Identification of the co-regulatory siRNAs of “miRNA→target” in Oryza sativa

Zhihong Yang; Lan Yu; Yeqing Jiang; Yijun Meng; Chaogang Shao

doi:10.1371/journal.pone.0321182

Abstract

The current “small interfering RNA(siRNA)→Target” mining tools can only search for targets of known siRNAs, and cannot discover co-regulatory siRNAs of unknown sequences that may exist, which means that the “microRNA(miRNA)→Target” database obtained by these mining tools is incomplete. Using the previously developed sRNATargetDigger, we re-mined the rice “miRNA→Target” database supported by the degradome and found 86.2% of the target genes were co-regulated by one or more miRNAs\siRNAs. Besides the known miRNAs, 30 miRNA isoforms (isomiRs) and 12 siRNAs were identified to be involved in co-regulation, which play important roles in rice response to external auxin regulation, rice blast resistance, adventitious root formation, cold resistance, and tillering etc. Some isomiRs even have higher expression levels than miRNAs. In addition, we also found that the regulatory relationship between 51 miRNAs and 48 target genes in the original database could not be verified due to the low expression levels of miRNA, poor complementarity between miRNA and target, or no specific cleavage signal detected by degradome in the middle of the miRNA binding site in the targets. Four miRNAs (osa-miR530-5p,osa-miR319b,osa-miR172c and osa-miR395a) only found isomiRs involved in regulation. In addition, we also found a number of miRNA→target regulatory relationships missed in the database. This study improved the rice “miRNA→target” database which will contribute to the research of rice miRNA and molecular breeding.

Citation: Yang Z, Yu L, Jiang Y, Meng Y, Shao C (2025) Identification of the co-regulatory siRNAs of “miRNA→target” in Oryza sativa . PLoS ONE 20(4): e0321182. https://doi.org/10.1371/journal.pone.0321182

Editor: Keqiang Wu, National Taiwan University, TAIWAN

Received: October 24, 2024; Accepted: March 2, 2025; Published: April 3, 2025

Copyright: © 2025 Yang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The sRNA HTS data sets (GSM1081563, GSM1081564, GSM1081565, GSM1409671, GSM1409672, GSM1409673, GSM1409674, GSM1547539, GSM816687, GSM816688, GSM816689, GSM816690, GSM816691, GSM816692, GSM816693, GSM816694, GSM816695, GSM816696, GSM816697, GSM816698, GSM816699, GSM647192, GSM647193, GSM647194, GSM647195, GSM816730, GSM816731, GSM816732, GSM816733, GSM816735, GSM816736, GSM816737, GSM816738, GSM816739, GSM816740, GSM816741, GSM816742, GSM816743, GSM816744) and the degradome sequencing data sets (GSM434596, GSM455938, GSM4113551, GSM455939, GSM476257) of Oryza sativa were retrieved from GEO (Gene Expression Omnibus; http://www.ncbi.nlm.nih.gov/geo/). The cDNAs and the gene annotations of Oryza sativa were retrieved from the FTP site of the rice genome annotation project (Release7; https://rice.uga.edu/pub/data/Eukaryotic_Projects/o_sativa/annotation_dbs/pseudomolecules/version_7.0/). The miRNA-target database of Oryza sativa were retrieved from the TarDB (http://www.biosequencing.cn/TarDB/download.html) and the miRNAs of Oryza sativa were retrieved from miRBase (release 22.1; https://www.mirbase.org/). The mining software, sRNATargetDigger, can be downloaded from the Project sRNATargetDigger in OSF database (https://osf.io/n7uc9/files/osfstorage).

Funding: This work was supported by the National Natural Sciences Foundation of China[31771457].

Competing interests: The authors have declared that no competing interests exist.

Introduction

MicroRNA (miRNA) is a class of evolutionarily conserved, endogenous single-stranded small RNA(sRNA) with a length of about 19-24 nucleotides. It can cause the cleavage and degradation or translation inhibition of target genes through specific base complementary pairing with target mRNAs, and affecting almost all biological processes from development, physiology to stress response. It is a key factor in post-transcriptional gene expression regulation [1,2]. Recent studies have shown that trans-acting small interfering RNA(ta-siRNA), natural antisense small interfering RNA(nat-siRNA), coding transcript-derived small interfering RNA(ct-siRNA) and other sources of short interfering RNA (siRNA) also have similar gene regulatory functions as miRNA, and play an important role in plant growth and development [3].

Identifying these siRNAs target genes is an important step in our understanding of their functions. Bioinformatics technology can help predict the target genes of siRNAs. Since plant miRNAs mainly bind to target sites through near-perfect complementary pairing, which leads to direct cleavage of target mRNAs, the pattern of plant miRNA recognizing target sites is relatively simple [4]. However, whether the predicted target genes are accurate generally needs biological verification. The commonly used methods are as follows: one is the Agrobacterium injection method, in which the constructed vector containing miRNA and target genes is injected into tobacco leaves to detect the cleavage results of miRNA on target genes [5], the second is to use wheat germ lysate to detect the cleavage activity of miRNA on target genes in vitro [6], and the third is to directly analyze the products of miRNA cleavage using 5’rapid-amplification of cDNA ends (5’RACE). This method involves attaching a linker to the 5’ end of the miRNA cleavage product, and after Reverse Transcription-Polymerase Chain Reaction(RT-PCR) and sequencing, it can not only verify the cleavage characteristics of miRNA on target genes, but also identify the precise cleavage sites of miRNA. It is currently the most reliable method for identifying miRNA target genes [5]. However, the above experimental verification methods are relatively cumbersome, which greatly limits the functional identification of miRNA.

High-throughput sequencing (HTS) technology is a breakthrough sequencing technology developed in recent years, which can determine millions of nucleic acid sequences at a time. Depending on the RNA separation conditions, HTS can be used to determine the transcriptome of mRNAs, the degradome of mRNAs, and the transcriptome of sRNA. It can even obtain various sRNAs loaded in the Argonaute (AGO) complex through specific antibody capture technology, which provides a powerful tool for the identification of miRNAs and their target genes [7–9]. In particular, the degradome high-throughput sequencing technology [10], or named as parallel analysis of RNA ends (PARE)) [11] and nanoPARE techniques [12], can capture all the degradation fragments of transcripts, and by using bioinformatics tools (CleaveLand [13], SeqTar [14], PAREsnip2 [15], sPARTA [16], etc.) to match with reference sequences, it can detect specific cleavage signals on transcripts, which can provide experimental evidence for miRNAs to perform cutting functions, and can be used for large-scale identification of the targets of plant miRNAs [10–12].

