes and MmiR5408.

Conclusion
In conclusion, this study identified the 104 miRNAs and 7610 temperature responsive lncRNAs from

mango transcriptome sequences. And the interactions of these ncRNAs with their target genes were

also predicted. According to the result, newly identified mango ncRNAs, like other plant ncRNAs, have

potential role in biological and metabolic pathways including plant growth and developmental

process, pathogen defence mechanism, and stress responsive process. Therefore, the resulting data

of this project may help for the further prediction of the specific functions of mango ncRNAs and wet

lab experiments.

Methods
Data collection

A total of 10,415 plant miRNAs (release 21) were downloaded from miRBase database

(http://www.mirbase.org/cgi-bin/browse.pl) and the redundancy sequences were removed. The

12
resulting 6042 non-redundant known miRNAs were used as the reference for the prediction of

conserved miRNAs.

107,744 mango unigenes collected from the mango RNA-Seq database

(http://bioinfo.bti.cornell.edu/cgi-bin/mango/index.cgi) were used for the identification of mango

miRNAs.

As the whole genome of mango is not available, the publically available de novo assembled

transcripts were used for the prediction of candidate lncRNAs. A total of 277,071 mango transcripts

obtained from Zill, Shelly and Keitt mango cultivars [40, 52-54] were used in this study. A total of

1998 mango mRNAs downloaded from NCBI were used for the target prediction of miRNAs and

lncRNAs.

Identification of miRNAs and their precursors

First, the homology search of mango unigenes against non-redundant plant miRNAs was performed by

using BLASTn with default parameters. The following criteria were used to choose the candidate

miRNAs; the length of candidate miRNA should be greater than or equal to 18 nt without gap and the

number of mismatches between mango sequences and plant miRNAs should not be more than 2. The

sequences of 100 nt upstream and 100 nt downstream from the BLAST hit were extracted for

precursor sequences. If the length of query sequence was less than 200 nt, the entire sequence was

selected. BLASTx against NCBI non-redundant (Nr) protein databases was used to remove the protein-

coding sequences from the extracted precursor sequences with the e-value cut off 0.01. The

secondary structures of remaining precursor sequences were predicted by using the Zuker folding

algorithm in MFOLD software [55] with default parameters. The workflow for the identification of

miRNAs was briefly described in Figure 1. Based on the parameters of Zhang [25], the potential pre-

miRNAs were predicted as follows:

1. the minimum length of precursor should be at least 60 nt;

2. the pre-miRNA sequence should be folded into an appropriate stem-loop hairpin

secondary structure;

13
3. it should contain the mature miRNA within one arm of the hairpin;

4. the predicted mature miRNAs and its opposite miRNA* sequence in the other arm

should not have more than 6 nt mismatches;

5. loop or break should not be contained between miRNA/miRNA* duplex;

6. maximum size of a bulge in the mature miRNA sequences should not be more than 3

nt;

7. the predicted secondary structures should have higher negative minimal free

energies (MFE) and minimal free energy index (MFEI);

8. MFEI of pre-miRNA should be greater than 0.7;

9. A+U content should be within 30-70%.

The following equations were used to calculate the minimal free energy index (MFEI) and adjusted

minimal free energy (AMFE).

AMFE = (MFE/length of pre-miRNA) x 100

MFEI = AMFE/(G+C)%

Prediction of candidate lncRNAs

To predict the lncRNAs, the transcripts smaller than 200 nt were firstly removed. The coding potential

of remaining transcripts was then calculated by CPC [56] and LncFinder [57]. Only sequences with the

CPC score less than -1 and LncFinder score less than 0.5 were used for further prediction. The protein

coding sequences were removed by BLASTx against NCBI Nr protein databases and ORFfinder

(https://www.ncbi.nlm.nih.gov/orffinder/) was used to predict the open reading frame (ORF) of the

remaining sequences and the minimal ORF cutoff less than 102 amino acids was applied for

prediction. Then, house-keeping genes were removed against Rfam database (http://rfam.xfam.org)

with e-value 0.001. Finally, to remove the lncRNAs acting as precursors of known or novel miRNAs,

lncRNAs were aligned with precursors of known non-redundant plant miRNAs from the miRBase

database (http://www.mirbase.org/) using BLASTn with the default parameters (Figure 1).

