Differential subgenome expression underlies biomass accumulation in allotetraploid Pennisetum giganteum

Xing, Longsheng; Wang, Meijia; He, Qiang; Zhang, Hongyu; Liang, Hanfei; Zhou, Qinghong; Liu, Yu; Liu, Ze; Wang, Yu; Du, Cailian; Xiao, Yao; Liu, Jianan; Li, Wei; Liu, Guixia; Du, Huilong

doi:10.1186/s12915-023-01643-w

Research article
Open access
Published: 21 July 2023

Differential subgenome expression underlies biomass accumulation in allotetraploid Pennisetum giganteum

Longsheng Xing^1,2^na1,
Meijia Wang¹^na1,
Qiang He^1,2,
Hongyu Zhang¹,
Hanfei Liang¹,
Qinghong Zhou¹,
Yu Liu¹,
Ze Liu¹,
Yu Wang¹,
Cailian Du¹,
Yao Xiao¹,
Jianan Liu¹,
Wei Li^1,2,
Guixia Liu^1,2 &
…
Huilong Du ORCID: orcid.org/0000-0001-7755-6611^1,2

BMC Biology volume 21, Article number: 161 (2023) Cite this article

1687 Accesses
1 Altmetric
Metrics details

Abstract

Background

Pennisetum giganteum (AABB, 2n = 4x = 28) is a C4 plant in the genus Pennisetum with origin in Africa but currently also grown in Asia and America. It is a crucial forage and potential energy grass with significant advantages in yield, stress resistance, and environmental adaptation. However, the mechanisms underlying these advantageous traits remain largely unexplored. Here, we present a high-quality genome assembly of the allotetraploid P. giganteum aiming at providing insights into biomass accumulation.

Results

Our assembly has a genome size 2.03 Gb and contig N50 of 88.47 Mb that was further divided into A and B subgenomes. Genome evolution analysis revealed the evolutionary relationships across the Panicoideae subfamily lineages and identified numerous genome rearrangements that had occurred in P. giganteum. Comparative genomic analysis showed functional differentiation between the subgenomes. Transcriptome analysis found no subgenome dominance at the overall gene expression level; however, differentially expressed homoeologous genes and homoeolog-specific expressed genes between the two subgenomes were identified, suggesting that complementary effects between the A and B subgenomes contributed to biomass accumulation of P. giganteum. Besides, C4 photosynthesis-related genes were significantly expanded in P. giganteum and their sequences and expression patterns were highly conserved between the two subgenomes, implying that both subgenomes contributed greatly and almost equally to the highly efficient C4 photosynthesis in P. giganteum. We also identified key candidate genes in the C4 photosynthesis pathway that showed sustained high expression across all developmental stages of P. giganteum.

Conclusions

Our study provides important genomic resources for elucidating the genetic basis of advantageous traits in polyploid species, and facilitates further functional genomics research and genetic improvement of P. giganteum.

Background

Grasses provide most of the calories in human diets and are the most abundant but still not fully exploited sustainable resource [1,2,3]. Pennisetum species are widely distributed worldwide and many of them are C4 plants, which generally exhibit rapid growth, high arid tolerance, and high biomass production [4, 5]. Pennisetum giganteum is native to Africa and is well known for its huge biomass and production [6, 7]. P. giganteum can grow in tropical, subtropical, and temperate regions and has been cultivated in more than 30 provinces in China and up to 80 countries worldwide [8]. P. giganteum is a typical C4 plant with a solar energy conversion rate that is 4–7.5 times that of broad-leaved trees; they are generally grown to a height of 3–5 m and the maximum height can be up to 7.08 m [9]. The annual yield of fresh P. giganteum grass was reported to be up to 200–400 tons per hectare, with crude protein content of 10.8% after 4 weeks of growth. The cellulose-rich P. giganteum can be used for biomass power generation and for producing nano-cellulose, which is a raw material of non-wood fiber that is suitable for pulping and papermaking [7, 10]. Therefore, P. giganteum is regarded as a high-yield and high-quality forage grass and an extremely important raw material for bioenergy generation. The synthesis of cellulose is mainly mediated by cellulose synthase (CesA) complexes (CSCs) that is located on the plasma membrane [11, 12]. In monocots, the CesA gene superfamily are mainly composed of CesA and seven cellulose synthase-like (Csl) clades, such as CslA, CslC, CslD, CslE, CslF, CslH, and CslJ [12,13,14]. It was widely accepted that the CesA genes are mainly participating in cellulose synthesis and the Csl genes mainly encode enzymes that synthesize hemicellulose [14, 15]. Due to their essential roles in cell wall synthesis, the members of this superfamily have been extensively investigated since the discovery [11, 13, 14, 16,17,18]. However, the systematic identification and expression profiling of CesA/Csl genes in P. giganteum are still unexplored.

P. giganteum is an allotetraploid plant with many excellent characteristics of typical polyploid plants, such as huge size, high yield, and strong stress resistance (Fig. 1a). Previous studies have shown that polyploidization is an important driving force for increasing biomass and improving stress tolerance of plants during evolution [19,20,21]. However, the underlying mechanism of these traits in polyploid plants is still not fully elucidated. Besides, the nutrient content such as crude protein and cellulose was found to vary greatly in different growth stages of P. giganteum, but the underlying mechanism is also largely unexplored [8]. For example, the crude protein content in P. giganteum (height = 50 cm) at 4 weeks was 10.8%, but only 5.9% at 12 weeks [8]. Therefore, in-depth study of gene expression between the P. giganteum A and B subgenomes during different developmental stages will provide better insights into the potential mechanisms underlying biomass and nutrient content accumulation in polyploid plants.

So far, few transcriptome datasets on P. giganteum have been released; one reported the response of P. giganteum to chilling cold [22]. Only a few genome assemblies of Pennisetum species are available, such as pearl millet and elephant grass [23,24,25]. This lack of omics data has seriously hampered comprehensive evolutionary and comparative genomics studies of Pennisetum species. Therefore, to obtain more high-quality genome assemblies, it is necessary to expand the genetic resources of Pennisetum and elucidate the genetic basis of many important agronomic traits in genus Pennisetum. C4 plants generally have high photosynthetic efficiency, and the genetic basis of this is largely resolved [26, 27]. Additionally, many studies have focused on the evolution of C4 photosynthesis and the regulation of gene expression that are required for C4 photosynthesis previously [28,29,30]. However, the regulation and expression patterns of photosynthesis-related genes between subgenomes in polyploid C4 plants have rarely been reported.

Here, we present a high-quality genome assembly of the allotetraploid P. giganteum that was divided into A and B subgenomes. We examined the evolutionary relationship and the history of genomes across the Panicoideae lineages, and discovered the functional differentiation and homoeolog-biased expression between the A and B subgenomes by combining comparative genomics analysis with transcriptome meta-analysis. The results implied that the hybrid used the favorable copies of the genes and maintained their high expression. The comprehensive transcriptome analysis of P. giganteum leaf and stem at seven developmental stages revealed the genetic basis of biomass accumulation in this C4 plant. We also identified key candidate genes at each step of the C4 photosynthesis pathway that were consistently and highly expressed at all seven developmental stages. This genomic resource provides insights into the genetic basis of the huge biomass generated in polyploid plant species.

Results

High-quality genome assembly and separate A and B subgenomes of P. giganteum

We obtained a total of 65.25 Gb PacBio circular consensus sequencing (CCS) data and 284.94 Gb of high-throughput chromosome conformation capture (Hi-C) sequencing data and used them to generate a high-quality genome assembly of P. giganteum (Table S1). De novo assembly of CCS data yielded an initial contig-level assembly. We then used the Juicer and 3D-DNA algorithms to anchor and correct the assembled contigs by integrating the Hi-C data. The final assembly contained 14 chromosome pseudomolecules with 99.17% of the sequence anchored onto the chromosomes with contig N50 of 88.47 Mb. This assembly is one of the highest continuity grass genome assemblies currently available (Fig. 1b; Additional file 2: Figure S1; Table 1; Additional file 1: Tables S3 and S5). The total length of the assembly spanned 2.03 Gb, which is consistent with the estimated genome size from k-mer frequency analysis based on 115.01 Gb MGI-Seq short reads (Table 1; Additional file 2: Figure S2; Additional file 1: Tables S1–S3). Noticeably, four chromosomes consisted of one single contig, and the remaining 10 chromosomes harbored only 1–5 gaps (Additional file 1: Table S4), indicating the genome assembly had high contiguity. Benchmarking Universal Single-Copy Orthologs (BUSCO) evaluation showed that 98.50% of the genes identified in the P. giganteum genome were complete, and the average LTR assembly index score was 19.32 (Table 1; Additional file 1: Table S6), indicating that the genome assembly was close to the gold standard [31]. The average of RNA sequencing (RNA-seq) mapping rates was 96.09%, further confirming the completeness of the genome assembly (Additional file 1: Table S7). Additionally, the base accuracy evaluation by Illumina short reads indicated > 99.999% accuracy rate of the P. giganteum genome (Additional file 1: Table S8). Together, these results confirmed the high quality of our P. giganteum genome assembly.