Due to the regulatory relationship between miRNAs and target genes being a “multi” to “multi” regulatory network, a target gene may be co-regulated by multiple miRNAs or siRNAs at the same or different sites. However, current “siRNAs→Target” mining tools based on degradome can only search for target genes of known sequence siRNAs, and cannot discover co-regulatory siRNAs of unknown sequences that may exist. Therefore, the miRNA→target regulatory network obtained by these mining tools may be incomplete.Besides, according to the previous reports, the effectiveness of siRNA in cleaving and inhibiting mRNA is positively correlated with the concentration of siRNA [17].So, if there are unknown co-regulatory siRNAs with much higher expression levels than known miRNAs, the contribution of this miRNA to target gene cleavage is very little, and the previously reported miRNA→target regulatory relationship may be a false positive result.

To address this issue, we integrated reverse mining technology with the classic “siRNA→Target” mining algorithm to develop a new mining tool, sRNATargetDigger [18]. This tool can search for all possible co-regulatory siRNAs from HTS data of sRNAs, and determine the degree of participation of these siRNAs in target regulation based on their expression levels, thereby obtaining a more complete and reliable “miRNAs\siRNA→Target” regulatory relationship and also enabling bidirectional mining [18]. Testing of “miRNAs→Target” in Arabidopsis thaliana showed that 69 miRNA target genes found co-regulatory siRNAs in one or more tissues, forming 170 new siRNA→target regulatory relationships [18].

Rice is an important food crop and a model plant for life science research. The “miRNA→Target” database is of great significance for basic research and molecular breeding of rice. In this study, we used the sRNATargetDigger tool and massive HTS data from public databases to re-mine the reported rice “miRNA→Target” database, searching for missing co-regulatory siRNAs information.

Materials and methods

Data sources

The cDNAs and the gene annotations of Oryza sativa (Nipponbare) were retrieved from the FTP site of the rice genome annotation project (Release 7; https://rice.uga.edu/pub/data/Eukaryotic_Projects/o_sativa/annotation_dbs/pseudomolecules/version_7.0/) [19]. The sRNA HTS data sets (GSM1081563, GSM1081564, GSM1081565, GSM1409671, GSM1409672, GSM1409673, GSM1409674, GSM1547539, GSM816687, GSM816688, GSM816689, GSM816690, GSM816691, GSM816692, GSM816693, GSM816694, GSM816695, GSM816696, GSM816697, GSM816698, GSM816699, GSM647192, GSM647193, GSM647194, GSM647195, GSM816730, GSM816731, GSM816732, GSM816733, GSM816735, GSM816736, GSM816737, GSM816738, GSM816739, GSM816740, GSM816741, GSM816742, GSM816743, GSM816744) and the degradome sequencing data sets (GSM434596, GSM455938, GSM4113551, GSM455939, GSM476257) of Oryza sativa were retrieved from GEO (Gene Expression Omnibus; http://www.ncbi.nlm.nih.gov/geo/) [20]. See S1 Table for the sequencing library information of these data sets. In order to allow cross-library comparison, the normalized read count (in RPM, reads per million) of a short read from a specific library was calculated by dividing the raw count of this read by the total counts of the library, and then multiplied by 10⁶.

Prediction of miRNA→target binding sites and identification of specific cleavage signals in target genes

We downloaded the newly published miRNA→target database of Oryza sativa from the TarDB (http://www.biosequencing.cn/TarDB/download.html) [21], and extracted the sequences of miRNAs and targets from the miRBase (release 22.1, https://www.mirbase.org/) [22] and rice cDNA databases(Release 7; https://rice.uga.edu/pub/data/Eukaryotic_Projects/o_sativa/annotation_dbs/pseudomolecules/version_7.0/) [19], respectively.Then the binding sites of miRNA on target were predicted by miRU algorithm [23] to distinguish two categories of co-regulatory siRNAs which have the same or different binding site with miRNAs on target genes. Next, we matched the degradome to the target, recorded the 5’-end site information of the exact match (if there are multiple exact matches, they all need to be recorded) and calculated the degradation signal intensity of cleavage site and background as previously reported [18]:

Degradation signal intensity of cleavage site = sum of the expression counts of all matching degradome sequences at the cleavage site/sum of the number of matching degradome sequences at the cleavage site;

Degradation signal intensity of background = sum of the expression counts of the matching degradome sequences in full-length (except for the cleavage site)/sum of the number of matching degradome sequences in full-length(except for the cleavage site).

Screening of specific cleavage sites: The degradation signal intensity of the cleavage site should be 5 times or more of the background degradation signal intensity.

Co-regulatory siRNA identification

The samples were divided into two different tissues, seedling and panicle, and the mining was conducted separately. There were a total of 21 sets of sRNA sequencing data and 3 degradome data for seedling, and 18 sets of sRNA sequencing data and 2 degradome data for panicle (S1 Table).

(1) Search for co-regulatory siRNAs with the same cleavage site as miRNA: Using the forward mining module of sRNATargetDigger (https://osf.io/n7uc9/files/osfstorage), input the degradation group data, sRNA sequencing data, rice miRNA and target gene sequences, and search co-regulatory siRNAs with the same cleavage site as miRNA. The expression counts of co-regulatory siRNAs needs to be greater than or equal to 10 rpm, and not less than 1/10 of the highest value of the regulatory miRNA/siRNA at that site.
(2) Search for co-regulatory siRNAs with different cleavage sites from miRNA: Utilize the reverse mining module of sRNATargetDigger, input specific cleavage site information which is not in the middle of miRNA binding site, sRNA sequencing data, and target gene sequence to search co-regulatory siRNAs. The expression value of co-regulatory siRNA needs to be greater than or equal to 10 rpm, and not less than 1/10 of the highest value of the regulatory siRNA at that site.

The co-regulatory siRNAs obtained will be annotated as miRNA, isomiR or other siRNA by searching the miRNA and pre-miRNA databases.

