Genome‐wide identification and functional profile analysis of long non‐coding RNAs in Avicennia marina

Avicennia marina, known for its remarkable adaptability to the challenging coastal environment, including high salinity, tide, and anaerobic soils, holds pivotal functions in safeguarding the coastal ecosystem. Long non‐coding RNAs (lncRNAs) have emerged as significant players in various natural processes of plants such as development. However, lncRNAs in A. marina remain largely unknown and uncharacterized. Here, we employed the transcriptome datasets from multiple tissues, such as root, leaf, and seed, to detect and characterize the lncRNAs of A. marina. Analyzing synthetically, we finally identified 6333 lncRNAs in the A. marina. These lncRNAs exhibited distinct features compared to messenger RNAs, including larger exons, lower guanine‐cytosine contents, lower expression levels, and higher tissue specificities. Moreover, we identified thousands of tissue‐specific lncRNAs across the examined tissues and further found that these tissue‐specific lncRNAs were significantly enriched in biological processes related to the major functions of their corresponding tissues. For instance, leaf‐specific lncRNAs showed prominent enrichment in photosynthesis, oxidation–reduction processes, and light harvesting. By providing a comprehensive dataset and functional annotations for A. marina lncRNAs, this study offers a valuable overview of lncRNAs in A. marina and lays the fundamental foundation for further functional exploring of them.


INTRODUCTION
More than 90% of eukaryotic genomes are transcribed (Ariel et al., 2015).However, only approximately 2% of these transcripts get translated, the majority are non-coding RNAs (ncRNAs) possessing little to no discernable protein-coding ability (Chekanova et al., 2007;Kapranov et al., 2007;Wilusz et al., 2009).ncRNAs can be classified into different types according to the nucleotide (nt) lengths (Yu et al., 2019).Long non-coding RNAs (lncRNAs) are RNA transcripts longer than 200 nt, without coding capacity, and constitute one of the most extensive categories of ncRNAs (Charles Richard & Eichhorn, 2018;Derrien et al., 2012;Quinn & Chang, 2016).Mounting evidence has substantiated the significance of lncRNAs as vital regulators participating in a wide array of biological processes as development and environmental responses (Ariel et al., 2015;Wierzbicki, 2012).Within animal kingdom, the functions and regulation mechanisms of lncRNAs have been well studied (Anastasiadou et al., 2018;Batista & Chang, 2013;Bond et al., 2009;Fatica & Bozzoni, 2014;Zhou et al., 2020), while the genome-wide identification and functional explorations of plant lncRNAs are far behind and less comprehensive.In recent years, benefiting from advanced sequencing technology and bioinformatics analysis, numerous plant lncRNAs have been identified and reported.Following, the biological functions of several lncRNAs have been clarified in certain plant species.In Arabidopsis, 200 transcriptome datasets were used to identify lncRNAs; as a result, more than 36,000 candidates, including natural antisense transcripts (>30,000) and long intergenic ncRNAs (lincRNAs) (>6000), were found (Jin et al., 2013;Wang, Chung, et al., 2014).In addition, HID-DEN TREASURE 1, a 236 nt length lncRNA, can promote Arabidopsis photomorphogenesis by regulating the expression of PHYTOCHROME INTERACTING FACTOR 3 and PROTOCHLOROPHYLLIDE OXIDOREDUCTASES (Wang, Fan, et al., 2014;Wang, Li, et al., 2018).The AUXIN REG-ULATED PROMOTER LOOP RNA/npc34 lncRNA, which is located upstream of PINOID (PID, coding an auxin signaling kinase) gene, positively regulates the expression of PID and ROOT HAIR DEFECTIVE 6 and then has a significant effect on root development (Moison et al., 2021).Besides, Arabidopsis lncRNAs are reported to be involved in regulating flowering (Tian Y et al., 2019) and biotic and abiotic stress responses (Liu et al., 2019;Qin et al., 2017).In maize, 1535 drought-responsive lncRNAs are identified under arid stress (Pang et al., 2019).Furthermore, over-expression of LRK (leucine-rich repeat receptor kinase) antisense intergenic RNA (LAIR) enhances grain yield by promoting the expression of LRK neighboring genes in rice (Wang, Luo et al., 2018).To date, lncRNAs have been identified and studied not only in Arabidopsis, maize, and rice (Huang et al., 2021) but also in other non-model plant species, such as rubber (Li et al., 2021;Liu et al., 2022;Wang et al., 2022;Yin et al., 2019).
Avicennia marina (Forssk.)Vierh.(Acanthaceae) is a mangrove tree with a broad distribution, which is known as the guardian of the coast and plays important ecological as well as environmental roles in protecting the water quality, maintaining coast stability, mitigating climate change, conserving biodiversity, etc. (Kuenzer et al., 2011).As the main species of the coastal mangrove ecosystem, A. marina can survive in extreme ecological environments, such as external salt concentration and heavy metal toxicity (Zeinali et al., 2017).However, the relevant molecular mechanisms implicated in regulating the development, growth, and environmental adaptions of A. marina under different conditions were still largely unknown.Most of the previous studies are mainly focused on physiology and biochemistry studies or the functional role of coding genes (Kavitha et al., 2008;Natarajan et al., 2021).In recent years, although an accumulating number of research indicates that lncRNAs are significant regulators involved in plant development and stress responses (Moison et al., 2021;Tian et al., 2023), the identification, characterization, and biological function exploration of lncRNAs in A. marina was still unreported.
To overview the quantity, types, and features, as well as pursue further biological function information about the lncRNAs of A. marina, we conducted comprehensive identification, characterization, and annotation of them with transcriptome datasets from root, pneumatophore, stem, leaf, flower, seed, etc., tissues.These results will provide valuable insights and clues to understand the functions of A. marina lncRNAs and lay theoretical foundations for further functional studies of A. marina lncRNAs.