Table 1 Statistics of the Pennisetum giganteum genome assembly and annotation

Full size table

Overall, 66.12% of P. giganteum genome sequence was identified as transposable elements, and long terminal repeat retrotransposons (LTR-RTs) were the most abundant class of repetitive sequences, accounting for 52.14% of the P. giganteum genome. Among the LTR-RTs, Gypsy elements (31.45%) were the most abundant, followed by Copia elements (11.50%) (Fig. 1c; Table 1; Additional file 1: Tables S9–S10). A total of 26,679 intact LTR-RTs were identified, including 13,312 Gypsy elements (49.90%) and 8216 Copia elements (30.80%) (Additional file 1: Table S11). A total of 85,765 high-confidence protein-coding genes were predicted by combing an evidence-based method and de novo prediction (Table 1; Additional file 1: Table S12); 95.89% of these genes were functionally annotated by homologous sequence searches and with Pfam protein domains by InterProScan (Table 1; Additional file 1: Table S13). The annotated gene set covered 99.40% of the complete BUSCO genes (Additional file 1: Table S6), indicating the completeness of the gene annotation. The average gene length and average exon number were 3042 bp and 4.29, respectively (Additional file 1: Table S12), which is consistent with those in the genomes of two other Pennisetum species, elephant grass (Pennisetum purpureum) and pearl millet (Cenchrus americanus) [24].

To divide P. giganteum genome into subgenomes, resequencing reads from C. americanus (2n = 14), potential donor of the A subgenome in the genus Pennisetum [5], were mapped onto the P. giganteum genome. The mapping rates of seven chromosomes were 4.67–30.05%, whereas those of the other seven chromosomes were < 2.47% (Additional file 2: Figure S3; Additional file 1: Table S14). We also identified and analyzed the frequencies of 13-mers enriched in the P. giganteum genome and found that the enriched 13-mer species divided the 14 chromosomes into two clear groups (Additional file 2: Figure S4; Additional file 1: Table S15). Phylogenetic analysis based on the alignment of single-copy genes of homologous chromosome pairs of P. giganteum, C. americanus, and Setaria viridis showed that the 14 chromosomes of P. giganteum again divided into two clear groups; the seven chromosomes in one of the groups were closer to C. americanus in phylogenetic distance than the seven chromosomes in the other group (Additional file 2: Figures S5–S7). Together these results indicate that the P. giganteum genome was successfully sorted into A and B subgenomes that contained seven chromosomes each (Fig. 1c).

Comparative genomics and gene family evolution analysis

To investigate the evolution of P. giganteum, we constructed a phylogenetic tree of 12 plant species (P. giganteum and P. purpureum were divided into two subgenomes) using 307 single-copy genes (Additional file 1: Table S17). Phylogenomic analysis showed that the A subgenome of P. giganteum and C. americanus originated from a common ancestor 5.9 million years ago (Mya), followed by the divergence of the A subgenome of P. purpureum and P. giganteum at 4.0 Mya (Fig. 2a). The B subgenome of P. purpureum and P. giganteum diverged more recently (approximately 2.6 Mya). Formation of the allotetraploid P. giganteum genome (AABB) occurred at approximately 8.3 Mya, at which time the two subgenomes of P. giganteum diverged from each other. Gene family expansion and contraction analysis showed that 2781 and 2848 gene families were expanded in the A and B subgenomes of P. giganteum after divergence from the common ancestor, respectively, and 901 of these gene families were unique to the P. giganteum genome (Fig. 2a; Additional file 1: Tables S16–S17). Most of the expanded gene families originated from dispersed duplication in both subgenomes, followed by proximal and tandem duplications (Fig. 2d). Notably, the percentage of dispersed duplication in the B subgenome (3310, 7.78%) was significantly higher than that in the A subgenome (2327, 5.44%) (P < 2.2 × 10⁻¹⁶, one-tailed Fisher’s exact test), while the percentages of tandem (P = 4.8 × 10⁻¹⁵), proximal (P = 5.73 × 10⁻¹²), WGD/segmental duplication (P = 3.03 × 10⁻⁹) in the B subgenome were significantly lower than those in the A subgenome.

Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment analysis showed that the rapidly expanded gene families in the A subgenome (539 gene families) and B (647 gene families) subgenomes were commonly enriched in nucleosome (GO: 0005509; A: 15/1710, Benjamini-Hochberg (BH)-adjusted P = 9.70 × 10⁻³; B: 13/1608, BH-adjusted P = 4.25 × 10⁻²), polysaccharide binding (GO:0030247; A:54/1710, BH-adjusted P = 4.23 × 10⁻¹⁹; B: 47/1608, BH-adjusted P = 5.00 × 10⁻¹⁸), and translation factors (A: K15032, 29/399, BH-adjusted P = 2.36 × 10⁻²⁸; B: K09419, 5/271, BH-adjusted P = 1.84 × 10⁻⁵), implying these gene families were involved in maintaining basic life activities (Fig. 2c; Additional file 1: Tables S18–S19). The expanded gene families in the A subgenome were uniquely enriched in recognition of pollen (GO:0048544, 23/1710, BH-adjusted P = 1.52 × 10⁻⁵), chitin binding (GO:0008061, 11/1710, BH-adjusted P = 2.11 × 10⁻⁶), tryptophan synthase activity (GO:0004834, 16/1710, BH-adjusted P = 8.79 × 10⁻¹³), nutrient reservoir activity (GO:0045735, 18/1710, BH-adjusted P = 1.95 × 10⁻⁵), and protein dimerization activity (GO:0046983, 61/1710, BH-adjusted P = 5.10 × 10⁻⁷) (Fig. 2c; Additional file 1: Table S18). The KEGG pathway analysis showed that the expanded gene families in the A subgenome were uniquely involved in phenylpropanoid biosynthesis (K00083, 6/399, BH-adjusted P = 7.89 × 10⁻⁶), calcium signaling pathway (K04150, 27/399, BH-adjusted P = 1.34 × 10⁻⁸), and glycosyltransferases (K08237, 8/399, BH-adjusted P = 1.59 × 10⁻⁹) (Fig. 2c; Additional file 1: Table S19). The expanded gene families in the B subgenome were uniquely enriched in nucleic (GO:0000786, 13/1608, BH-adjusted P = 1.33 × 10⁻²), ADP binding (GO:0043531, 70/1608, BH-adjusted P = 4.25 × 10⁻²), and response to wounding (GO:0009611, 15/1608, BH-adjusted P = 1.68 × 10⁻⁹) (Fig. 2c; Additional file 1: Table S18). The KEGG pathway analysis showed that the expanded gene families in the B subgenome were uniquely involved in plant hormone signal transduction (K13464, 6/271, BH-adjusted P = 2.99 × 10⁻⁵), replication and repair (K07497, 4/271, BH-adjusted P = 1.07 × 10⁻³), and transporters (K24193, 3/271, BH-adjusted P = 4.63 × 10⁻²) (Fig. 2c; Additional file 1: Table S19). Together, the functional enrichment results suggested that the expanded gene families may improve the material synthesis capacity and environmental adaptability of P. giganteum, which is consistent with its high tolerance of abiotic stress and significantly high biomass. The functional differentiation of the expanded gene families between the A and B subgenomes of P. giganteum indicated that the A subgenome was related mainly to nutrient synthesis whereas the B subgenome was related mainly to abiotic stress tolerance.

Interestingly, gene families related to flavonoid biosynthesis were significantly expanded in the P. giganteum genome (Fig. 2c). Previous studies have demonstrated that flavonoids played important roles in the growth, development, and disease resistance of plants [32]. Specifically, we found that the chalcone and stilbene synthase (CHS/STS) gene family was significantly expanded in P. giganteum (A: 48, B: 33) compared with its presence in closely related species (Sorghum bicolor: 30, C. americanus: 29, Setaria viridis: 23, Setaria italica: 20) (Fig. 2e). Additionally, the copy number and expressed gene number in the CHS/STS family were much higher in the A subgenome (48) than they were in the B subgenome (33) (Fig. 2e, f; Additional file 2: Figure S8), implying that the A subgenome contributed more to flavonoid biosynthesis than the B subgenome, further confirming functional differentiation between the subgenomes in P. giganteum.

We compared the gene families in four typical C4 species (P. giganteum, C. americanus, S. viridis, and S. bicolor). Functional enrichment analysis found that the highly conserved genes among these C4 plants were significantly enriched in substance synthesis and stress resistance in the GO biological process category (Additional file 2: Figure S9; Additional file 1: Tables S20–S21), which may be related to features of these C4 plants. The unique gene families detected in the A and B subgenomes of P. giganteum by comparison were mainly enriched in DNA integration, photosynthesis, gene silencing by RNA, negative regulation of translation, defense response, and response to wounding (Additional file 2: Figure S10; Additional file 1: Tables S22–S24), which implied that P. giganteum evolved to better cope with biotic and abiotic stress environment compared with the other C4 plants. Furthermore, 2292 and 2608 unique genes in the A and B subgenomes were identified by comparing the two subgenomes. The unique genes in the A subgenome were involved mainly in substance metabolism and stress response, whereas the unique genes in the B subgenome were associated mainly with material biosynthesis (Additional file 2: Figure S11; Additional file 1: Table S25). We also found that there were significant differences in the expression levels of conserved and unique genes in P. giganteum (Additional file 2: Figure S12).