The remaining transcriptome sequences that were not captured as lncRNAs were used as queries

14
against the NCBI Nr protein database using BLASTx with a cutoff e-value of 1e-5. The sequences with

the blast hits were then analysed to remove the house-keeping RNAs. The final sequences were

identified as protein coding sequences in this study for target gene analysis.

Identification of significantly expressed temperature responsive lncRNAs

From the resulting lncRNA transcripts, the temperature responsive lncRNAs were filtered by two

parameters. The mango lncRNAs with the adjusted p-value of 0.05 and the log2 fold change of greater

than 2 or less than -2 were identified as the significantly expressed lncRNAs.

Target gene prediction of miRNAs and lncRNAs

Mango mRNAs downloaded from NCBI database and mango protein coding sequences previously

identified were used for the target genes prediction of miRNAs. The putative target sites of miRNAs

were identified by aligning the miRNA sequences using plant target prediction tool, psRNATarget

server (http:// plantgrn.noble.org/psRNATarget/) [58]. To reduce the number of false predictions, the

maximum expectation threshold was set to the value of 3.0. The cut-off length of nucleotides for

complementarity scoring, hsp size, was set as the length of the mature miRNAs. The maximum

energy of unpairing (UPE) the target site was set as 25 kcal. The flanking length around target site for

target accessibility analysis was 17 bp in upstream and 13 bp in downstream. The range of central

mismatch leading to translation inhibition was adjusted as 9-11 nt. No gap and no more than four

mismatches between miRNA and its target (G-U pair count as 0.5 mismatch) was allowed. The target

genes of mango lncRNAs were predicted by using LncTar tool [59] with the normalized binding free

energy (ndG) cutoff value less than 0.1.

Prediction of lncRNAs as miRNA target or target mimic

To predict the lncRNAs as the target genes of miRNAs, psRNATarget was used as previously

mentioned in the interaction prediction of miRNAs and mRNAs. For target mimic prediction, TAPIR

server [60] was used in this study. TAPIR is a web server for the prediction of plant miRNA targets

including target mimics,

Functional annotation and pathway analysis of target genes

The gene ontology (GO) analysis of the identified target transcripts was executed by combining both

15
BLASTx data and interproscan analysis data by means of the BLAST2go software [61]. The GO

enrichment analysis was performed by using Fisher’s exact test with multiple testing correction of

false discovery rate (FDR). KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis was

also performed for better understanding of the functions of the target genes.

Conservation analysis of lncRNAs

The analysis of conservation of mango lncRNAs was detected by using BLASTn against all lncRNA

sequences from the plant lncRNA database, CANTATAdb [62], with e-value cutoff 1e-20.

Interaction network of miRNAs, lncRNAs and their target genes

Finally, the interaction network of miRNAs, lncRNAs and their target genes were visualized by using

Cytoscape [63].