Transcriptome assembly of long and short reads
The RNA-seq datasets were obtained from the sequence read archive (SRA) database (https://www.ncbi.nlm.nih.gov/sra) of NCBI (National Center for Biotechnology Information) (Table S1).The reference genome of A. marina was downloaded from NCBI (https://www.ncbi.nlm.nih.gov)(Natarajan et al., 2021).The FastQC (v0.11.9) was used to check the RNA sequencing quality (Andrews, 2010) and AdapterRemoval (v2.3.1) was applied to remove remnant adapter sequences from the sequencing reads (Schubert et al., 2016).For the samples only with short-read RNA sequencing data, we mapped the reads to the genome sequence of A. marina using hisat2 (v2.2.1) (Figure 1) (Kim et al., 2019).For the PacBio long reads, we followed the pipeline in Isoseq3 (https://github.com/PacificBiosciences/IsoSeq) to generate the potential isoforms and utilized the pbmm2 to map them to the reference genome.Stringtie (v2.2.1) was used to assemble the reads to potential transcripts following the read mapping for each sample (Figure 1) (Kovaka et al., 2019).For the samples with both short and long reads, we use both of them for the transcriptome assembly.We merged the transcripts from different samples into a non-redundant set of transcripts and filtered them without strand information.

Identification of lncRNAs
To provide a comprehensive dataset for lncRNAs of A. marina, we identified the lncRNAs from RNA-sequencing data generated from multiple tissues, including root, leaf, and seed.First, the sequences were filtered with the following criteria: transcripts with length >200 bp and exon number >1 were retained, and the transcripts with exon overlapped with messenger RNAs (mRNAs) were removed (Li et al., 2014;Zhang et al., 2014).Next, we screened the transcripts based on their coding potential.The coding potential was assessed by (a) the longest open reading frame (ORF) length, (b) sequence similarity or domain with the known proteins, and (c) coding score.The longest ORF was calculated by the TransDecoder software with the parameter of -m 30 & -S (https://github.com/TransDecoder/TransDecoder).The transcripts that encode less than 100 aa (amino acid) were preserved and then used for sequence alignments (Blastx, evalue = 1e-3) and domain scan (pfamscan, -e_seq = 1e-3 and -e_dom = 1e-3) to remove the transcripts containing significant homology with the known protein sequences or Pfam domains (Camacho et al., 2009;Mistry et al., 2021).To further recognize the non-coding transcripts, we utilized coding potential calculator 2 (CPC2) and RNAplonc (v1.1) to calcu-