Comparative genomic analysis between P. giganteum and its closely related species P. purpureum

Considering the high similarity in morphological characteristics and biological traits between P. giganteum and P. purpureum, we undertook a detailed comparison analysis between two species to explore their genomic difference. Firstly, sequence similarity analysis showed that 20.22% (17,251/85,313) of P. giganteum proteins had no BLAST hit against P. purpureum proteins (Fig. 3a), and 23.50, 49.50, and 27.00% of homologous gene pairs showed low (identity < 30%), moderate (30% ≤ identity < 90%), and high (identity ≥ 90%) sequence similarity, respectively (Fig. 3a), suggesting that there exists substantial difference in sequence between two species. Secondly, the proteins in two subgenomes of P. giganteum and P. purpureum were clustered, and a total of 31,314 orthogroups (OGs) were assigned. Among them, 11,888 (37.96%) OGs were shared by four subgenomes, while a substantial proportion of OGs were identified only in either species (Fig. 3b), suggesting the large differences in gene family between two species. For example, there were more copies of cellulose synthase-related genes (P. giganteum: 73; P. purpureum: 63) and C4 photosynthesis-related genes (P. giganteum: 78; P. purpureum: 65) in P. giganteum compared with their presence in P. purpureum. Further analysis of different types of duplicated genes showed that there were great differences in percentage between P. giganteum and P. purpureum (Additional file 2: Figure S13). Thirdly, genome-wide synteny analysis showed that almost all the chromosomes displayed 1:1 correspondence between two species (Fig. 3c). However, extensive genome rearrangements were detected, especially for chromosomes PgiA1, PgiA2, PgiA5, PgiA6, and PgiB12, suggesting the obvious changes in chromosome structure between two species during evolution. Fourth, the structural variations (SVs) between two genome assemblies were identified, yielding a total of 18,062 large SVs (> 50 bp) (Fig. 3e). The results showed that deletion (7108) and insertion (10,350) are top two most abundant types of SVs, followed by translocation (476) and inversion (128) (Fig. 3d, upper panel). As with the SV length, deletion, insertion, inversion, and translocation events spanned 248.28, 233.38, 226.09, and 135.09 Mb, respectively, suggesting that SVs involved a large fraction of genomic region (40.68%) (Fig. 3d, lower panel). Notably, several large inversions were observed in PgiA1 (0.28–45.26 Mb; 45.45–54.12 Mb; 54.14–61.34 Mb; 82.02–92.76 Mb), PgiA4 (112.85–127.17 Mb), PgiA6 (3.50–32.64 Mb), and PgiB12 (126.45–147.19 Mb) (Fig. 3e). Additionally, we made a comparison of SV count and length with four other groups of plants and found that the count and length of SVs varied greatly among these groups (Additional file 1: Table S26). In terms of insertion, we observed the longest average insertion length between P. purpureum and C. americanus, followed by P. purpureum and P. giganteum (Additional file 1: Table S26). These results indicated the occurrence of extensive genome rearrangements and great divergences between P. giganteum and P. purpureum. Fifth, we calculated the ratios of nonsynonymous substitution rate to synonymous substitution rate (Ka/Ks) for each subgenome of two species. The results showed that both sets of subgenomes exhibited significant differences in evolutionary rate between two species (A: P < 2.22 × 10⁻¹⁶; B: P < 2.22 × 10⁻¹⁶, Wilcoxon rank-sum test) (Fig. 3f). Interestingly, Ka/Ks ratios were significantly different between the A and B subgenomes of P. purpureum (P < 2.9 × 10⁻¹¹), while no significant difference was observed between two subgenomes of P. giganteum (P = 0.4) (Fig. 3f). Altogether, the above evidences confirmed the substantial differences between P. giganteum and P. purpureum, suggesting that P. giganteum is a different species from P. purpureum despite their high similarity in phenotype.

Genome evolution analysis of the A and B subgenomes

A total of 22,295 pairs of syntenic genes were identified between C. americanus and P. giganteum (A: 21,336; B: 19,490) (Fig. 4b; Additional file 2: Figure S14a, b). The A subgenome of P. giganteum was strongly collinear with C. americanus, and their chromosomes showed an almost 1:1 syntenic relationship (Fig. 4b; Additional file 2: Figure S14c). Conversely, many large-scale chromosomal rearrangements were identified between the B subgenome and C. americanus (Fig. 4b). Interestingly, we found that the B subgenome was highly syntenic with S. viridis, except for several rearrangements (Fig. 4c), and more highly conserved genes were found between the B subgenome and S. viridis than were found between the B subgenome and C. americanus (Additional file 2: Figure S15; Additional file 1: Table S27). Together, our results show that the A subgenome was evolutionarily closer to C. americanus, and the B subgenome was evolutionarily closer to S. viridis indicating it had a closer relationship to the ancestral status than the A subgenome had.

The synteny analysis between the A and B subgenomes of P. giganteum identified numerous genome rearrangements between them (Fig. 4d). Chromosome B8 contained the major parts of chromosomes A1 and A3, and the remaining parts of the A1 and A3 chromosomes were inserted into the middle part of chromosome A7 to create chromosome B9. Chromosome A4 contributed to the major part of chromosome B11 and the left part of chromosome B13, and the remaining part of chromosome B11 showed synteny with chromosome A6. Chromosome A5 was split into the major part of chromosome B14 and the right part of chromosome B13 (Fig. 4d; Additional file 2: Figure S16). Interestingly, these chromosomal rearrangement events tended to occur frequently in candidate pericentromeric regions of the P. giganteum chromosomes (Additional file 2: Figures S17-S18). Previous studies found that the A and B compartments are active and inactive chromatin regions, respectively [33, 34]. Thus, the switching of A/B compartment status might influence gene expression in corresponding genomic regions. To examine the effect of genomic rearrangement on the switching of A/B compartment status, we investigated the switching patterns of chromatin A/B compartment for syntenic blocks between the two subgenomes of P. giganteum. We found that most 3D chromatin A/B compartments were stable between the A and B subgenomes, with stable A/B compartment ratios of 57.24–83.42% (Additional file 1: Table S28), indicating a highly conserved chromatin spatial structure. However, we also found extensive compartment status switching from A to B or B to A between the two subgenomes (Additional file 2: Figure S18a). Further, we found that the rearranged genomic regions (41.84%) were more likely to undergo A/B compartment switching compared with the non-rearranged genomic regions (27.04%) (P < 2.2 × 10⁻¹⁶, one-tailed Fisher’s exact test) (Additional file 2: Figure S18b), suggesting that genome rearrangement may be an important factor that contributes to changes in gene regulation and expression. Additionally, an inter-genomic comparison identified eight inversions of large fragments (> 1 Mb) between the A and B subgenomes that were validated by the Hi-C contact map (Additional file 2: Figure S16). Together, these findings show that many genome rearrangements occurred between the A and B subgenomes of P. giganteum. These rearrangements may be an important driving force of species evolution.

We reconstructed the ancestral karyotypes of P. giganteum, S. viridis, S. bicolor, C. americanus, and Zea mays (Fig. 4d, e). The ancestral karyotypes showed that the common ancestor of P. giganteum and C. americanus had seven chromosomes. Prior to polyploidization of P. giganteum, ancestor 5 split from ancestor 3 through one fission and one fusion. The allotetraploid P. giganteum emerged after ancestor 5 underwent hybridization and 19 fission and 19 fusion events. The fewest rearrangements were recorded in S. bicolor relative to ancestor 2 (three fissions and three fusions), and the most rearrangements were detected in Z. mays relative to ancestor 4 (117 fissions and 119 fusions), partly because of the lineage-specific whole-genome duplication (WGD) event that occurred in Z. mays. Specifically, a large segmental inversion was identified between the front of chromosome 3 of S. bicolor and chromosome 5 of S. viridis, and this genomic region of S. viridis was completely collinear with chromosomes B9 and A7 of P. giganteum. Thus, we speculated that this inversion occurred only in the S. bicolor genome and that the ancestral state was likely maintained in P. giganteum and S. viridis. The B8 chromosome of P. giganteum was formed by the insertion of chromosome 6 of S. viridis into the near-middle part of chromosome 3 of S. viridis, and chromosome 7 of S. bicolor was completely collinear with chromosome 6 of S. viridis, suggesting that S. bicolor and S. viridis harbored the ancestral state of these regions. The reconstructed ancestral karyotypes provide a global view of genome rearrangements across these species, further supporting chromosomal rearrangement as an important driving force of genome evolution and speciation.