Abbreviations
ABA

abscisic acid

ACO1

1-aminocyclopropane-1-carboxylate oxidase 1

ADH1

alcohol dehydrogenase 1

AMFE

adjusted minimal free energy

BLAST

basic local alignment search tool

bp

base pair

cytosine

CNGC

cyclic nucleotide-gated channel

CPC

coding potential calculator

DCL1

dicer-like 1 enzyme

EST

16
expressed sequence tag

FDR

false discovery rate

guanine

GO

gene ontology

GSS

genome survey sequence

Hsp70

heat shock protein 70

KAT2

3-ketoacyl-CoA thiolase 2

Kcal

kitocalorie

KEGG

Kyoto Encyclopedia of Genes and Genomes

lncRNA

long non-coding RNA

Lrr

leucine-rich repeat receptor

MFE

minimal free energies

MFEI

minimal free energy index

miRNA

microRNA

mRNA

messenger RNA

NCBI

national center for biotechnology information

ncRNA

non-coding RNA

ndG

17
normalized binding free energy

Nr

non-redundant

nt

nucleotide

ORF

open reading frame

piRNA

piwi-interacting RNA

RC12B

related cDNA 12 B

RISC

RNA-induced silencing complex

RNA

ribonucleic acid

SCL14

scarecrow-like 14

siRNA

small interfering

snoRNA

small nucleolar RNA

SPL

squamosa promoter binding protein-Like

STP13

sugar transport protein 13

UPE

maximum energy of unpairing

Declarations
Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and materials

18
The unigene sequence data are available from the mango RNA-Seq database

(http://bioinfo.bti.cornell.edu/cgi-bin/mango/index.cgi) and mango transcriptome data are available

from the supplementary data of the published journals [40, 52-54]. All data generated and analysed

during this study are included in this published article (and its supplementary information files).

Competing interests

The authors have declared no competing interests.

Funding

This research was supported by the Talented Young Scientist Program organized by the China Ministry

of Science and Technology. Ming Chen’s Lab are grateful to the supports from MOST

(2018YFC0310600, 2016YFA0501704), NSFC (31771477, 31571366), and JCIC-MCP/CIC-MCP.

Authors’ contributions

NMMM and MC designed the research. NMMM and PZ performed bioinformatics analysis.

NMMM and YC analyzed plant genes data. All authors approved the final manuscript.

Acknowledgements

The author thanks all lab members for their suggestions during this research work.

Author information

Affiliations

Biotechnology Research Department, Ministry of Education, Kyaukse, Myanmar

Nann Miky Moh Moh

Department of Bioinformatics, Key State Laboratory of Plant Physiology and Biochemistry, College of

Life Sciences, Zhejiang University, Hangzhou 310058, PR China

Nann Miky Moh Moh, Peijing Zhang, Ming Chen

Life Sciences and Food College, Inner Mongolia University for the Nationalities, Tongliao, Inner

Mongolia, PR China

Yujie Chen, Ming Chen

Authors’ information

Not applicable.

19
Corresponding authors

Correspondence to Ming Chen.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and

institutional affiliations.

References
1. Kurihara Y, Watanabe Y: Arabidopsis micro-RNA biogenesis through Dicer-like 1

protein functions. Proceedings of the National Academy of Sciences 2004,

101(34):12753–12758.

2. Panda D, Dehury B, Sahu J, Barooah M, Sen P, Modi MK. Computational identification

and characterization of conserved miRNAs and their target genes in garlic (Allium

sativum L.) expressed sequence tags. Gene. 2014;537(2):333–42.

3. Bartel DP: MicroRNAs: genomics, biogenesis, mechanism, and function. cell 2004,

116(2):281–297.

4. Zhang B, Pan X, Cobb GP, Anderson TA. Plant microRNA: a small regulatory molecule

with big impact. Developmental biology. 2006;289(1):3–16.

5. Rhoades MW, Reinhart BJ, Lim LP, Burge CB, Bartel B, Bartel DP: Prediction of plant

microRNA targets. cell 2002, 110(4):513–520.

6. Dieci G, Fiorino G, Castelnuovo M, Teichmann M, Pagano A. The expanding RNA

polymerase III transcriptome. TRENDS in Genetics. 2007;23(12):614–22.

7. Geisler S, Coller J. RNA in unexpected places: long non-coding RNA functions in

diverse cellular contexts. Nature reviews Molecular cell biology. 2013;14(11):699–

712.

8. Zhang Y-C, Chen Y-Q. Long noncoding RNAs: new regulators in plant development.

Biochem Biophys Res Commun. 2013;436(2):111–4.

20
 9. Jalali S, Bhartiya D, Lalwani MK, Sivasubbu S, Scaria V. Systematic transcriptome

wide analysis of lncRNA-miRNA interactions. PloS one 2013, 8(2).