Core Ideas
• Avicennia marina thrives in a harsh ecological environment and plays an important role in coastal ecosystem safeguarding.However, molecular mechanisms are still in its infancy.• We present an overview of lncRNAs of multiple tissues from Avicennia marina based on RNA-seq data analysis.• The features of these lncRNAs were analyzed, and their biological functions were annotated and enriched.
• The functions of identified tissue-specific lncR-NAs help elucidate the development mechanisms of Avicennia marina.• Our research will lay theoretical foundations for further functional studies of Avicennia marina long non-coding RNAs.
late their coding scores, and eliminated the coding transcripts with default criteria (Kang et al., 2017;Negri et al., 2019).The remaining transcripts were recognized as the lncRNAs of A. marina.

Classification distribution of A. marina lncRNAs
We employed the FEELnc to sort the identified lncRNAs into different categories, including intergenic lncRNAs, antisense lncRNAs, and intronic lncRNAs, and other lncRNAs based on the localization and transcriptional direction of proximal mRNAs (Wucher et al., 2017).

Chromosome distribution of lncRNAs in A. marina
To examine the chromosomal assignment of lncRNAs, we first calculated the number and density of lncRNAs for each chromosome.The lncRNAs density is calculated with the following equation:

Expression profile
In order to overview and compare the expression levels of A. marina mRNAs and lncRNAs, we used the salmon (v1.4.0) to calculate the raw read counts for each transcript (Patro et al., 2017).The raw read counts were normalized by the trimmed mean of the M-values (TMM) approach and converted to frag-ments per kilobase of sequence per million mapped reads with edgeR (v3.30.3) (Robinson et al., 2010).

Tissue-specificity analysis
We adopted the expression data from five tissues (root, leaf, stem, flower, and seed) to investigate the tissue-specificity of lncRNAs and mRNAs.The fractional expression for each RNA (lncRNA or mRNA) in a given tissue was the proportion of its expression against the total expressions of this RNA across all five tested tissues (Ding et al., 2018).The maximum fractional expression for each lncRNA is defined as its tissue-specific score across the examined tissues.The criterion of tissue-specific score >0.6 is employed to identify the tissue-specific lncRNAs (Wang et al., 2022).

Functional analysis for lncRNAs
Generally, lncRNA is tightly correlated with its potential targets in the expression level, the function of lncRNA can be predicted by its co-expression mRNAs.Thus, we conducted the functional analysis for the A. marina lncRNAs depending on the functional enrichment of their co-expression mRNAs.We performed the Pearson correlation analysis between lncR-NAs and mRNAs and employed the criteria of Pearson correlation >0.5 and adjusted p-value < 0.05 to find the co-expression mRNAs for each lncRNA.The gene ontology (biological process category) for the co-expression mRNAs of lncRNAs was annotated by the software interproscan (v5.48-83.0)with the parameter of -pa -goterms (Jones et al., 2014).We conducted the enrichment analysis for the co-expression mRNAs of each lncRNA to carry out function annotations using hypergeometric test (Kolberg et al., 2020).The p-value adjusted by Benjamini and Hochberg and a cutoff of adjusted p-value < 0.1 was used to determine the enrichment significance (Benjamini & Hochberg, 1995).Only the significantly enriched gene ontologies were used to annotate the function of lncRNAs.The functional enrichment analysis for the tissuespecific lncRNAs is also operated by the hypergeometric test with the same criteria.