Analysis of differentially expressed homoeologous genes between the A and B subgenomes

To investigate the gene expression patterns between the subgenomes in P. giganteum, we performed a comprehensive transcriptome data analysis of six P. giganteum tissues, apical tip, inverted third tip, root, stem, and aboveground and underground seedling parts. The mapping ratios of the two subgenomes and gene regions were similar for each tissue (A: 49.09–50.63%, B: 49.37–50.91%) with difference of only 0.55–1.82% (Additional file 1: Table S29). The number of genes expressed in the B subgenome (22,245–25,643) was similar to the number in the A subgenome (22,419–25,742) for each tissue (Fig. 5a). Gene expression analysis showed that most of the expressed genes (87–89%) in the different tissues were homoeologous genes between the subgenomes, and the remaining 11–13% were subgenome-specific genes (Fig. 5b). These findings implied that there was no significant subgenome dominance in gene expression in P. giganteum. However, the differential expression analysis between paired tissues revealed a clear tissue-specific pattern of gene expression in P. giganteum (Additional file 1: Table S31; Additional file 2: Figures S19–S21). Functional enrichment analysis showed that the highly expressed genes (top 30% based on the FPKM) in the A subgenome were related to protein synthesis (GO:0005840, BH-adjusted P = 7.17 × 10⁻³–2.06 × 10⁻²), whereas the highly expressed genes in the B subgenome were enriched in protein metabolism (GO:0019752, BH-adjusted P = 2.34 × 10⁻²–6.70 × 10⁻²), substance transportation (GO:0030127, BH-adjusted P = 1.06 × 10⁻²–1.69 × 10⁻²), and gene expression regulation (GO:0003723, BH-adjusted P = 1.08 × 10⁻²–4.26 × 10⁻²) (Fig. 5c; Additional file 1: Table S30).

The differential expression of homoeologous genes is considered to be an important mechanism of heterosis in polyploid species [35]. A total of 21,248 pairs of homologous genes were identified between the A and B subgenomes. The homoeologous genes that showed significantly biased expression between the two subgenomes were defined as homoeologous differentially expressed genes (HDEGs). The homoeologous genes that were expressed only in one subgenome were defined as homoeolog-specific expressed genes (HSEGs). Notably, most of the expressed homoeologous gene pairs exhibited balanced expression (57.86–72.11%) between the two subgenomes for each tissue (Fig. 5d; Additional file 1: Table S32), further supporting the idea that there was no significant subgenome dominance in gene expression in P. giganteum. However, 22.42–32.62% of homoeologous genes displayed expression bias between the two subgenomes, and the genes with higher expression levels were defined as A or B dominant genes. We found that there were slightly more B dominant homoeologous genes (2451–3503) than A dominant homoeologous genes (2313–3429) for each tissue (Fig. 5d; Additional file 1: Table S32). Interestingly, 637 HDEGs were consistently dominant in the A subgenome (Additional file 2: Figure S22a), and 831 HDEGs were consistently dominant in the B subgenome across all tissues (Additional file 2: Figure S22b). The enriched functions of the HDEGs in different tissues varied remarkably between the two subgenomes (Additional file 2: Figure S22c). For example, the HDEGs in apical leaf tip and inverted third tip were involved mainly in material hydrolysis (GO:0015914, BH-adjusted P = 3.33 × 10⁻²–4.28 × 10⁻²) in the A subgenome, while the B subgenome contributed to lipid metabolic process (GO:0006629, BH-adjusted P = 1.07 × 10⁻²–1.67 × 10⁻²), providing sufficient energy for plant growth (Additional file 2: Figure S22c). Intriguingly, more HSEGs were identified in the A subgenome (445–548) than in the B subgenome (286–455) for most tissues (Additional file 1: Table S3). These HSEGs were divided into two groups according to their expression patterns among different tissues: (1) consistent group in which the HSEGs were uniquely expressed in the same subgenome in all tissues (A: 64, B: 102) (Additional file 2: Figure S23a, b) and (2) discordant group in which the HSEGs were uniquely expressed in only one subgenome in some tissues (A: 1376, B: 1228) (Additional file 2: Figure S23c). This distinction may have important implications for the formation of allotetraploid advantage. Overall, the genome-wide expression analysis results indicated that gene expression was balanced between the two subgenomes. Nonetheless, many homoeologous genes showed expression bias between the two subgenomes among different tissues, implying that polyploidization can enable the use of favorable copies of genes, which are then expressed at high levels, strongly suggesting functional differentiation between the two subgenomes of P. giganteum.

P. giganteum is a cellulose-rich tetraploid plant, and unsurprisingly, we found that there were many more members of Ces and Csl gene families in P. giganteum (73, A: 39, B: 34) than there were in Arabidopsis thaliana (40), Oryza sativa (43), and S. viridis (32) (Fig. 5e; Additional file 1: Table S34), implying that tetraploidization may have contributed to the high content of cellulose in P. giganteum. Transcriptome analysis showed that the cellulose synthesis genes were highly expressed in the stem and leaf of P. giganteum (Fig. 5e; Additional file 1: Table S35), indicating their important contribution to cellulose synthesis in these tissues. The phylogenetic analysis identified 30 homologous pairs of cellulose synthase genes between the two subgenomes; however, their expression patterns in the A and B subgenomes were significantly different among tissues (Fig. 5e; Additional file 1: Table S36). There were 28 and 23 genes that were expressed in all tissues of the A and B subgenomes, respectively. Additionally, 8 ~ 18 pairs of genes related to cellulose synthase were differentially expressed between two subgenomes in different tissues. For example, CesA1, CesA6, CesA8, and CslF showed sustained higher expression levels in the A subgenome than they did in the B subgenome across different tissues, and many cellulose synthase-like genes showed higher expression in the B subgenome than their homologs did in the A subgenome (Fig. 5e; Additional file 1: Table S37). Gene structure comparison of the homoeologous genes showed that the LTR insertions may be the main force for the significant differences in expression of the cellulose synthase-related genes and their subsequent functional differentiation (Additional file 2: Figure S24). A pair of CslF homologs (Pgi0G0010261 and Pgi0G0073710) in the two subgenomes showed significant differential expression among different tissues, and further analysis showed that two LTRs were inserted into the gene region of Pgi0G0073710 (Fig. 5f), which might be related to the differential expression of CslF between the subgenomes. Together, these results indicate that the A subgenome, which has more copy numbers and higher expression levels of cellulose synthase-related genes, may contribute more to the highly efficient cellulose biosynthesis for biomass accumulation than the B subgenome does.

Strengthened cellulose synthesis and C4 photosynthesis in P. giganteum

P. giganteum is a typical polyploid and C4 plant that is well known for its rapid growth and huge biomass production. To discover the genetic basis of biomass accumulation of P. giganteum, we collected 42 transcriptome samples of leaf and stem at seven different developmental periods, including three developing stages, tillering, jointing, and maturing (Fig. 6a). The transcriptome analysis identified that a 20,517 and 26,596 differentially expressed genes (DEGs) among the different developmental periods of leaves and stems, respectively (Additional file 2: Figures S25–S28). Interestingly, the GO enrichment of these DEGs showed that the genes were significantly correlated with the growth and development requirements of each developmental stage in leaves and stems (Additional file 1: Tables S38–S40). Especially in leaves, the GO enrichment of DEGs among the different developmental stages showed that the highly expressed genes at the tillering stage were enriched mainly in cell proliferation (GO:2,001,109, BH-adjusted P = 1.24 × 10⁻²–4.08 × 10⁻²) and growth factor activity (GO:0008083, BH-adjusted P = 1.24 × 10⁻²–3.23 × 10⁻²), which mainly promoted the tillering of plants. At the jointing stage, the highly expressed genes were enriched mainly in photosynthesis (GO:0019684, BH-adjusted P = 1.02 × 10⁻¹²–2.58 × 10⁻⁶), cellulose biosynthesis (GO:2,001,006, BH-adjusted P = 3.62 × 10⁻⁵–4.28 × 10⁻²), material transport (GO:2,000,880, BH-adjusted P = 2.42 × 10⁻⁷–7.10 × 10⁻³), cytokinin (GO:0080038, BH-adjusted P = 3.33 × 10⁻²–4.28 × 10⁻²) and xyloglucan metabolic process (GO:0010411, BH-adjusted P = 1.69 × 10⁻²–3.74 × 10⁻²), which contribute mainly to plant growth and metabolite biosynthesis to meet the need of rapid growth. At the maturing stage when plant development had slowed down, the highly expressed genes were enriched mainly in metabolism-phenylalanine catabolic process (GO:0006559, BH-adjusted P = 3.58 × 10⁻⁴) and cyclin-dependent protein serine/threonine kinase inhibitor activity (GO:1,904,993, BH-adjusted P = 3.98 × 10⁻²) (Additional file 1: Table S39).

The transcriptome analysis showed that the cellulose synthase-related genes were significantly differentially expressed among different developmental stages in leaves (Fig. 6b; Additional file 2: Figure S29) and stems (Additional file 2: Figure S30). In leaves, five CslE, five CslF, and three CslD genes were highly expressed at the tillering stage; two CslE and two CslH genes were highly expressed at the jointing stage; and 23 CesA and five CslF genes were highly expressed at the maturing stage. In stems, 15 CesA, five CslD and six CslF genes were highly expressed at the tillering stage to promote primary wall synthesis and seedling growth; and eight CslE, eight CslF, and 12 CesA genes were highly expressed at the jointing and maturing stages to promote plant morphogenesis and ensure biomass yield. Therefore, at different developmental stages of leaf and stem, cellulose biosynthesis involved multiple cellulose synthase genes with different expression patterns that cooperated with each other, which may explain, at least partially, why P. giganteum is rich in cellulose.