10. Salmena L, Poliseno L, Tay Y, Kats L, Pandolfi PP. A ceRNA hypothesis: the Rosetta

Stone of a hidden RNA language? Cell. 2011;146(3):353–8.

11. Yoon J-H, Abdelmohsen K, Gorospe M. Functional interactions among microRNAs and

long noncoding RNAs. In: Seminars in cell & developmental biology: 2014. Elsevier:

9–14.

12. Xin M, Wang Y, Yao Y, Song N, Hu Z, Qin D, Xie C, Peng H, Ni Z, Sun Q. Identification

and characterization of wheat long non-protein coding RNAs responsive to powdery

mildew infection and heat stress by using microarray analysis and SBS sequencing.

BMC plant biology. 2011;11(1):61.

13. Zhang J, Mujahid H, Hou Y, Nallamilli BR, Peng Z. Plant long ncRNAs: a new frontier

for gene regulatory control. 2013.

14. Megha S, Basu U, Rahman MH, Kav NN: The role of long non-coding RNAs in abiotic

stress tolerance in plants. In: Elucidation of Abiotic Stress Signaling in Plants.

Springer; 2015: 93–106.

15. Mukherjee S. Origin of mango (Mangifera indica). Econ Bot. 1972;26(3):260–4.

16. Slippers B, Johnson GI, Crous PW, Coutinho TA, Wingfield BD, Wingfield MJ.

Phylogenetic and morphological re-evaluation of the Botryosphaeria species causing

diseases of Mangifera indica. Mycologia. 2005;97(1):99–110.

17. Hirano R, Htun Oo T, Watanabe K. Myanmar mango landraces reveal genetic

uniqueness over common cultivars from Florida, India, and Southeast Asia. Genome.

2010;53(4):321–30.

18. Galán Saúco V. Worldwide mango production and market: current situation and future

prospects. In: IX International Mango Symposium 992: 2010. 37–48.

21
19. Siddiq M, Akhtar S, Siddiq R. Mango Pocessing, Products and Nutrition. Tropical and

Subtripical Fruits: Posthrvest and Physiology, Processing and Packaging 2012.

20. Mukherjee S. The mango—Its botany, cultivation, uses and future improvement,

especially as observed in India. Econ Bot. 1953;7(2):130–62.

21. Lauricella M, Emanuele S, Calvaruso G, Giuliano M, D’Anneo A. Multifaceted health

benefits of Mangifera indica L.(Mango): the inestimable value of orchards recently

planted in Sicilian rural areas. Nutrients. 2017;9(5):525.

22. Dezulian T, Palatnik JF, Huson D, Weigel D. Conservation and divergence of microRNA

families in plants. Genome Biol. 2005;6(11):P13.

23. Weber MJ. New human and mouse microRNA genes found by homology search. FEBS J.

2005;272(1):59–73.

24. Zhang B, Pan X, Cannon CH, Cobb GP, Anderson TA. Conservation and divergence of

plant microRNA genes. Plant J. 2006;46(2):243–59.

25. Zhang BH, Pan XP, Wang QL, George PC, Anderson TA. Identification and

characterization of new plant microRNAs using EST analysis. Cell research.

2005;15(5):336–60.

26. Frazier TP, Zhang B: Identification of plant microRNAs using expressed sequence tag

analysis. In: Plant Reverse Genetics. Springer; 2011: 13–25.

27. Pan X, Zhang B, Francisco MS, Cobb GP. Characterizing viral microRNAs and its

application on identifying new microRNAs in viruses. Journal of cellular physiology.

2007;211(1):10–8.

28. Pérez-Quintero ÁL, Sablok G, Tatarinova TV, Conesa A, Kuo J, López C. Mining of

miRNAs and potential targets from gene oriented clusters of transcripts sequences of

the anti-malarial plant, Artemisia annua. Biotechnol Lett. 2012;34(4):737–45.