Identification of lncRNAs in A. marina
To provide a comprehensive profile of lncRNAs in A. marina, we employed the RNA-seq datasets that included both long reads and short reads sequencing data from multiple tissues to identify the lncRNAs (Figure 1 and Table S1).The reads were mapped to reference genome of A. marina and subsequently assembled into transcripts.Though merging the transcripts from different samples, we obtained a total of 123,183 transcripts.We identified lncRNAs in A. marina through a two-step process.Initially, we filtered the transcripts based on their lengths, exon numbers and exon overlapped with mRNA, and obtained 11,443 residual transcripts.Subsequently, we conducted a screening process based on coding potential, which was assessed by ORF length, Blastx, CPC2, etc.As a result, 6333 lncRNAs were finally identified in A. marina.

Distribution and category of A. marina lncRNAs
At the chromosome level, we observed that chromosomes 1, 2, and 3 contained more than 300 lncRNAs each (Figure 2A).Notably, both lncRNAs and mRNAs exhibited a preferred localization on these particular chromosomes (Figure 2A and Figure S1A).The number of lncRNAs and mRNAs on each chromosome may be influenced by chromosome's length.To account for this potential bias, we normalized the lncRNA and mRNA counts by scaling the chromosome length to 10 6 .Notably, even after this normalization, chromosomes 1, 2, and 3 were still ranking the top chromosomes in terms of lncRNA and mRNA distribution (Figure S1B,C).To further examine the detailed distribution of A. marina lncRNAs on each chromosome, we calculated the number of lncRNAs within 20 kb sliding window across each chromosome.Our analysis revealed that the distribution of lncRNAs on the same chromosome differed to some extent from that of mRNAs (Figure 2B).For instance, mRNAs exhibited relatively less distributions in the middle region of the chromosomes, whereas lncRNAs did not (Figure 2B).Furthermore, considering the genomic location relationship between lncRNAs and mRNAs, we categorized the lncRNAs into different categories, including intergenic, antisense, and intronic lncRNAs.Notably, more than 4000 lncRNAs fell into the category of intergenic lncRNAs (Figure S1D).Approximately 30% and 3% of the lncRNAs were identified as antisense and intronic lncRNAs, respectively (Figure S1D).

Distinct features of lncRNAs in comparison with mRNAs
To investigate whether lncRNAs possess particular features, we examined and compared exon size, exon number, transcript length, and GC content between lncRNAs and mRNAs.The results indicated that lncRNAs exhibit distinct characteristics, such as larger exon sizes and longer introns than mRNAs (Figure 3A,B).These features are consistent with the characteristics of lncRNAs in other plants, such as Hevea brasiliensis and Solanum lycopersicum (Wang et al., 2022, Wang, Zhao, et al., 2018).Additionally, we observed that lncRNAs in A. marina tend to contain fewer exons than mRNAs (Figure 3C).As approximately 15% mRNAs contain >10 exons, only ∼3% lncRNAs possess over 10 exons, suggesting that lncRNAs in A. marina commonly have relatively fewer exons compared to mRNAs.Furthermore, we found that lncRNAs in A. marina have shorter sequence lengths than mRNAs (Figure 3D), which is consistent with the features of lncRNAs in soybean and Camellia sinensis (Lin et al., 2020;Wan et al., 2020).
In general, lncRNAs are known to possess lower GC content than mRNAs (Haerty & Ponting, 2015).Consistently, we found that lncRNAs in A. marina also exhibit lower GC content compared to mRNAs (Figure 3E).It suggests that the lower GC content in lncRNAs may potentially affect the coding potential of their sequences.Moreover, we observed that lncRNAs exhibit lower expression levels than mRNAs across the examined tissues (Figure 3F), suggesting that GC content may also associate with the potential of DNA transcription in A. marina.