To discover the regulation and expression patterns of genes related to photosynthesis between subgenomes in C4 plants with polyploidization, C4 photosynthesis-related genes in nine gene families were identified in P. giganteum (78, A: 39, B:39), which is more than the number in P. purpureum (56), C. americanus (20), and S. italica (24) (Fig. 5c). Most of the C4 photosynthesis-related genes in P. giganteum (64/79, 81.01%) were expressed at all seven leaf developmental periods, and 31 homologous 1:1 gene pairs were identified between the A and B subgenomes. The transcriptome analysis showed that there was no significant difference in expression between most of these homologous gene pairs (30/31) across all developmental periods. The C4 photosynthesis-related homologous genes were more highly conserved than other gene families such as protein phosphatases and ubiquitin-mediated proteolysis genes (Fig. 6d). We inferred that these highly conserved C4 photosynthesis-related genes between the A and B subgenomes may have similar expression patterns, which implied that the A and B subgenomes conservatively retained and even expanded photosynthesis-related gene families after the formation of allotetraploid P. giganteum. Interestingly, we found that there was at least one A/B homologous gene pair at each step of the C4 photosynthesis pathway. These photosynthesis-related gene pairs were consistently and highly expressed at all developmental stages and may be key candidate genes for genetically improving plant growth and biomass production through the C4 photosynthesis pathway. We analyzed the gene structures of the photosynthesis-related homologous gene pairs, Pgi0G0022029 and Pgi0G0048817, that was significantly differentially expressed, and found that Pgi0G0048817 had a large deletion that led to the loss of one exon, which may have resulted in its lower expression levels in different developmental periods (Fig. 6e). In summary, the complementarity and additivity of genetic effects of C4 photosynthesis-related genes in the A and B subgenomes may drive massive biomass accumulation and contribute to the rapid growth of P. giganteum, which also indicated that both of the A and B subgenomes contributed greatly and almost equally to the highly efficient C4 photosynthesis.

Discussion

Due to its huge biomass and high application potential as bioenergy and forage grass, P. giganteum has been cultivated worldwide. Decoding the genetic base underlying the excellent traits of P. giganteum for molecular breeding and genetic improvement is urgently needed to meet the dramatically increased demand for sustainable forage production. Here, we generated a high-quality genome assembly of P. giganteum, which represents a major advancement in grass genomics. The quality of the P. giganteum assembly (contig N50 88.47 Mb) was significantly higher than other recently published forage genomes including those of P. purpureum cv. Purple (contig N50 1.83 Mb) [24], orchard grass (Dactylis glomerata L.; contig N50 0.93 Mb) [36], and Miscanthus lutarioriparius (contig N50 1.71 Mb) [37], broomcorn millet (contig N50 2.55 Mb) [38], S. spontaneum Np-X (contig N50 0.38 Mb) [39], A. longiglumis (contig N50 7.30 Mb), A. insularis (contig N50 5.64 Mb), A. sativa ssp. nuda cv. Sanfensan (contig N50 75.27 Mb) [40].

Gene family expansion and contraction analysis showed many gene families were significantly expanded in the A and B subgenomes of P. giganteum (Fig. 2a), and there was clear functional differentiation of the expanded gene families between the two subgenomes. WGD analysis showed that both subgenomes of P. giganteum underwent only the ancient ρ-WGD event shared by grass species, and did not experience any independent WGD event, which is supported by the different types of duplicated genes (Fig. 2d). On the basis of the insertion time of intact LTR-RTs, we detected two LTR peaks at 0.24 Mya and 1.19 Mya in P. giganteum (Additional file 2: Figure S31b) and one peak at 1.07 Mya in C. americanus. No apparent peak for LTR insertion was detected in S. italica or S. viridis. These findings indicated that the recent LTR burst in genus Pennisetum possibly contributed to their significantly larger genome sizes compared with those in genus Setaria (Fig. 3a).

Comparative genomic analysis strongly supported the idea that P. giganteum and P. purpureum are two different species, despite their high similarity in many morphological features and agronomic traits (Additional file 2: Figure S31; Additional file 1: Table S26). The reconstructed ancestral karyotype of grass species identified massive chromosomal rearrangements among them and a clear change trajectory in chromosome numbers, from ten in S. bicolor, nine in S. viridis, to seven in C. americanus and P. giganteum. The synteny analysis showed the A5 segment (50 Mb) in the A5–B14 collinear region was remarkably shorter than the B14 counterpart segment (80 Mb) (Additional file 1: Table S41; Additional file 2: Figure S31c). Interestingly, we found that the length and number of transposable elements in B14 (57.31 Mb) were much more than those in A5 (22.50 Mb). These results indicated that abundant transposable elements were non-uniformly and specifically inserted into chromosome B14, which may provide a potentially excellent resource for studying the mechanism of transposable element amplification or genome evolution.

It has been widely accepted that C4 plants show a distinguishing ability to concentrate CO₂. Compared with three other closely related C4 plants, P. giganteum harbored more copies of C4 photosynthesis-related genes, probably suggesting that P. giganteum possesses stronger photosynthesis ability for rapid biomass accumulation. More than 80% of C4 photosynthesis-related genes showed sustained expression across the whole developmental stages of P. giganteum, implying that the majority of genes in the C4 photosynthesis pathway were transcriptionally active. Additionally, almost all the homologous gene pairs showed no significant expression bias between two subgenomes, suggestive of the equal contribution of two subgenomes to the photosynthesis. Cellulose is one of the most important components of cell wall in higher plants, which has great impact on the growth and development of plants. On the one hand, more copies of Ces superfamily members were identified in P. giganteum compared with three other diploid species, implying the contribution of polyploidization to this gene family size. On the other hand, we also observed difference in expression between homologous Ces genes in two subgenomes. For example, a pair of CslF homologous genes (Pgi0G0010261 and Pgi0G0073710) showed significant difference in expression, and we hypothesized that the difference was possibly caused by the insertion of two LTRs in the genic region of Pgi0G0073710. Transcriptome data analysis showed difference in expression between leaf and stem tissues across three developmental stages. In leaf, most of the CesA genes (24) showed the highest expression in the maturing stage. By contrast, the numbers of CesA genes showing higher transcript abundance were stable across three developmental stages. These results might suggest that the synthesis of cellulose in stem started earlier than in leaf, indicating the distinct roles of different tissues in biomass accumulation.

The existence of subgenome dominance in polyploids has been debated by researchers for a long time and controversies still remain [21, 25, 41,42,43,44,45,46]. In this study, we compared our results with those in previous studies on allopolyploid species, including Gossypium hirsutum [47], C. purpureus [24], and Brassica juncea [21]. The results showed that the average percentage of homoeologous genes showing differential expression between subgenomes was the lowest in P. giganteum (Additional file 2: Figure S32a). To exclude the bias that might be caused by different approaches, we re-analyzed the RNA-seq data from different studies using the same method as used by us. The results indicated that the percentage of homoeologous genes showing biased expression between subgenomes is highest in P. giganteum compared with three other polyploid plants (Additional file 2: Figure S32b). However, the ratios of homoeologs that were dominant in two subgenomes were almost equal in P. giganteum (Additional file 1: Table S42; Additional file 2: Figure S32b). Therefore, we made a conclusion that no significant subgenome dominance was observed in P. giganteum at the overall expression level, which is consistent with several allopolyploid species [21, 48, 49], but a large number of differentially expressed homoeologous genes and homoeolog-specific expressed genes were identified between the two subgenomes in different tissues and at different developmental periods and different developing stages (Fig. 5a, b). The functional enrichment analysis showed that there were significant differences in functional categories of the DEGs between the two subgenomes. The C4 photosynthesis-related genes were expanded in P. giganteum and highly conserved between the two subgenomes, which resulted in the balanced and high expression of the homoeologous gene pairs (30/31) between the A and B subgenomes at different developmental periods. This finding indicated that both the A and B subgenomes contributed greatly and almost equally to the highly efficient C4 photosynthesis of P. giganteum, which drives massive biomass accumulation and contributes to its rapid growth. Key candidate genes in each gene family in the C4 photosynthesis pathway were identified and were found to be consistently and highly expressed at all developmental stages.

Conclusions

In summary, this study reported a high-quality genome assembly for the allotetraploid forage grass P. giganteum. Our results showed clearly about the functional differentiation between the two subgenomes in P. giganteum. The combined comparative genomics analysis and transcriptome meta-analysis showed that complementarity and additivity of genetic effects between the A and B subgenomes may have important impacts on biomass accumulation and rapid growth of P. giganteum. The fundamental resources described here will become important references for further studies of the genetic basis of advantageous traits in polyploid species and will facilitate further functional genomics-related research and genetic improvements of P. giganteum.