29. Naganeeswaran S, Fayas T, Rachana K, Rajesh M. Computational prediction and

22
 characterization of miRNA from coconut leaf transcriptome. Journal of Applied

Horticulture. 2015;17(1):12–7.

30. Yao F, Zhu H, Yi C, Qu H, Jiang Y. MicroRNAs and targets in senescent litchi fruit

during ambient storage and post-cold storage shelf life. BMC plant biology.

2015;15(1):181.

31. Asha S, Sreekumar S, Soniya E. Unravelling the complexity of microRNA-mediated

gene regulation in black pepper (Piper nigrum L.) using high-throughput small RNA

profiling. Plant cell reports. 2016;35(1):53–63.

32. Zhang B, Pan X, Cox S, Cobb G, Anderson T. Evidence that miRNAs are different from

other RNAs. Cellular Molecular Life Sciences CMLS. 2006;63(2):246–54.

33. Pandit SS, Kulkarni RS, Giri AP, Köllner TG, Degenhardt J, Gershenzon J, Gupta VS.

Expression profiling of various genes during the fruit development and ripening of

mango. Plant Physiol Biochem. 2010;48(6):426–33.

34. Gandikota M, Birkenbihl RP, Höhmann S, Cardon GH, Saedler H, Huijser P. The

miRNA156/157 recognition element in the 3′ UTR of the Arabidopsis SBP box gene

SPL3 prevents early flowering by translational inhibition in seedlings. Plant J.

2007;49(4):683–93.

35. Wu G, Park MY, Conway SR, Wang J-W, Weigel D, Poethig RS. The sequential action of

miR156 and miR172 regulates developmental timing in Arabidopsis. Cell.

2009;138(4):750–9.

36. Yamaguchi A, Wu M-F, Yang L, Wu G, Poethig RS, Wagner D. The microRNA-regulated

SBP-Box transcription factor SPL3 is a direct upstream activator of LEAFY, FRUITFULL,

and APETALA1. Developmental cell. 2009;17(2):268–78.

37. Chuck G, Cigan AM, Saeteurn K, Hake S. The heterochronic maize mutant Corngrass1

results from overexpression of a tandem microRNA. Nat Genet. 2007;39(4):544–9.

23
38. Jiao Y, Wang Y, Xue D, Wang J, Yan M, Liu G, Dong G, Zeng D, Lu Z, Zhu X. Regulation

of OsSPL14 by OsmiR156 defines ideal plant architecture in rice. Nat Genet.

2010;42(6):541.

39. Jeong D-H, Park S, Zhai J, Gurazada SGR, De Paoli E, Meyers BC, Green PJ. Massive

analysis of rice small RNAs: mechanistic implications of regulated microRNAs and

variants for differential target RNA cleavage. Plant Cell. 2011;23(12):4185–207.

40. Sivankalyani V, Sela N, Feygenberg O, Zemach H, Maurer D, Alkan N. Transcriptome

dynamics in mango fruit peel reveals mechanisms of chilling stress. Frontiers in plant

science. 2016;7:1579.

41. Xue H, Chen X, Li G. Involvement of phospholipid signaling in plant growth and

hormone effects. Curr Opin Plant Biol. 2007;10(5):483–9.

42. Lederman IE, Zauberman G, Weksler A, Rot I, Fuchs Y. Ethylene-forming capacity

during cold storage and chilling injury development in ‘Keitt’mango fruit. Postharvest

Biol Technol. 1997;10(1):107–12.

43. Li Y-H, Zou M-H, Feng B-H, Huang X, Zhang Z, Sun G-M. Molecular cloning and

characterization of the genes encoding an auxin efflux carrier and the auxin influx

carriers associated with the adventitious root formation in mango (Mangifera indica

L.) cotyledon segments. Plant Physiol Biochem. 2012;55:33–42.

44. Li Y-H, Zhang H-N, Wu Q-S, Muday GK. Transcriptional sequencing and analysis of

major genes involved in the adventitious root formation of mango cotyledon

segments. Planta. 2017;245(6):1193–213.