Screening and verification of the results

The co-regulatory siRNAs discovered by sRNATargetDigger were further screened and validated by psRNATarget (https://www.zhaolab.org/psRNATarget/) [24], a plant sRNA target analysis server. “Submit small RNAs and Target” module of psRNATarget and the default parameters were selected. Only the predicted results with “Inhibition: Cleavage” were retained, and the cleavage signal found by degradome on the target should be located in the middle of the binding site of the co-regulatory siRNA.

For miRNA→target in the TarDB database [21] that failed to pass the sRNATargetDigger, we further analyzed the expression of the miRNA, the cleavage signal on the target, and the binding of miRNA to the target for verification. If the miRNA is not expressed or expressed very low (<10rpm), or if no cleavage signal is detected by the degradome in the middle of the binding site of the miRNA on the target, or if both the miRU and psRNATarget detections indicate that the miRNA cannot bind to the target, it is determined that the miRNA→target regulatory relationship is incorrect in these tissues or conditions.

Results and discussion

Discovery of anomalies in the rice “miRNA→target” database

We downloaded 480 miRNA→target pairs of rice supported by the degradome from the newly published TarDB database [21], including 98 miRNAs and 165 target genes, for re-mining. The regulatory relationships between 62 miRNAs and 116 genes were validated by sRNATargetDigger (S2 and S3 Tables). The rest which could not pass the verification were further analyzed. miRU and psRNATarget found osa-miR408-5p, osa-miR5541 and osa-miR2098-3p could not bind with their corresponding targets, LOC_Os04g39350.1 (OsHIPP19, heavy-metal-associated isoprenylated plant protein), LOC_Os09g32700.1(expressed protein) and LOC_Os09g39750.1(domain of unknown function DUF966 domain containing protein), respectively (S4 Table) .Forty one target genes did not find cleavage signals in the middle of the miRNA binding sites by degradome. (Fig 1, S1 Fig, S4 Table) and expression analysis found 22 miRNAs which regulate 44 targets were almost not expressed in all samples (S4 Table). Considering that the sRNA data used by sRNATargetDigger for mining is exactly the same as that provided by the TarDB database, and the degradome data have one more group (GSM476257) than that provided by the TarDB (theoretically, more cleavage signals can be detected, improving the success rate of miRNA→target prediction), the impact of the datasets on the results can be ruled out. So, the regulation relationship between these miRNA→target pairs that could not be validated by sRNATargetDigger may not be correct in these tissues or conditions. Due to the limitation of samples, we cannot rule out the possibility that the regulatory relationship may be exist in other tissues or under specific conditions.

Download:

Fig 1. Four examples of the target plots (t-plots) which did not find cleavage signals in the middle of the miRNA binding sites by degradome.

All 5 degradome sequencing data libraries (GSM434596, GSM455938, GSM4113551, GSM455939, GSM476257) were recruited for T-plot profiling. The IDs of the target transcripts and the corresponding miRNAs are listed on the top of sub-figures. The x axes measure the positions of the degradome signals along the transcripts, and the y axes measure the degradome signal intensities based on normalized counts (in RPM, reads per million).The binding sites of the miRNA on their target transcripts were denoted by red horizontal lines. No specific cleavage signals are found in the middle of the miRNA binding sites in these target genes.

https://doi.org/10.1371/journal.pone.0321182.g001

In addition, we found that many miRNAs belonging to the same family of rice in the TarDB database share similar sequences and co-regulate the same target gene. However, in the co-regulation relationship, the TarDB database does not provide information on the contribution degree of each miRNA. Through sRNATargetDigger mining, we found that there are significant differences in the expression levels of different members of many miRNA families. For example, three members of the osa-miR164 family (osa-miR164d,osa-miR164e and osa-miR164f) co-regulate LOC_Os06g46270.1(OMTN4, no apical meristem protein) in rice seedlings, but high-throughput sequencing shows the abundant of osa-miR164f is more than four times than that of osa-miR164d and osa-miR164e in various datasets, and even more than 100 times in some datasets. According to the previous reports, the effectiveness of siRNA in cleaving and inhibiting mRNA is positively correlated with the concentration of siRNA [17]. So, osa-miR164f is the main regulator of LOC_Os06g46270.1 (Fig 2A). In rice panicle, osa-miR396c-5p and osa-miR396e-5p co-regulate LOC_Os02g45570.2(OsGRF10, growth regulating factor), but the expression level of osa-miR396e-5p in various datasets also far exceeds that of osa-miR396c-5p, making it the main regulator of LOC_Os02g45570.2 (Fig 2B). Besides, we found 6 miRNAs in 5 miRNA families were expressed at less than 1/10 of the highest expression miRNA which cut the target. In theory, their contribution to target gene cleavage is very little (S4 Table).

Download:

Fig 2. Expression bar graph of the miRNAs belonging to the same family involved in the co-regulation of target genes.

A. osa-miR164d,osa-miR164e and osa-miR164f co-regulate LOC_Os06g46270.1 in rice seedlings. The x axes represents the GSM numbers of 21 samples in seedling and the y axes represents the expression values of miRNAs (in RPM, reads per million).The abundant of osa-miR164f is more than that of osa-miR164d and osa-miR164e in various datasets, and it is obviously the main regulator of LOC_Os06g46270.1(OMTN4). B. osa-miR396c-5p and osa-miR396e-5p co-regulate LOC_Os02g45570.2 (OsGRF10, growth regulating factor) in rice panicle.The x axes represents the GSM numbers of 18 samples in panicle and the y axes represents the expression values of miRNAs (in RPM, reads per million).The expression level of osa-miR396e-5p in various datasets far exceeds that of osa-miR396c-5p, making it the main regulator of LOC_Os02g45570.2.

https://doi.org/10.1371/journal.pone.0321182.g002

During the re-mining, sRNATargetDigger also found some missing miRNA→target regulatory relationships in the TarDB database (Table 1).For example, in the original database, osa-miR444c.2 only regulates LOC_Os02g49840.3(OsMADS57, MADS-box gene) and LOC_Os02g49840.4 (OsMADS57), but in rice seedlings, we found that osa-miR444c.2 also participates in the regulation of LOC_Os02g49840.2 (OsMADS57) (S2 Table). In addition to being regulated by osa-miR444c.2, LOC_Os02g36924.1(OsMADS27, MADS-box family gene)is also found to be co-regulated by osa-miR444e at same site (cutsite 575) and osa-miR444c.1 at different site (cutsite 569) (Fig 3, S2 Table).All these missing miRNA→target pairs identified by sRNATargetDigger passed the psRNATarget validation (S5 Table). We further investigated the relevant literature on the mining of miRNA target genes in rice and found our partial results, the regulatory relationships of “osa-miR444c.1→LOC_Os02g36924.1 (OsMADS27)”, “osa-miR444c.2→LOC_Os02g49840.2(OsMADS57)”, “osa-miR444e→LOC_Os02g36924.1 (OsMADS27)” and “osa-miR444e→ LOC_Os04g38780.1(transcription factor)” were also supported by the research of Fei et al. [25].