Tissue specificity of lncRNAs and mRNAs
To assess the tissue specificities of both lncRNAs and mRNAs, we utilized expression data from various tissues to calculate the tissue-specific scores for them.
Using a criterion of tissue-specific score >0.6, we identified more than a thousand of tissue-specific lncRNAs from the examined tissues (Figure S2A).Most of these tissue-specific lncRNAs were primarily identified in the leaf and seed tissues, indicating their important functions in seed and leaf development, growth, as well as physiology (Figure S2A,B).Additionally, around 200 lncRNAs were respectively identified as tissue-specific in flower and root tissues, and only about 100 members were found to be the stem tissue-specific lncRNAs (Figure S2A).
Moreover, as shown in Figure 4A, we found that lncRNAs exhibited significantly higher tissue-specific scores compared The Plant Genome with mRNAs.Meanwhile, regardless of the threshold used to define tissue-specific RNAs, we consistently observed a relatively higher percentage of tissue-specific lncRNAs than mRNAs (Figure 4B).This consistent tendency demonstrated that lncRNAs possess higher tissue specificities in A. marina, indicating their potential roles in regulating the development, growth, and biological function processes of the particular tissues.

Tissue-specific lncRNAs exhibit tissue-specific functional relevance
Although the aforementioned analysis has successfully identified around a thousand of tissue-specific lncRNAs in A. marina, their functional roles remain unclear.To unravel the potential functions of tissue-specific lncRNAs, we employed functional annotation of co-expression mRNAs to predict their biological functions.

Oxidation-reduction process
The functional analysis revealed that flower-specific lncR-NAs are significantly enriched in the biological processes related to the cell wall modification and organization (Figure 5A and Table S2).While the root tissue-specific lncRNAs were predominantly enriched in the nitric oxide biosynthetic process (Figure 5B and Table S3).In the case of seed-specific lncRNAs, they were found to be associated with transcriptional regulation and cell wall biogenesis processes The Plant Genome  (Figure 5C and Table S4), both of which are connected to the preparation for the seed development.Differently, the stemspecific lncRNAs were enriched in the biological process of growth regulation (Figure 5D and Table S5), suggesting their potential roles in regulating the developmental and living processes of stems.
In comparison to other tissues, leaf-specific lncRNAs are enriched in a greater number of biological processes (Figure 6A and Table S6).Remarkably, the topmost enriched pathways for leaf-specific lncRNAs include photosynthesis, oxidation-reduction processes, and light harvesting (Figure 6A), which are highly relevant to the main function of leaf tissues.Notably, dozens of leaf-specific lncRNAs, including Amarina.29041.3 and Amarina.1520.3, are involved in these biological processes (Figure 6B), emphasizing the contribution of leaf-specific lncRNAs to the primary function of leaves.Furthermore, we discovered that leaf-specific lncR-NAs, such as Amarina.29041.3 and Amarina.1520.3,not only regulate the major functions of leaves but also participate in other biological processes, such as translation, defense responses, and response to gibberellin (Figure S3 and Table S7), indicating their crucial functions in regulating leaf development, growth, and stress responses in A. marina.

Functional analysis of non-tissue-specific lncRNAs
Non-tissue-specific lncRNAs also constitute an important component of the A. marina lncRNAs.Except for the tissue-specific lncRNAs, we concurrently conducted functional analysis for the non-tissue-specific lncRNAs.Our findings revealed that the majority of non-tissue-specific lncRNAs are referred to biological processes, such as photosynthesis, translation, glycerol ether metabolic process, chloroplast ribulose bisphosphate carboxylase complex assembly, oxidation-reduction process, and response to aluminum ion, acidic pH, as well as nitrate (Figure 7).This suggests that non-tissue-specific lncRNAs are likely to play significant roles in different biological processes involved in both plant growth and environment adaptions across various tissues in A. marina.