45. Medina Jn, Catalá R, Salinas J. Developmental and stress regulation of RCI2A

andRCI2B, two cold-inducible genes of Arabidopsis encoding highly conserved

hydrophobic proteins. Plant physiology. 2001;125(4):1655–66.

46. Ramamoorthy R, Jiang S-Y, Kumar N, Venkatesh PN, Ramachandran S. A

24
 comprehensive transcriptional profiling of the WRKY gene family in rice under various

abiotic and phytohormone treatments. Plant cell physiology. 2008;49(6):865–79.

47. Ling J, Jiang W, Zhang Y, Yu H, Mao Z, Gu X, Huang S, Xie B. Genome-wide analysis of

WRKY gene family in Cucumis sativus. BMC Genomics. 2011;12(1):471.

48. Jiang T, Zhang X-F, Wang X-F, Zhang D-P. Arabidopsis 3-ketoacyl-CoA thiolase-2

(KAT2), an enzyme of fatty acid β-oxidation, is involved in ABA signal transduction.

Plant cell physiology. 2011;52(3):528–38.

49. Singh RK, Sane VA, Misra A, Ali SA, Nath P. Differential expression of the mango

alcohol dehydrogenase gene family during ripening. Phytochemistry.

2010;71(13):1485–94.

50. Wu H-J, Wang Z-M, Wang M, Wang X-J. Widespread long noncoding RNAs as

endogenous target mimics for microRNAs in plants. Plant physiology.

2013;161(4):1875–84.

51. Franco-Zorrilla JM, Valli A, Todesco M, Mateos I, Puga MI, Rubio-Somoza I, Leyva A,

Weigel D, García JA, Paz-Ares J. Target mimicry provides a new mechanism for

regulation of microRNA activity. Nat Genet. 2007;39(8):1033–7.

52. Wu H-x, Jia H-m, Ma X-w, Wang S-b, Xu YQ. W-t, Zhou Y-g, Gao Z-s, Zhan R-l:

Transcriptome and proteomic analysis of mango (Mangifera indica Linn) fruits.

Journal of proteomics. 2014;105:19–30.

53. Luria N, Sela N, Yaari M, Feygenberg O, Kobiler I, Lers A, Prusky D. De-novo assembly

of mango fruit peel transcriptome reveals mechanisms of mango response to hot

water treatment. BMC Genomics. 2014;15(1):957.

54. Tafolla-Arellano JC, Zheng Y, Sun H, Jiao C, Ruiz-May E, Hernández-Oñate MA,

González-León A, Báez-Sañudo R, Fei Z, Domozych D. Transcriptome analysis of

mango (Mangifera indica L.) fruit epidermal peel to identify putative cuticle-

25
 associated genes. Scientific reports. 2017;7:46163.

55. Zuker M. Mfold web server for nucleic acid folding and hybridization prediction.

Nucleic acids research. 2003;31(13):3406–15.

56. Kong L, Zhang Y, Ye Z-Q, Liu X-Q, Zhao S-Q, Wei L, Gao G. CPC: assess the protein-

coding potential of transcripts using sequence features and support vector machine.

Nucleic acids research. 2007;35(suppl_2):W345–9.

57. Han S, Liang Y, Ma Q, Xu Y, Zhang Y, Du W, Wang C, Li Y. LncFinder: an integrated

platform for long non-coding RNA identification utilizing sequence intrinsic

composition, structural information and physicochemical property. Brief Bioinform.

2019;20(6):2009–27.

58. Dai X, Zhao PX. psRNATarget: a plant small RNA target analysis server. Nucleic acids

research. 2011;39(suppl_2):W155–9.

59. Li J, Ma W, Zeng P, Wang J, Geng B, Yang J, Cui Q. LncTar: a tool for predicting the

RNA targets of long noncoding RNAs. Brief Bioinform. 2015;16(5):806–12.