Download:

Table 1. Missing or novel miRNA→target regulatory relationships discovered by sRNATargetDigger in rice.

https://doi.org/10.1371/journal.pone.0321182.t001

Download:

Fig 3. osa-miR444c.2 and co-regulating miRNAs with their target LOC_Os02g36924.1 in rice seedling.

A. Expression of osa-miR444c.2, osa-miR444e and osa-miR444c.1 in the samples of wt2003, wt2007 and wt2011, and their alignments with target genes. B and C. Degradome sequencing data libraries GSM4113551 (B) and GSM434596 (C) were recruited for T-plot profiling, respectively. The ID of the target LOC_Os02g36924.1(OsMADS27) is listed on the top of the figure. The y axes measure the normalized reads (in RMP, reads per million) of the degradome signals, and the x axes represent the position of the cleavage signals on the target transcripts. The binding site of the osa-miR444e\osa-miR444c.2 and osa-miR444c.1 on their target was denoted by purple and green horizontal line, respectively.The vertical dotted line indicates the specific cleavage site.

https://doi.org/10.1371/journal.pone.0321182.g003

Co-regulatory siRNAs identified by sRNATargetDigger in rice

Among the 116 target genes verified by sRNATargetDigger in rice seedlings and panicles, 100 target genes were found to be co-regulated by two or more miRNAs\siRNAs (S2 and S3 Tables). Most of the co-regulation occurred at the same site except for 5 targets. Only 16 target genes were regulated by one miRNA or isomiR. Among them, four miRNAs, namely osa-miR530-5p, osa-miR319b, osa-miR172c and osa-miR395a, have low expression levels and sRNATargetDigger only found corresponding isomiRs involved in the regulation (S2 and S3 Tables).

IsomiRs are predominantly derived from the alternative and imprecise cleavage of Drosha and Dicer during pri-miRNA/pre-miRNA processing, 3’ addition events in miRNA maturation processes or base modification of pri-miRNA/pre-miRNA/miRNA [26]. They are very close to miRNA sequences and RNAi machinery generates isomiRs as a potential strategy to increase the specificity and efficiency of silencing target genes [26]. In this study, sRNATargetDigger identified 30 isomiRs which play an important role in co-regulation of the targets in rice (S2 and S3 Tables) and these newly discovered isomiR → target regulatory relationships were further confirmed by psRNATarget (S6 Table).Some isomiRs have very high expression levels, even exceeding miRNAs. For example, osa-miR167d-5p, which plays an important role in rice response to external auxin regulation [27] and rice blast resistance [28], has been found co-regulation with osa-miR167c-5p and isomiR of osa-miR167d-5p to target ARF8 (LOC_Os04g57610) in seedling and panicle (S2 and S3 Tables). In most sRNA sequencing data, the expression level of isomiR of osa-miR167d-5p far exceeds that of osa-miR167d-5p and osa-miR167c-5p (Fig 4).Although literature reported that 22-nt miRNAs had the ability to trigger the phased siRNAs production from the cleaved targets [29], further analysis did not find that this 22-nt isomiR of osa-miR167d-5p can stimulate ARF8 to produce phased siRNAs. In Arabidopsis, Li et al. also found that many MIR genes produce 22-nt isoforms due to imprecision in DCL1 processing, yet most 22-nt miRNA isoforms do not trigger phasiRNA production from their target RNAs [30].The target gene OsDRM2 (LOC_Os03g02010.4) of osa-miR820c play an important role in regulating rice vegetative and reproductive growth through DNA methylation [31]. sRNATargetDigger found that there are five other isomiRs of osa-miR820c and four siRNAs involved in the co-regulation.One of the isomiRs of osa-miR820c,“UCGGCCUCGUGGAUGGACCAGGAG”, has a higher expression level than osa-miR820c in most samples (S2 Table).

Download:

Fig 4. osa-miR167d-5p and co-regulating siRNAs with their targetARF8 (LOC_Os04g57610) in rice seedling.

A. Expression of osa-miR167d-5p, osa-miR167c-5p and isomiR of osa-miR167d-5p in the samples of control, drought, salt, cold and heat, and their alignments with target genes. The red number indicates that the sRNA plays the most important regulatory role. B. Degradome sequencing data library GSM434596 was recruited for T-plot profiling. The ID of the target ARF8 is listed on the top of the figure. The y axes measure the normalized reads (in RMP, reads per million) of the degradome signals, and the x axes represent the position of the cleavage signals on the target transcripts. The binding site of the osa-miR167d-5p, osa-miR167c-5p and isomiR of osa-miR167d-5p on their target was denoted by red horizontal line.The red vertical dotted line indicates the specific cleavage site.

https://doi.org/10.1371/journal.pone.0321182.g004

The main root of rice will die after one week of germination, and the adventitious roots that grow later are the main organs for rice to fix, absorb nutrients and water, affecting the growth, yield and resistance of rice. Highly expressed osa-miR156j-5p regulates the growth hormone transport and signaling by inhibiting the activation of OsMADS50 by OsSPL3 (LOC_Os02g04680)/OsSPL12 (LOC_Os06g49010), promoting the occurrence of adventitious roots in rice [32]. We found in rice seedlings that the isomiR of osa-miR156j-5p also participates in the co-regulation of OsSPL3/OsSPL12, and its expression level is comparable to that of osa-miR156j-5p (S2 Table), indicating that the isomiR of osa-miR156j-5p also plays an important role in the occurrence of adventitious roots.