DISCUSSION
LncRNAs act as a vital player in diverse biological processes, such as tissue/ organ development, stress responses, aging, and diseases (Statello et al., 2021;Wang et al., 2023;Zhou et al., 2020).Due to the well-developed high-throughput sequencing and bioinformatics analysis techniques, thousands of lncRNAs have been identified in various species.Nevertheless, the identification and functional analysis of lncRNAs in A. marina have not been conducted yet.
In this study, we integrated the transcriptomic data from multiple tissues of A. marina and identified 6333 lncRNAs, which were unevenly distributed on different chromosomes.But for chromosomes 1, 2 and 3, both lncRNAs and mRNAs exhibited a preferred localization on these particular chromosomes.Among these lncRNAs, more than 4000 of them were intergenic lncRNAs, which is consistent with previous research (Wang, Zhao, et al., 2018).The chromosomal distribution of lncRNAs usually showed close relationships with coding genes.These results suggesting that there were more coding genes on chromosome 1, 2 and 3, thus more lncRNAs were found on them.This potentially indicated the importance of these chromosomes during the evolution, growth, and development of A. marina.
Besides, these identified lncRNAs presented remarkable higher tissue specificities than mRNAs, which maintain highly consistent with previous research in other plants (Wang et al., 2022).Therefore, hundreds of tissue-specific lncRNAs were identified from the examined tissues.The tissue-specific lncRNAs were likely to play important roles in the development and growth of related tissues.These findings suggesting that lncRNAs exhibited more tissue specific than mRNAs, potentially indicating their significant functions in regulating tissues' development and maintaining their special biological functions (Ma et al., 2021;Zhang et al., 2014).Simultaneously, we found that flower organ-specific lncRNAs showed an enrichment in cell wall modification and organization processes.Cell wall modification plays a crucial role in the formation of floral organs (Cruz-Valderrama et al., 2021), indicating that flower-specific lncRNAs may play a regulatory role in floral organ development.
Additionally, as the underground part of plant, root can hold the plant as well as absorb water and water-soluble minerals.Meanwhile, root can synthesize organic products, such as amino acids, organic nitrogen, and hormones, which are essential for physiological processes of plant.Besides, several A. marina root tissue-specific lncRNAs were enriched in nitric oxide biosynthetic process, implying their significant functions in root development and stress responses (Ageeva-Kieferle et al., 2021).Furthermore, we found several leaf-specific lncRNAs significantly enriched in leaf interrelated biological processes.Especially for Amarina.29041.3 and Amarina.1520.3, as the leaf tissue-specific lncRNAs, they were significantly enriched in leaf function related processes, such as photosynthesis, photosynthesis light harvesting, and oxidation-reduction process.Interestingly, Amarina.1520.3 is the potential orthologous lncRNA of CNT2072811 (CANTATAdb ID), which is defined as a confident lncRNA and enriched in response to abiotic and biotic stimulus, response to light stimulus and photosynthesis processes in Populus trichocarpa (Szcześniak et al., 2019).In summary, our comprehensive identification, characterization, and functional analysis of the lncRNAs provide theoretical foundation for further studies of lncRNAs in A. marina as well as in other plant species.

AU T H O R C O N T R I B U T I O N S
Lingling Wang: Funding acquisition; project administration; writing-original draft; writing-review and editing.Zixin Yuan: Data curation; resources.Jingyi Wang: Formal analysis; validation.Yali Guan: Investigation; project administration; supervision.

C O N F L I C T O F I N T E R E S T S T A T E M E N T
The authors declare no conflicts of interest.

D A T A AVA I L A B I L I T Y S T A T E M E N T
The data are available in the Supporting Information.