Methods

Plant materials, DNA and RNA sample collection and sequencing

P. giganteum plants were obtained from National Engineering Research Center of Juncao Technology (Fuzhou City, Fujian Province, China) and planted in Hebei University (Baoding, Hebei Province, China). For genome sequencing, high molecular weight DNA was extracted from the leaves of fresh plants using a DNA extraction kit. Two PCR-free single-molecule real-time (SMRT) bell libraries were constructed from the extracted genomic DNA and sequenced on a third-generation PacBio Sequel II platform (Annoroad Gene Technology, Beijing, China) to obtain high-quality sequencing data. For Hi-C sequencing, the leaves of fresh plants were cross-linked with formaldehyde and the reaction was terminated using glycine solution. The Hi-C library was constructed following the standard protocol and paired-end sequenced on an Illumina HiSeq 2000 platform (Annoroad Gene Technology). For transcriptome sequencing (RNA-seq), P. giganteum plants were regenerated by stem propagation and six tissues were harvested from fresh plants, namely parietal tip, inverted third leaf tip, root, stem, leaf, and aboveground and underground parts. The tissue samples were rinsed using ddH₂O, snap frozen in liquid nitrogen, and stored in − 80 ℃ until used. Three biological replicates were collected for each tissue sample. Total RNA was extracted from the tissue samples using a RNeasy Plant Mini Kit (Qiagen), cDNA libraries were constructed following the manufacturer’s instructions, and paired-end sequencing was performed on a NovaSeq sequencing platform (Illumina) (Berry Genomics, Beijing, China).

Genome assembly and quality assessment

To estimate the genome size and heterozygosity ratio of P. giganteum, k-mer frequency analysis was conducted using the genomic character estimate (GCE v1.0.2) [50] with default parameters. To achieve high-quality genome assembly, high-fidelity (HiFi) reads were generated on a PacBio Sequel II platform using the circular consensus sequencing (CCS) mode. First, the P. giganteum genome was de novo assembled using Hifiasm (v0.16.0-r369) [51] with default parameters. Then, the Hi-C reads were used to anchor the assembled contigs onto chromosome pseudomolecules through sorting, orientation, and ordering using 3D-DNA (v170123) [52] and Juicer (v1.6) [53] to yield a final version of the P. giganteum genome assembly. The quality of the genome assembly was evaluated using the Benchmarking Universal Single-Copy Orthologs (BUSCO v4.1.4) [54] based on the embryophyta_odb10 dataset, which contains 1614 core genes across land plants and the LTR assembly index [31].

Genome annotation

Repetitive sequences in the P. giganteum genome were annotated using ab initio and homology-based methods. The Tandem Repeat Finder (TRF v4.07b) [55] was used to identify tandem repeat sequences. For the de novo searches, RepeatModeler (v1.0.11), LTR_FINDER (v1.05) [56], LTRharvest (v1.5.11) [57], and LTR_retriever (v1.9) [58] were used to construct de novo repeat libraries, and RepeatMasker (v4.1.1) was used to identify genome-wide repeat sequences based on the de novo library. For homology-based searches, repeat elements were identified by RepeatMasker using a known repeat library (Repbase 15.02) (https://www.girinst.org/server/RepBase/index.php). Additionally, tRNA genes were predicted using tRNAscan-SE [59]. Other non-coding RNAs such as rRNA, miRNA, and snRNA were predicted using INFERNAL (v1.1) [60] by searching against the Rfam database (v9.1) [61].

To predict protein-coding genes in the P. giganteum genome, gene models were obtained by combining three prediction approaches, namely ab initio, homology-based, and transcriptome-based methods. For ab initio prediction, PASA (v2.3) [62] was used to predict gene structure using transcripts assembled by Trinity (v2.12) [63]; these transcripts were also used in AUGUSTUS (v3.2.3) [64] to train the gene model. For homology-based prediction, GenomeThreader (v1.7.3) [65] was used to search against the protein sequences of seven plant species, namely A. thaliana, P. purpureum, O. sativa, S. viridis, S. bicolor, and C. americanus. For transcriptome-based prediction, Trinity was used to assemble the RNA-seq data into transcripts, and PASA software was used for gene structure prediction based on transcriptome assembly. In addition, the RNA-seq data were mapped to the P. giganteum genome using HISAT2 (v2.2.1) [66], and StringTie (v2.1.6) [67] was used for reference-guided transcript assembly. ORFs (open reading frames) in the assembled transcripts were predicted using TransDecoder (v5.1.0). The prediction results from the different sources were combined using EVidenceModeler (EVM) [68]. The final gene set was evaluated by BUSCO based on the embryophyta_odb10 dataset.

Subgenome division of the P. giganteum genome

We used three methods to divide the P. giganteum genome into subgenomes. First, the next-generation sequencing data of the potential A subgenome donor C. americanus was mapped onto the P. giganteum genome using BWA-MEM (v0.7.17) [69] and two subgenomes were identified based on mapping rates. Second, protein-coding genes from the two subgenomes of P. giganteum,S. viridis, and C.americanus were used to construct orthogroups using OrthoFinder [70]. The single-copy gene clusters shared by them were extracted and aligned for each group of homologous chromosomes using MUSCLE (v3.8.1551) [71]. Collinear blocks among the two subgenomes of P. giganteum, C.americanus, S. viridis, and S. bicolor were obtained by JCVI (v1.2.1). The 1:1:1:1:1 syntenic blocks (> 50 genes) were selected to construct phylogenetic trees. Third, subgenome-specific repetitive DNA sequences were identified using SubPhaser [72]. Jellyfish v2.2.10 [73] was used to scan for and count 13-bp sequences (13-mers) in the 14 P. giganteum chromosomes. After normalizing the differential k-mer matrices, the k-means algorithm was used to cluster chromosomes into two groups (subgenomes).

Functional annotation

To determine the functional relevance of the predicted gene models in the P. giganteum genome, protein-coding genes were retrieved and functionally annotated by BLAST searches against known databases, namely UniProtKB/SwissProt and the NCBI non-redundant (NR) protein database. Gene ontology (GO) functional and Pfam domain annotations were assigned using InterProScan (v5.32) [74, 75]. KEGG orthology (KO) terms for each gene were assigned by homology searches against the hidden Markov model (KOfam) database using KofamScan [76] with default parameters.

Hi-C read mapping and normalization

Clean Hi-C sequencing data were mapped to P. giganteum genome using Bowtie2 with default parameters. The HiC-Pro pipeline (v3.0.0) [77] was used for removing singleton reads, multiple-mapping reads and duplicated read pairs, and only pairs for which both reads are uniquely aligned were kept to identify valid interactions. Raw contact matrices were built using bin sizes of 500 kb and normalized with the iterative correction and eigenvector decomposition (ICE) method using HiC-Pro.

Identification of A and B compartments

The eigenvector module implemented in the Juicer tools (v1.6) [53] was applied to identify A and B compartments on the 500-kb corrected matrix of each chromosome in P. giganteum. For each chromosome, genomic bins for which the first eigenvector (PC1) with a positive or negative value were assigned to compartment A or B, respectively. The genes located in each genomic bin were assigned with the corresponding compartment. To identify the switching of A/B compartment status between two subgenomes, we compared the compartment status change of syntenic gene pairs.

Variant calling and validation of accuracy rate of single base

The paired-end Illumina reads of P. giganteum were aligned to the assembled genome using the BWA-MEM [69] algorithm. Only reads with mapping quality > 30 were retained for downstream analysis. Variant calling was performed using the GATK pipeline (v4.1.6.0; https://github.com/broadinstitute/gatk) with the best practice protocol. After filtering out variants with low coverage or low quality, we selected only the biallelic single-nucleotide polymorphisms (SNPs) for subsequent analysis. The single-base error rate of the assembled genome was calculated as the percentage of homozygous SNP loci in the whole genome, thus the single-base accuracy rate was calculated as: accuracy rate = 1 − error rate.

Phylogenetic tree construction and evolutionary rate estimation

Orthologous gene clusters in the A and B subgenomes of P. giganteum and 11 other representative plants, namely Setaria viridis (GenBank: GCA_005286985.1), Pennisetum purpureum (https://ngdc.cncb.ac.cn/search/?dbId=gwh &q = GWHAORA00000000 +), Brachypodium distachyon (RefSeq: GCF_000005505.3), Dichanthelium oligosanthes (GenBank: GCA_001633215.2), Oryza sativa (Phytozome), Cenchrus americanus (http://cegsb.icrisat.org/ipmgsc/genome.html), Sorghum bicolor (RefSeq: GCF_000003195.3), Setaria italica (RefSeq: GCF_000263155.2), Zea mays (RefSeq: GCF_000005005.2) and two outgroup species, Arabidopsis thaliana (GenBank: GCA_000001735.2) and Glycine max (GenBank: GCA_000004515.5), were identified using the OrthoFinder (v2.3.14) pipeline [70]. A total of 2561 orthologous groups containing 2961 genes were identified among these species, including 307 single-copy orthologs. These single-copy orthologous genes were used to build a maximum likelihood tree with FastTree (v2.1.9) [78]. The species divergence time was calibrated using r8s [79] based on the TimeTree website [80]. For example, 42.0–52.0 Mya was used for B. distachyon and D. oligosanthes. FigTree (v1.4.3) was used fto visualize the species tree.

Gene family analysis

CAFE version 4.0.1 [81] with the default parameters was used to explore the change of gene families. Gene families that were identified in at least five of the species were selected for further analysis. The changes in gene families along each lineage of the phylogenetic tree were evaluated using the built-in model. The change of gene families of all the nodes was analyzed.