In addition, we found that some miRNAs and isomiRs regulate target genes under different conditions. For example, seedling under normal conditions or drought, salt, cold treatment, osa-miR172d-3p was involved in the regulation of AP2 genes (LOC_Os03g60430, LOC_Os04g55560, LOC_Os05g03040, LOC_Os07g13170), but under heat treatment or heat + salt treatment, only the isomiR of osa-miR172d-3p was involved in the regulation (S2 Table).

Besides the 30 isomiRs, 12 other co-regulatory siRNAs were also identified by sRNATargetDigger (S2 and S3 Tables) and passed psRNATarget validation (S6 Table). The expressions of some siRNAs are tissue or condition specific. For example, siRNA_12 is only expressed in panicle, while other co-regulatory siRNAs are only expressed in seedling. In addition, siRNA_3 and siRNA_12 are only expressed in the heat condition.Compared to isomiRs, the expression levels of these siRNAs are relatively low and may play a supporting role in co-regulation.

Conclusions

In this study, we re-mined the rice “miRNA→Target” database by sRNATargetDigger and found the regulatory relationship between 51 miRNAs and 48 target genes in the original database could not be verified due to the low expression levels of miRNA, poor complementarity between miRNA and target, or no specific cleavage signal detected by degradome in the middle of the miRNA binding site in the targets. These miRNA→target pairs may be incorrect in these tissues or conditions. In addition, a number of miRNA→target regulatory relationships missed in the database have been discovered. We also found 86.2% of the target genes were co-regulated by one or more miRNAs\siRNAs. Besides the known miRNAs, 30 isomiRs and 12 siRNAs were identified to be involved in co-regulation, which play important roles in rice response to external auxin regulation, rice blast resistance, adventitious root formation, etc. Some isomiRs even have higher expression levels than miRNAs.

Supporting information

S1 Fig. Target plots (t-plots) of forty one target genes which did not find cleavage signals in the middle of the miRNA binding sites by degradome.

All 5 degradome sequencing data libraries (GSM434596, GSM455938, GSM4113551, GSM455939, GSM476257) were recruited for T-plot profiling. The IDs of the target transcripts and the corresponding miRNAs are listed on the top of sub-figures. The x axes measure the positions of the degradome signals along the transcripts, and the y axes measure the degradome signal intensities based on normalized counts (in RPM, reads per million).The binding sites of the miRNA on their target transcripts were denoted by red horizontal lines. No specific cleavage signals have been found in the middle of the miRNA binding sites in these target genes.

https://doi.org/10.1371/journal.pone.0321182.s001

(PDF)

S1 Table. Sequencing library information of sRNA-seq and degradome data for rice miRNA-target re-mining.

https://doi.org/10.1371/journal.pone.0321182.s002

(XLS)

S2 Table. miRNAs and co-regulatory siRNAs identified in seedling.

https://doi.org/10.1371/journal.pone.0321182.s003

(XLS)

S3 Table. miRNAs and co-regulatory siRNAs identified in panicle.

https://doi.org/10.1371/journal.pone.0321182.s004

(XLS)

S4 Table. miRNA-target pairs are not found by sRNATargetDigger.

https://doi.org/10.1371/journal.pone.0321182.s005

(XLS)

S5 Table. Validation of the Newly discovered miRNA-target pairs in rice by psRNATarget.

https://doi.org/10.1371/journal.pone.0321182.s006

(XLS)

S6 Table. Validation of the co-regulatory siRNA-target pairs by psRNATarget.

https://doi.org/10.1371/journal.pone.0321182.s007

(XLSX)

Acknowledgments

We would like to thank all the publicly available data sets and the scientists behind them.