where
ChrN indicates the chromosome N. Subsequently, we split each chromosome into 200-kbwidth bins from the start site to the end with a 200 kb step, then counted the number of lncRNAs and mRNA in each bin, separately.length > 200 bp b.Exon number > 1 c.Remove transcript with exon overlap with the exon of mRNA a. Longest ORF < 100 aa b.Blastx and Pfam c. CPC2 and RNAplonc long non-coding RNAs F I G U R E 1 The pipeline to identify the long non-coding RNAs (lncRNAs) in Avicennia marina.This pipeline demonstrates the process from transcriptome assemble to the lncRNA identification.The short blue line represents the short reads RNA-seq data, while the long sandybrown line indicates the long reads RNA-seq data.ORF, open reading frame; mRNA, messenger RNAs; CPC2, coding potential calculator 2.
Long non-coding RNAs' (lncRNAs) distribution across the chromosomes.(A) Number of lncRNAs across the chromosomes.(B) The distribution of lncRNAs and messenger RNAs (mRNAs) on each chromosome.The out layer represents the chromosomes.The purple, blue, pink, and orange colors indicate the lncRNAs on + strand, mRNAs on + strand, lncRNA on − strand, and mRNA on − strand, respectively.
Characteristics of long non-coding RNAs (lncRNAs) in Avicennia marina.The cumulative probability of (A) exon size and (B) intron size of lncRNAs and messenger RNAs (mRNAs).p-Value was calculated by Kolmogorov-Smirnov test.(C) Percentage of lncRNAs and mRNAs with different exon numbers.(D) The transcript length of lnRNAs and mRNAs.The statistical significance was performed by the Mann-Whitney test (*p < 0.05).(E) Density distribution of guanine and cytosine (GC) content (%) in lncRNAs and mRNAs.(F) The expression level of lncRNAs and mRNAs; Y axis indicates the expression level (log2(FPKM), where FPKM is fragments per kilobase of sequence per million mapped reads).The statistical significance was conducted by the Mann-Whitney test (*p < 0.05).
Tissue specificity of the long non-coding RNAs (lncRNAs) and messenger RNAs (mRNAs).(A) Cumulative probability of tissue-specific score of lncRNAs and mRNAs.(B) Percentage of tissue-specific lncRNAs and mRNAs under varying thresholds of the tissue-specific score.Functional enrichment for the tissue-specific long non-coding RNAs (lncRNAs).The significant enrichment BPs (biological processes) for tissue-specific lncRNAs in (A) flower, (B) root, (C) seed, and (D) stem.FDR, false discovery rate.
Photosynthetic electron transport in photosystem I Photosystem II assembly Chlorophyll biosynthetic process Autophagosome assembly Autophagy tRNA aminoacylation Glycerol ether metabolic process Aromatic amino acid family biosynthetic process Photosynthesis, light harvesting Oxidation−reduction process Photosynthesis photosynthesis photosynthesis, light harvesting translation oxidation−reduction process glycerol ether metabolic process chloroplast ribulose bisphosphate carboxylase complex assembly aromatic amino acid family biosynthetic process photosystem II assembly trichome morphogenesis glycine decarboxylation via glycine cleavage system ATP synthesis coupled proton transport cellular aromatic compound metabolic process trichome differentiation protein folding protein catabolic process cell redox homeostasis chlorophyll biosynthetic process xylan biosynthetic process abscisic acid−activated signaling pathway RNA processing proteolysis involved in cellular protein catabolic process double−strand break repair via homologous recombination response to aluminum ion response to acidic pH photosynthetic electron transport in photosystem I response to nitrate nitrate transport iron−sulfur cluster assembly L−methionine salvage from methylthioadenosine nucleus organization ubiquitin−dependent protein catabolic process translational initiation protein arginylation pigment biosynthetic process carboxylic acid metabolic process cell wall modification tRNA aminoacylation multicellular organism development response to high light intensity sulfate transport RNA methylation response to gibberellin negative regulation of transcription, DNA−templated mRNA splicing, via spliceosome transmembrane transport arginyl−tRNA aminoacylation cell wall biogenesis malate metabolic process systemic acquired resistance regulation of cyclin−dependent protein serine/threonine kinase activity glucose metabolic process ribosomal small subunit biogenesis copper ion transmembrane transport phloem development xyloglucan metabolic process defense response regulation of cell cycle potassium ion transmembrane transport intracellular protein transport mRNA processing carbohydrate metabolic process hydrogen peroxide catabolic process cortical microtubule organization mitotic sister chromatid