60. Bonnet E, He Y, Billiau K, Van de Peer Y. TAPIR, a web server for the prediction of

plant microRNA targets, including target mimics. Bioinformatics. 2010;26(12):1566–

8.

61. Conesa A, Götz S, García-Gómez JM, Terol J, Talón M, Robles M. Blast2GO: a universal

tool for annotation, visualization and analysis in functional genomics research.

Bioinformatics. 2005;21(18):3674–6.

62. Szcześniak MW, Rosikiewicz W, Makałowska I. CANTATAdb: a collection of plant long

non-coding RNAs. Plant Cell Physiol. 2016;57(1):e8–8.

63. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski

B, Ideker T. Cytoscape: a software environment for integrated models of biomolecular

interaction networks. Genome research. 2003;13(11):2498–504.

26
Supplementary Materials
Supplementary Table 1 miRNAs data

Supplementary Table 2 miRNAs targets data

Supplementary Table 3 Heat responsive lncRNAs data

Supplementary Table 4 Cold responsive lncRNAs data

Supplementary Table 5 KEGG pathways of the target of heat responsive lncRNAs

Supplementary Table 6 KEGG pathways of the target of cold responsive lncRNAs

Supplementary Table 7 Heat responsive lncRNAs as target of miRNAs

Supplementary Table 8 Cold responsive lncRNAs as target of miRNAs

Supplementary Table 9 Cold responsive lncRNAs as target mimic of miRNAs

Supplementary Data Set 1 Heat responsive lncRNAs sequences

Supplementary Data Set 2 Cold responsive lncRNAs sequences

Figures

27
 Figure 1

Workflow for identification of ncRNAs (miRNAs and LncRNAs) and their targets. The yellow

rounded rectangle represents the data input and green rounded rectangles for data output.

The blue rectangles are the processing steps and grey cans represent databases.

28
 Figure 2

Mango miRNAs (A) Length distribution of miRNAs; (B) Hairpin structures of five precursor

miRNAs (Mature miRNA are highlighted by yellow colour) involved in the development

process of mango; (C) GO enrichment analysis of highly enriched target genes of miRNAs.

29
 Figure 3

Temperature responsive lncRNAs of mango. (A) Volcano plot of cold responsive lncRNAs of

mango; green plots represent significantly expressed cold responsive lncRNAs with false

discovery rate (adjusted p-value) of less than 0.05 and log2 fold change of less than -2 or

greater than 2; (B) Conserved number of mango lncRNAs in different plant species; (C)

Interaction subnetwork between three low temperature responsive proteins and cold

responsive lncRNAs of mango; (D) GO enrichment analysis of highly enriched genes

targeted by heat responsive lncRNAs; (E) GO enrichment analysis of highly enriched target

genes of cold responsive lncRNAs.

30
 Figure 4

Interaction network among mango ncRNAs and their targets; Green triangles represent cold

responsive lncRNAs, red triangles represent heat responsive lncRNAs, yellow ellipses are

used for target genes, blue rectangles are for miRNAs and purple V-shapes for target

mimics. (A) Network of interaction among newly identified miRNAs, newly identified lncRNAs

(cold responsive lncRNAs and heat responsive lncRNAs), newly identified target mimic of

miRNAs and mRNAs; (B) subnetwork of interaction among MmiR5408, its target mimic

CRlnc31221, target CRlncRNAs and 47 target genes; (C) Base-pairing interaction between

MmiR5408 and its target mimic cold responsive lncRNA, CRlnc31221.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.

SupplementaryDataSet1.txt
SupplementaryDataSet2.txt
SupplementaryTable4.xlsx
SupplementaryTable5.xlsx
SupplementaryTable6.xlsx
SupplementaryTable7.xlsx
SupplementaryTable8.xlsx
SupplementaryTable9.xlsx

31
 SupplementaryTable2.xlsx
SupplementaryTable3.xlsx
SupplementaryTable1.xlsx

32