References

1. Carthew RW, Sontheimer EJ. Origins and Mechanisms of miRNAs and siRNAs. Cell. 2009;136(4):642–55. pmid:19239886
- View Article
- PubMed/NCBI
- Google Scholar
2. Chen X. Small RNAs and their roles in plant development. Annu Rev Cell Dev Biol. 2009;25:21–44. pmid:19575669
- View Article
- PubMed/NCBI
- Google Scholar
3. Vaucheret H, Voinnet O. The plant siRNA landscape. Plant Cell. 2024;36(2):246–75. pmid:37772967
- View Article
- PubMed/NCBI
- Google Scholar
4. Rhoades MW, Reinhart BJ, Lim LP, Burge CB, Bartel B, Bartel DP. Prediction of plant microRNA targets. Cell. 2002;110(4):513–20. pmid:12202040
- View Article
- PubMed/NCBI
- Google Scholar
5. Llave C, Xie Z, Kasschau KD, Carrington JC. Cleavage of Scarecrow-like mRNA targets directed by a class of Arabidopsis miRNA. Science. 2002;297(5589):2053–6. pmid:12242443
- View Article
- PubMed/NCBI
- Google Scholar
6. Mallory AC, Bartel DP, Bartel B. MicroRNA-directed regulation of Arabidopsis AUXIN RESPONSE FACTOR17 is essential for proper development and modulates expression of early auxin response genes. Plant Cell. 2005;17(5):1360–75. pmid:15829600
- View Article
- PubMed/NCBI
- Google Scholar
7. Fahlgren N, Howell MD, Kasschau KD, Chapman EJ, Sullivan CM, Cumbie JS, et al. High-throughput sequencing of Arabidopsis microRNAs: evidence for frequent birth and death of MIRNA genes. PLoS One. 2007;2(2):e219. pmid:17299599
- View Article
- PubMed/NCBI
- Google Scholar
8. Varshney RK, Nayak SN, May GD, Jackson SA. Next-generation sequencing technologies and their implications for crop genetics and breeding. Trends Biotechnol. 2009;27(9):522–30. pmid:19679362
- View Article
- PubMed/NCBI
- Google Scholar
9. Johnson C, Bowman L, Adai AT, Vance V, Sundaresan V. CSRDB: a small RNA integrated database and browser resource for cereals. Nucleic Acids Res. 2007;35(Database issue):D829–33. pmid:17169981
- View Article
- PubMed/NCBI
- Google Scholar
10. Addo-Quaye C, Eshoo TW, Bartel DP, Axtell MJ. Endogenous siRNA and miRNA Targets Identified by Sequencing of the Arabidopsis Degradome. Current Biology. 2008;18(10):758–62.
- View Article
- Google Scholar
11. German MA, Pillay M, Jeong D-H, Hetawal A, Luo S, Janardhanan P, et al. Global identification of microRNA-target RNA pairs by parallel analysis of RNA ends. Nat Biotechnol. 2008;26(8):941–6. pmid:18542052
- View Article
- PubMed/NCBI
- Google Scholar
12. Schon MA, Kellner MJ, Plotnikova A, Hofmann F, Nodine MD. NanoPARE: parallel analysis of RNA 5’ ends from low-input RNA. Genome Res. 2018;28(12):1931–42. pmid:30355603
- View Article
- PubMed/NCBI
- Google Scholar
13. Addo-Quaye C, Miller W, Axtell MJ. CleaveLand: a pipeline for using degradome data to find cleaved small RNA targets. Bioinformatics. 2009;25(1):130–1. pmid:19017659
- View Article
- PubMed/NCBI
- Google Scholar
14. Zheng Y, Li Y-F, Sunkar R, Zhang W. SeqTar: an effective method for identifying microRNA guided cleavage sites from degradome of polyadenylated transcripts in plants. Nucleic Acids Res. 2012;40(4):e28. pmid:22140118
- View Article
- PubMed/NCBI
- Google Scholar
15. Joshua T, Leighton F, Zahara M, Xu P, Tamas M, Vincent D. PAREsnip2: a tool for high-throughput prediction of small RNA targets from degradome sequencing data using configurable targeting rules. Nucleic Acids Research. 2018;46(17):8730–9.
- View Article
- Google Scholar
16. Kakrana A, Hammond R, Patel P, Nakano M, Meyers BC. sPARTA: a parallelized pipeline for integrated analysis of plant miRNA and cleaved mRNA data sets, including new miRNA target-identification software. Nucleic Acids Res. 2014;42(18):e139. pmid:25120269
- View Article
- PubMed/NCBI
- Google Scholar
17. Doench JG, Petersen CP, Sharp PA. siRNAs can function as miRNAs. Genes Dev. 2003;17(4):438–42. pmid:12600936
- View Article
- PubMed/NCBI
- Google Scholar
18. Ye X, Yang Z, Jiang Y, Yu L, Guo R, Meng Y, et al. sRNATargetDigger: A bioinformatics software for bidirectional identification of sRNA-target pairs with co-regulatory sRNAs information. PLoS One. 2020;15(12):e0244480. pmid:33370386
- View Article
- PubMed/NCBI
- Google Scholar
19. Kawahara Y, de la Bastide M, Hamilton JP, Kanamori H, McCombie WR, Ouyang S, et al. Improvement of the Oryza sativa Nipponbare reference genome using next generation sequence and optical map data. Rice (N Y). 2013;6(1):4. pmid:24280374
- View Article
- PubMed/NCBI
- Google Scholar
20. Barrett T, Troup DB, Wilhite SE, Ledoux P, Rudnev D, Evangelista C, et al. NCBI GEO: archive for high-throughput functional genomic data. Nucleic Acids Res. 2009;37(Database issue):D885–90. pmid:18940857
- View Article
- PubMed/NCBI
- Google Scholar
21. Liu J, Liu X, Zhang S, Liang S, Luan W, Ma X. TarDB: an online database for plant miRNA targets and miRNA-triggered phased siRNAs. BMC Genomics. 2021;22(1):348. pmid:33985427
- View Article
- PubMed/NCBI
- Google Scholar
22. Kozomara A, Birgaoanu M, Griffiths-Jones S. miRBase: from microRNA sequences to function. Nucleic Acids Res. 2019;47(D1):D155–62. pmid:30423142
- View Article
- PubMed/NCBI
- Google Scholar
23. Zhang Y. miRU: an automated plant miRNA target prediction server. Nucleic Acids Res. 2005;33(Web Server issue):W701-4. pmid:15980567
- View Article
- PubMed/NCBI
- Google Scholar
24. Dai X, Zhuang Z, Zhao PX. psRNATarget: a plant small RNA target analysis server (2017 release). Nucleic Acids Research. 2018;46:W49–54.
- View Article
- Google Scholar
25. Fei Y, Wang R, Li H, Liu S, Zhang H, Huang J. DPMIND: degradome-based plant miRNA-target interaction and network database. Bioinformatics. 2018;34(9):1618–20. pmid:29280990
- View Article
- PubMed/NCBI
- Google Scholar
26. Fard EM, Moradi S, Salekdeh NN, Bakhshi B, Ghaffari MR, Zeinalabedini M, et al. Plant isomiRs: origins, biogenesis, and biological functions. Genomics. 2020;112(5):3382–95. pmid:32561347
- View Article
- PubMed/NCBI
- Google Scholar
27. Yang JH, Han SJ, Yoon EK, Lee WS. “Evidence of an auxin signal pathway, microRNA167-ARF8-GH3, and its response to exogenous auxin in cultured rice cells”. Nucleic Acids Res. 2006;34(6):1892–9. pmid:16598073
- View Article
- PubMed/NCBI
- Google Scholar
28. Yong ZhuHe WangViswanathan ChandranZhao Z-X, Feng Q, Cao X-L, Zhu Y, Wang H, Chandran V, et al. Osa-miR167d facilitates infection of Magnaporthe oryzae in rice. J Integr Plant Biol. 2020;62(5):702–15. pmid:31001874
- View Article
- PubMed/NCBI
- Google Scholar
29. Chen H-M, Chen L-T, Patel K, Li Y-H, Baulcombe DC, Wu S-H. 22-Nucleotide RNAs trigger secondary siRNA biogenesis in plants. Proc Natl Acad Sci U S A. 2010;107(34):15269–74. pmid:20643946
- View Article
- PubMed/NCBI
- Google Scholar
30. Li S, Le B, Ma X, Li S, You C, Yu Y, et al. Biogenesis of phased siRNAs on membrane-bound polysomes in Arabidopsis. Elife. 2016;5:e22750. pmid:27938667
- View Article
- PubMed/NCBI
- Google Scholar
31. Moritoh S, Eun C, Ono A, Asao H, Okano Y, Yamaguchi K, et al. Targeted disruption of an orthologue of DOMAINS REARRANGED METHYLASE 2, OsDRM2, impairs the growth of rice plants by abnormal DNA methylation. Plant J. 2012;71(1):85–98. pmid:22380881
- View Article
- PubMed/NCBI
- Google Scholar
32. Shao Y, Zhou H-Z, Wu Y, Zhang H, Lin J, Jiang X, et al. OsSPL3, an SBP-domain protein, regulates crown root development in rice. Plant Cell. 2019;31(6):1257–75. pmid:30940685
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Carthew RW, Sontheimer EJ. Origins and Mechanisms of miRNAs and siRNAs. Cell. 2009;136(4):642–55. pmid:19239886
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Chen X. Small RNAs and their roles in plant development. Annu Rev Cell Dev Biol. 2009;25:21–44. pmid:19575669
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Vaucheret H, Voinnet O. The plant siRNA landscape. Plant Cell. 2024;36(2):246–75. pmid:37772967
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. Rhoades MW, Reinhart BJ, Lim LP, Burge CB, Bartel B, Bartel DP. Prediction of plant microRNA targets. Cell. 2002;110(4):513–20. pmid:12202040
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref5] 5. Llave C, Xie Z, Kasschau KD, Carrington JC. Cleavage of Scarecrow-like mRNA targets directed by a class of Arabidopsis miRNA. Science. 2002;297(5589):2053–6. pmid:12242443
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref6] 6. Mallory AC, Bartel DP, Bartel B. MicroRNA-directed regulation of Arabidopsis AUXIN RESPONSE FACTOR17 is essential for proper development and modulates expression of early auxin response genes. Plant Cell. 2005;17(5):1360–75. pmid:15829600
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref7] 7. Fahlgren N, Howell MD, Kasschau KD, Chapman EJ, Sullivan CM, Cumbie JS, et al. High-throughput sequencing of Arabidopsis microRNAs: evidence for frequent birth and death of MIRNA genes. PLoS One. 2007;2(2):e219. pmid:17299599
View Article
PubMed/NCBI
Google Scholar