cohesion lipid transport translational termination polysaccharide catabolic process fatty acid metabolic process intracellular transport ATP metabolic process endocytosis regulation of flower development response to oxidative stress plant−type cell wall organization protein phosphorylation histone modification ribosomal large subunit biogenesis nicotianamine biosynthetic process transcription elongation from RNA polymerase II promoter isoprenoid biosynthetic process nitric oxide biosynthetic process nitrate assimilation response to wounding regulation of DNA endoreduplication root development regulation of amino acid export vacuolar transport tRNA processing regulation of transcription, DNA−templated protein ubiquitination terpenoid biosynthetic process translational elongation protein ufmylation DNA replication initiation secondary shoot formation sterol biosynthetic process protein glycosylation SRP−dependent cotranslational protein targeting to membrane tricarboxylic acid cycle tRNA wobble uridine modification gene silencing by RNA steroid biosynthetic process cellular glucan metabolic process phosphatidylcholine metabolic process ammonium transport nucleosome assembly L−serine biosynthetic process ammonium transmembrane transport galactose metabolic process nucleotide−excision repair tRNA modification photosystem II stabilization nuclear−transcribed mRNA catabolic process, deadenylation−dependent decay microtubule−based process phospholipid transport circadian rhythm protein import into mitochondrial matrix autophagosome assembly response to desiccation jasmonic acid biosynthetic process triterpenoid biosynthetic process cysteine biosynthetic process from serine positive regulation of translational termination lipid metabolic process DNA replication radial pattern formation proteolysis glutamine biosynthetic process photosystem I assembly cell−cell signaling involved in cell fate commitment histone acetylation cellular manganese ion homeostasis inositol phosphate dephosphorylation endoplasmic reticulum to Golgi vesicle−mediated transport root hair cell development autophagy malate transport acetyl−CoA biosynthetic process from pyruvate glycine biosynthetic process from serine tetrahydrofolate interconversion RNA catabolic process DNA duplex unwinding L−phenylalanine biosynthetic process regulation of circadian rhythm organic substance metabolic process regulation of translation vesicle−mediated transport chromatin remodeling positive regulation of translational elongation purine nucleotide biosynthetic process D−amino acid catabolic process rRNA processing cellular amino acid metabolic process electron transport chain regulation of gene expression TOR signaling carbon utilization response to light stimulus dTDP biosynthetic process response to heat transcription initiation from RNA polymerase II promoter mRNA export from nucleus leaf formation acetyl−CoA metabolic process farnesyl diphosphate biosynthetic process, mevalonate pathway deadenylation−dependent decapping of nuclear−transcribed mRNA cellular glucose homeostasis protein retention in ER lumen nuclear−transcribed mRNA catabolic process, exonucleolytic, 3'−5' systemic acquired resistance, salicylic acid mediated signaling pathway regulation of jasmonic acid mediated signaling pathway regulation of salicylic acid mediated signaling pathway DNA metabolic process response to UV−B cytidine to uridine editing protein refolding histidine biosynthetic process DNA repair pseudouridine synthesis cellular protein modification process protein import into nucleus sucrose transport adenylate cyclase−modulating G protein−coupled receptor signaling pathway histone deacetylation exon−exon junction complex disassembly glutathione metabolic process actin filament depolymerization protein peptidyl−prolyl isomerization regulation of root meristem growth response to red or far red light chloride transport aromatic amino acid family metabolic process DNA−templated transcription, initiation alanyl−tRNA aminoacylation tRNA methylation vacuolar proton−transporting V−type ATPase complex assembly intra−Golgi vesicle−mediated transport response to photooxidative stress asymmetric cell division chorismate metabolic process positive regulation of transcription by RNA polymerase II organelle organization recognition of pollen sulfur compound metabolic process DNA topological change vitamin B6 biosynthetic process pyridoxal phosphate biosynthetic process carbohydrate phosphorylation cellular metabolic process leaf development signal transduction protein transport regulation of protein catabolic process blue light signaling pathway ribosome biogenesis cell surface receptor signaling pathway Golgi vesicle transport