Whole-genome duplication (WGD) and synteny analysis

Synonymous substitution rate (Ks) estimation was used to detect WGD events in P. giganteum. First, the protein sequences of P. giganteum, C. americanus, S. bicolor, and S. viridis were aligned against self using BLASTP (E-value = 1 × 10⁻¹⁰). Then, the collinear blocks of these plants were identified using MCScanX [82, 83]. The WGD events in each plant species were evaluated based on the Ks distribution [84]. Ks values of the homologous gene pairs in the collinear blocks were calculated using the KaKs_calculator (v2.0) [85] with the YN model; the median of Ks values was considered to be representative of the collinear blocks. The Ks values of all gene pairs were plotted to identify putative WGD events in P. giganteum. The time of WGD and species divergence was estimated as t = Ks/2r, where r is the neutral substitution rate. In this study, we used a neutral substitution rate of 6.5 × 10⁻⁹ as reported previously [25].

Ancestor genome structure inference among P. giganteum and other related grass species

We applied OrthoFinder to find orthogroups among P. giganteum, C. americanus, S. viridis, Z. mays, and S. bicolor. The orthogroups were filtered to generate an input file for DRIMM-Synteny [86] following the instructions in the processDrimm pipeline (https://github.com/xjtu-omics/processDrimm). Then, DRIMM-Synteny was run to generate synteny and blocks information. The DRIMM-Synteny output files were processed to retain the synteny blocks that met the expected copy number as the input file for the IAGS pipeline [87] (https://github.com/xjtu-omics/IAGS) using the Python script processDrimm.py (https://github.com/xjtu-omics/processDrimm/processDrimm.py). Finally, the ancestor genome structure was inferred following the IAGS pipeline instructions.

Genome-wide differential expression analysis

Syntenic gene pairs between the A and B subgenomes of P. giganteum were obtained using MCScanX with the default parameters, and 21,258 reciprocal best-match gene pairs that met the 1:1 relationship between the two subgenomes were identified as allelic gene pairs. To investigate differences in gene expression between the two subgenomes of P. giganteum, we calculated the FPKM (fragments per kilobase of exon model per million mapped fragments) values of the allelic genes in six tissues, namely apical tip, inverted third tip, root, stem, and the aboveground and underground parts of seedling, and at seven developmental stages of leaf and stem. Differentially expressed genes were identified using the DESeq2 (v1.32.0) package [88] with a minimum of twofold differential expression (|log2Foldchange|> 1) and P_adj < 0.05. Sequence alignment and phylogenetic tree construction of cellulose synthase and C4 photosynthesis-related genes were performed using MUSCLE (v3.8.1551) [71] and IQ-TREE 2 [89]. The phylogenetic tree was visualized using iTOL v5 [90].

Comparative genomic analysis of P. giganteum and P. purpureum

To compare the similarity of amino acid sequences between P. giganteum and P. purpureum, BLASTP search was performed with the E-value cutoff = 1 × 10⁻³, and the best hit was used for sequence similarity analysis. Additionally, OrthoFinder [70] was used to assign orthogroups (OGs) to the subgenomes of P. giganteum and P. purpureum. Genome-wide synteny analysis between P. giganteum and P. purpureum was conducted using JCVI with the default parameters. MUMmer (v4.0.0beta2) [91] was employed for whole genome alignment between P. giganteum and P. purpureum, and SyRI [92] was used to identify genomic structural variation (SVs) between them based on whole genome alignment result. The visualization of genome syntenic and rearrangement regions was performed using plotsr [93]. The original SVs generated by SyRI were filtered to retain only large SVs (> 50 bp) with the scripts SyRI_Parse.pl and Covert_CPV.pl in the pipeline GenomeSV (https://github.com/wolongac/GenomeSV). To compare the difference in nucleotide substitution rate between two species, Ka/Ks ratios were calculated for each subgenome of two species using KaKs_Calculator 2.0 [85], respectively.

Availability of data and materials

All data generated or analyzed during this study are included in this published article, its supplementary information files and publicly available repositories. PacBio HiFi whole-genome sequencing reads, Illumina genomic (including Hi-C experiment) and transcriptomic reads generated in this study have been deposited in the Genome Sequence Archive [94] in National Genomics Data Center [95], China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences (GSA: PRJCA011649) and are publicly accessible at https://ngdc.cncb.ac.cn/gsa. Additionally, the genome assembly and annotation data have been deposited in the Figshare database (https://doi.org/10.6084/m9.figshare.23118794) [96].

Abbreviations

CCS:: Circular consensus sequencing
Hi-C:: High-throughput chromosome conformation capture
BUSCO:: Benchmarking Universal Single-Copy Orthologs
RNA-seq:: RNA sequencing
LTR-RTs:: Long terminal repeat retrotransposons
Mya:: Million years ago
WGD:: Whole-genome duplication
GO:: Gene ontology
KEGG:: Kyoto Encyclopedia of Genes and Genomes
KO:: KEGG orthology
BH:: Benjamini-Hochberg
CHS/STS:: Chalcone and stilbene synthase
OGs:: Orthogroups
SV:: Structural variation
HDEG:: Homoeologous differentially expressed gene
HSEG:: Homoeolog-specific expressed gene
CesA :: Cellulose synthase
Csl :: Cellulose synthase-like
DEG:: Differentially expressed gene
SMRT:: Single-molecule real-time
HiFi:: High-fidelity
EVM:: EVidenceModeler
SNP:: Single-nucleotide polymorphism
Ks:: Synonymous substitution rate
Ka:: Nonsynonymous substitution rate
FPKM:: Fragments per kilobase of exon model per million mapped fragments
NR:: Non-redundant