[26] View Article

[27] PubMed/NCBI

[28] Google Scholar

[ref8] 8. Varshney RK, Nayak SN, May GD, Jackson SA. Next-generation sequencing technologies and their implications for crop genetics and breeding. Trends Biotechnol. 2009;27(9):522–30. pmid:19679362
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref9] 9. Johnson C, Bowman L, Adai AT, Vance V, Sundaresan V. CSRDB: a small RNA integrated database and browser resource for cereals. Nucleic Acids Res. 2007;35(Database issue):D829–33. pmid:17169981
View Article
PubMed/NCBI
Google Scholar

[34] View Article

[35] PubMed/NCBI

[36] Google Scholar

[ref10] 10. Addo-Quaye C, Eshoo TW, Bartel DP, Axtell MJ. Endogenous siRNA and miRNA Targets Identified by Sequencing of the Arabidopsis Degradome. Current Biology. 2008;18(10):758–62.
View Article
Google Scholar

[38] View Article

[39] Google Scholar

[ref11] 11. German MA, Pillay M, Jeong D-H, Hetawal A, Luo S, Janardhanan P, et al. Global identification of microRNA-target RNA pairs by parallel analysis of RNA ends. Nat Biotechnol. 2008;26(8):941–6. pmid:18542052
View Article
PubMed/NCBI
Google Scholar

[41] View Article

[42] PubMed/NCBI

[43] Google Scholar

[ref12] 12. Schon MA, Kellner MJ, Plotnikova A, Hofmann F, Nodine MD. NanoPARE: parallel analysis of RNA 5’ ends from low-input RNA. Genome Res. 2018;28(12):1931–42. pmid:30355603
View Article
PubMed/NCBI
Google Scholar

[45] View Article

[46] PubMed/NCBI

[47] Google Scholar

[ref13] 13. Addo-Quaye C, Miller W, Axtell MJ. CleaveLand: a pipeline for using degradome data to find cleaved small RNA targets. Bioinformatics. 2009;25(1):130–1. pmid:19017659
View Article
PubMed/NCBI
Google Scholar

[49] View Article

[50] PubMed/NCBI

[51] Google Scholar

[ref14] 14. Zheng Y, Li Y-F, Sunkar R, Zhang W. SeqTar: an effective method for identifying microRNA guided cleavage sites from degradome of polyadenylated transcripts in plants. Nucleic Acids Res. 2012;40(4):e28. pmid:22140118
View Article
PubMed/NCBI
Google Scholar

[53] View Article

[54] PubMed/NCBI

[55] Google Scholar

[ref15] 15. Joshua T, Leighton F, Zahara M, Xu P, Tamas M, Vincent D. PAREsnip2: a tool for high-throughput prediction of small RNA targets from degradome sequencing data using configurable targeting rules. Nucleic Acids Research. 2018;46(17):8730–9.
View Article
Google Scholar

[57] View Article

[58] Google Scholar

[ref16] 16. Kakrana A, Hammond R, Patel P, Nakano M, Meyers BC. sPARTA: a parallelized pipeline for integrated analysis of plant miRNA and cleaved mRNA data sets, including new miRNA target-identification software. Nucleic Acids Res. 2014;42(18):e139. pmid:25120269
View Article
PubMed/NCBI
Google Scholar

[60] View Article

[61] PubMed/NCBI

[62] Google Scholar

[ref17] 17. Doench JG, Petersen CP, Sharp PA. siRNAs can function as miRNAs. Genes Dev. 2003;17(4):438–42. pmid:12600936
View Article
PubMed/NCBI
Google Scholar

[64] View Article

[65] PubMed/NCBI

[66] Google Scholar

[ref18] 18. Ye X, Yang Z, Jiang Y, Yu L, Guo R, Meng Y, et al. sRNATargetDigger: A bioinformatics software for bidirectional identification of sRNA-target pairs with co-regulatory sRNAs information. PLoS One. 2020;15(12):e0244480. pmid:33370386
View Article
PubMed/NCBI
Google Scholar

[68] View Article

[69] PubMed/NCBI

[70] Google Scholar

[ref19] 19. Kawahara Y, de la Bastide M, Hamilton JP, Kanamori H, McCombie WR, Ouyang S, et al. Improvement of the Oryza sativa Nipponbare reference genome using next generation sequence and optical map data. Rice (N Y). 2013;6(1):4. pmid:24280374
View Article
PubMed/NCBI
Google Scholar

[72] View Article

[73] PubMed/NCBI

[74] Google Scholar

[ref20] 20. Barrett T, Troup DB, Wilhite SE, Ledoux P, Rudnev D, Evangelista C, et al. NCBI GEO: archive for high-throughput functional genomic data. Nucleic Acids Res. 2009;37(Database issue):D885–90. pmid:18940857
View Article
PubMed/NCBI
Google Scholar

[76] View Article

[77] PubMed/NCBI

[78] Google Scholar

[ref21] 21. Liu J, Liu X, Zhang S, Liang S, Luan W, Ma X. TarDB: an online database for plant miRNA targets and miRNA-triggered phased siRNAs. BMC Genomics. 2021;22(1):348. pmid:33985427
View Article
PubMed/NCBI
Google Scholar

[80] View Article

[81] PubMed/NCBI

[82] Google Scholar

[ref22] 22. Kozomara A, Birgaoanu M, Griffiths-Jones S. miRBase: from microRNA sequences to function. Nucleic Acids Res. 2019;47(D1):D155–62. pmid:30423142
View Article
PubMed/NCBI
Google Scholar

[84] View Article

[85] PubMed/NCBI

[86] Google Scholar

[ref23] 23. Zhang Y. miRU: an automated plant miRNA target prediction server. Nucleic Acids Res. 2005;33(Web Server issue):W701-4. pmid:15980567
View Article
PubMed/NCBI
Google Scholar

[88] View Article

[89] PubMed/NCBI

[90] Google Scholar

[ref24] 24. Dai X, Zhuang Z, Zhao PX. psRNATarget: a plant small RNA target analysis server (2017 release). Nucleic Acids Research. 2018;46:W49–54.
View Article
Google Scholar

[92] View Article

[93] Google Scholar

[ref25] 25. Fei Y, Wang R, Li H, Liu S, Zhang H, Huang J. DPMIND: degradome-based plant miRNA-target interaction and network database. Bioinformatics. 2018;34(9):1618–20. pmid:29280990
View Article
PubMed/NCBI
Google Scholar

[95] View Article

[96] PubMed/NCBI

[97] Google Scholar

[ref26] 26. Fard EM, Moradi S, Salekdeh NN, Bakhshi B, Ghaffari MR, Zeinalabedini M, et al. Plant isomiRs: origins, biogenesis, and biological functions. Genomics. 2020;112(5):3382–95. pmid:32561347
View Article
PubMed/NCBI
Google Scholar

[99] View Article

[100] PubMed/NCBI

[101] Google Scholar

[ref27] 27. Yang JH, Han SJ, Yoon EK, Lee WS. “Evidence of an auxin signal pathway, microRNA167-ARF8-GH3, and its response to exogenous auxin in cultured rice cells”. Nucleic Acids Res. 2006;34(6):1892–9. pmid:16598073
View Article
PubMed/NCBI
Google Scholar

[103] View Article

[104] PubMed/NCBI

[105] Google Scholar

[ref28] 28. Yong ZhuHe WangViswanathan ChandranZhao Z-X, Feng Q, Cao X-L, Zhu Y, Wang H, Chandran V, et al. Osa-miR167d facilitates infection of Magnaporthe oryzae in rice. J Integr Plant Biol. 2020;62(5):702–15. pmid:31001874
View Article
PubMed/NCBI
Google Scholar

[107] View Article

[108] PubMed/NCBI

[109] Google Scholar

[ref29] 29. Chen H-M, Chen L-T, Patel K, Li Y-H, Baulcombe DC, Wu S-H. 22-Nucleotide RNAs trigger secondary siRNA biogenesis in plants. Proc Natl Acad Sci U S A. 2010;107(34):15269–74. pmid:20643946
View Article
PubMed/NCBI
Google Scholar

[111] View Article

[112] PubMed/NCBI

[113] Google Scholar

[ref30] 30. Li S, Le B, Ma X, Li S, You C, Yu Y, et al. Biogenesis of phased siRNAs on membrane-bound polysomes in Arabidopsis. Elife. 2016;5:e22750. pmid:27938667
View Article
PubMed/NCBI
Google Scholar

[115] View Article

[116] PubMed/NCBI

[117] Google Scholar

[ref31] 31. Moritoh S, Eun C, Ono A, Asao H, Okano Y, Yamaguchi K, et al. Targeted disruption of an orthologue of DOMAINS REARRANGED METHYLASE 2, OsDRM2, impairs the growth of rice plants by abnormal DNA methylation. Plant J. 2012;71(1):85–98. pmid:22380881
View Article
PubMed/NCBI
Google Scholar

[119] View Article

[120] PubMed/NCBI

[121] Google Scholar

[ref32] 32. Shao Y, Zhou H-Z, Wu Y, Zhang H, Lin J, Jiang X, et al. OsSPL3, an SBP-domain protein, regulates crown root development in rice. Plant Cell. 2019;31(6):1257–75. pmid:30940685
View Article
PubMed/NCBI
Google Scholar

[123] View Article

[124] PubMed/NCBI

[125] Google Scholar

Figures

Abstract

Introduction

Materials and methods

Data sources

Prediction of miRNA→target binding sites and identification of specific cleavage signals in target genes

Co-regulatory siRNA identification

Screening and verification of the results

Results and discussion

Discovery of anomalies in the rice “miRNA→target” database

Co-regulatory siRNAs identified by sRNATargetDigger in rice

Conclusions

Supporting information

S1 Fig. Target plots (t-plots) of forty one target genes which did not find cleavage signals in the middle of the miRNA binding sites by degradome.

S1 Table. Sequencing library information of sRNA-seq and degradome data for rice miRNA-target re-mining.

S2 Table. miRNAs and co-regulatory siRNAs identified in seedling.

S3 Table. miRNAs and co-regulatory siRNAs identified in panicle.

S4 Table. miRNA-target pairs are not found by sRNATargetDigger.

S5 Table. Validation of the Newly discovered miRNA-target pairs in rice by psRNATarget.

S6 Table. Validation of the co-regulatory siRNA-target pairs by psRNATarget.

Acknowledgments

References