References

Lopez B, Hines PJ, Ash C. The unrecognized value of grass. Science. 2022;377:590–1.
Article PubMed Google Scholar
Strömberg CAE, Staver AC. The history and challenge of grassy biomes. Science. 2022;377:592–3.
Article PubMed Google Scholar
McSteen P, Kellogg EA. Molecular, cellular, and developmental foundations of grass diversity. Science. 2022;377:599–602.
Article CAS PubMed Google Scholar
Xu J, Song Y, Jing X, Li M. Characterization of the complete chloroplast genome sequence of Pennisetum glaucum and its phylogenetic implications. Mitochondrial DNA Part B. 2019;4(2):3764–5.
Article PubMed PubMed Central Google Scholar
Dos Reis GB, Mesquita AT, Torres GA, Andrade-Vieira LF, Pereira AV, Davide LC. Genomic homeology between Pennisetum purpureum and Pennisetum glaucum (Poaceae). Comp Cytogenet. 2014;8(3):199–209.
Article PubMed PubMed Central Google Scholar
Fulkerson WJ, Neal JS, Clark CF, Horadagoda A, Nandra KS, Barchia I. Nutritive value of forage species grown in the warm temperate climate of Australia for dairy cows: Grasses and legumes. Livest Sci. 2007;107:253–64.
Article Google Scholar
Samson R, Mani S, Boddey R, Sokhansanj S, Quesada D, Urquiaga S, et al. The potential of C4 perennial grasses for developing a global BIOHEAT industry. Crit Rev Plant Sci. 2005;24(5–6):461–95.
Article Google Scholar
Chen B, Lin J, Luo Z, Ye J, Jiao W. Determination of routine nutritional ingredients and amino acid contents of Jujuncao during different growth stages. Guizhou Agric Sci. 2016;44(1):101–3.
Google Scholar
Zheng H, Lin X, Lin H, Li J, Lu G, Lin Z. Research progress on JUJUNCAO (Pennisetum giganteum). J Fujian Agric For Univ (Natural Science Edition). 2019;48:681–7.
Google Scholar
Somerville C, Youngs H, Taylor C, Davis SC, Long SP. Feedstocks for Lignocellulosic biofuels. Science. 2010;329(5993):790–2.
Article CAS PubMed Google Scholar
McFarlane HE, Doring A, Persson S. The cell biology of cellulose synthesis. Annu Rev Plant Biol. 2014;65:69–94.
Article CAS PubMed Google Scholar
Polko JK, Kieber JJ. The regulation of cellulose biosynthesis in plants. Plant Cell. 2019;31(2):282–96.
Article CAS PubMed PubMed Central Google Scholar
Yin Y, Huang J, Xu Y. The cellulose synthase superfamily in fully sequenced plants and algae. BMC Plant Biol. 2009;9:99.
Article PubMed PubMed Central Google Scholar
Little A, Schwerdt JG, Shirley NJ, Khor SF, Neumann K, O’Donovan LA, et al. Revised phylogeny of the cellulose synthase gene superfamily: insights into cell wall evolution. Plant Physiol. 2018;177(3):1124–41.
Article CAS PubMed PubMed Central Google Scholar
Richmond TA, Somerville C. The cellulose synthase superfamily. Plant Physiol. 2000;124(2):495–8.
Article CAS PubMed PubMed Central Google Scholar
Carpita NC. Update on mechanisms of plant cell wall biosynthesis: how plants make cellulose and other (1->4)-beta-D-glycans. Plant Physiol. 2011;155(1):171–84.
Article CAS PubMed Google Scholar
Jobling SA. Membrane pore architecture of the CslF6 protein controls beta-glucan structure. Sci Adv. 2015;1: e1500069.
Article PubMed PubMed Central Google Scholar
Sarkar P, Bosneaga E, Auer M. Plant cell walls throughout evolution: towards a molecular understanding of their design principles. J Exp Bot. 2009;60(13):3615–35.
Article CAS PubMed Google Scholar
Corneillie S, De Storme N, Van Acker R, Fangel JU, De Bruyne M, De Rycke R, et al. Polyploidy affects plant growth and alters cell wall composition. Plant Physiol. 2019;179(1):74–87.
Article CAS PubMed Google Scholar
Van de Peer Y, Ashman TL, Soltis PS, Soltis DE. Polyploidy: an evolutionary and ecological force in stressful times. Plant Cell. 2021;33(1):11–26.
Article PubMed Google Scholar
Yang J, Liu D, Wang X, Ji C, Cheng F, Liu B, et al. The genome sequence of allopolyploid Brassica juncea and analysis of differential homoeolog gene expression influencing selection. Nat Genet. 2016;48(10):1225–32.
Article CAS PubMed Google Scholar
Li Q, Xiang C, Xu L, Cui J, Fu S, Chen B, et al. SMRT sequencing of a full-length transcriptome reveals transcript variants involved in C18 unsaturated fatty acid biosynthesis and metabolism pathways at chilling temperature in Pennisetum giganteum. BMC Genomics. 2020;21(1):52.
Article CAS PubMed PubMed Central Google Scholar
Varshney RK, Shi C, Thudi M, Mariac C, Wallace J, Qi P, et al. Pearl millet genome sequence provides a resource to improve agronomic traits in arid environments. Nat Biotechnol. 2017;35(10):969–76.
Article CAS PubMed PubMed Central Google Scholar
Yan Q, Wu F, Xu P, Sun Z, Li J, Gao L, et al. The elephant grass (Cenchrus purpureus) genome provides insights into anthocyanidin accumulation and fast growth. Mol Ecol Resour. 2021;21(2):526–42.
Article CAS PubMed Google Scholar
Zhang S, Xia Z, Li C, Wang X, Lu X, Zhang W, et al. Chromosome-scale genome assembly provides insights into speciation of allotetraploid and massive biomass accumulation of elephant grass (Pennisetum purpureum Schum.). Mol Ecol Resour. 2022;22(6):2363–78.
Article CAS PubMed Google Scholar
Evans JR. Improving photosynthesis. Plant Physiol. 2013;162(4):1780–93.
Article CAS PubMed PubMed Central Google Scholar
Li S, Moller CA, Mitchell NG, Martin DG, Sacks EJ, Saikia S, et al. The leaf economics spectrum of triploid and tetraploid C4 grass Miscanthus x giganteus. Plant Cell Environ. 2022;45(12):3462–75.
Article CAS PubMed PubMed Central Google Scholar
Langdale JA. C4 cycles: past, present, and future research on C4 photosynthesis. Plant Cell. 2011;23(11):3879–92.
Article CAS PubMed PubMed Central Google Scholar
Hibberd JM, Covshoff S. The regulation of gene expression required for C4 photosynthesis. Annu Rev Plant Biol. 2010;61:181–207.
Article CAS PubMed Google Scholar
Williams BP, Aubry S, Hibberd JM. Molecular evolution of genes recruited into C₄ photosynthesis. Trends Plant Sci. 2012;17(4):213–20.
Article CAS PubMed Google Scholar
Ou S, Chen J, Jiang N. Assessing genome assembly quality using the LTR Assembly Index (LAI). Nucleic Acids Res. 2018;46(21): e126.
PubMed PubMed Central Google Scholar
Nascimento LBDS, Tattini M. Beyond photoprotection: the multifarious roles of flavonoids in plant terrestrialization. Int J Mol Sci. 2022;23(9):5284.
Article CAS PubMed PubMed Central Google Scholar
Dixon JR, Jung I, Selvaraj S, Shen Y, Antosiewicz-Bourget JE, Lee AY, et al. Chromatin architecture reorganization during stem cell differentiation. Nature. 2015;518(7539):331–6.
Article CAS PubMed PubMed Central Google Scholar
Dong P, Tu X, Liang Z, Kang BH, Zhong S. Plant and animal chromatin three-dimensional organization: similar structures but different functions. J Exp Bot. 2020;71(17):5119–28.
Article CAS PubMed Google Scholar
Shao L, Xing F, Xu C, Zhang Q, Che J, Wang X, et al. Patterns of genome-wide allele-specific expression in hybrid rice and the implications on the genetic basis of heterosis. Proc Natl Acad Sci U S A. 2019;116(12):5653–8.
Article CAS PubMed PubMed Central Google Scholar
Huang L, Feng G, Yan H, Zhang Z, Bushman BS, Wang J, et al. Genome assembly provides insights into the genome evolution and flowering regulation of orchardgrass. Plant Biotechnol J. 2020;18(2):373–88.
Article CAS PubMed Google Scholar
Miao J, Feng Q, Li Y, Zhao Q, Zhou C, Lu H, et al. Chromosome-scale assembly and analysis of biomass crop Miscanthus lutarioriparius genome. Nat Commun. 2021;12(1):2458.
Article CAS PubMed PubMed Central Google Scholar
Shi J, Ma X, Zhang J, Zhou Y, Liu M, Huang L, et al. Chromosome conformation capture resolved near complete genome assembly of broomcorn millet. Nat Commun. 2019;10(1):464.
Article CAS PubMed PubMed Central Google Scholar
Zhang Q, Qi Y, Pan H, Tang H, Wang G, Hua X, et al. Genomic insights into the recent chromosome reduction of autopolyploid sugarcane Saccharum spontaneum. Nat Genet. 2022;54(6):885–96.
Article CAS PubMed Google Scholar
Peng Y, Yan H, Guo L, Deng C, Wang C, Wang Y, et al. Reference genome assemblies reveal the origin and evolution of allohexaploid oat. Nat Genet. 2022;54(8):1248–58.
Article CAS PubMed PubMed Central Google Scholar
Zhang G, Ge C, Xu P, Wang S, Cheng S, Han Y, et al. The reference genome of Miscanthus floridulus illuminates the evolution of Saccharinae. Nat Plants. 2021;7(5):608–18.
Article CAS PubMed PubMed Central Google Scholar
Bird KA, Niederhuth CE, Ou S, Gehan M, Pires JC, Xiong Z, et al. Replaying the evolutionary tape to investigate subgenome dominance in allopolyploid Brassica napus. New Phytol. 2021;230(1):354–71.
Article CAS PubMed PubMed Central Google Scholar
Shen Y, Li W, Zeng Y, Li Z, Chen Y, Zhang J, et al. Chromosome-level and haplotype-resolved genome provides insight into the tetraploid hybrid origin of patchouli. Nat Commun. 2022;13:3511.
Article CAS PubMed PubMed Central Google Scholar
Wang Z, Miao H, Liu J, Xu B, Yao X, Xu C, et al. Musa balbisiana genome reveals subgenome evolution and functional divergence. Nat Plants. 2019;5(8):810–21.
Article CAS PubMed PubMed Central Google Scholar
Wu H, Yu Q, Ran JH, Wang XQ. Unbiased subgenome evolution in allotetraploid species of Ephedra and its implications for the evolution of large genomes in gymnosperms. Genome Biol Evol. 2021;13(2):evaa236.
Article PubMed Google Scholar
Kamal N, Tsardakas Renhuldt N, Bentzer J, Gundlach H, Haberer G, Juhasz A, et al. The mosaic oat genome gives insights into a uniquely healthy cereal crop. Nature. 2022;606(7912):113–9.
Article CAS PubMed PubMed Central Google Scholar
Yoo MJ, Szadkowski E, Wendel JF. Homoeolog expression bias and expression level dominance in allopolyploid cotton. Heredity (Edinb). 2013;110(2):171–80.
Article CAS PubMed Google Scholar
Chalhoub B, Denoeud F, Liu S, Parkin IA, Tang H, Wang X, et al. Early allopolyploid evolution in the post-Neolithic Brassica napus oilseed genome. Science. 2014;345(6199):950–3.
Article CAS PubMed Google Scholar
Zhang T, Hu Y, Jiang W, Fang L, Guan X, Chen J, et al. Sequencing of allotetraploid cotton (Gossypium hirsutum L. acc. TM-1) provides a resource for fiber improvement. Nat Biotechnol. 2015;33(5):531–7.
Article CAS PubMed Google Scholar
Liu B, Shi Y, Yuan J, Hu X, Zhang H, Li N, et al. Estimation of genomic characteristics by analyzing k-mer frequency in de novo genome projects. arXiv. 2013. https://doi.org/10.48550/arXiv.1308.2012.
Cheng H, Concepcion GT, Feng X, Zhang H, Li H. Haplotype-resolved de novo assembly using phased assembly graphs with hifiasm. Nat Methods. 2021;18(2):170–5.
Article CAS PubMed PubMed Central Google Scholar
Dudchenko O, Batra SS, Omer AD, Nyquist SK, Hoeger M, Durand NC, et al. De novo asssembly of the Aedes aegypti genome using Hi-C yields chromosome-length scaffolds. Science. 2017;356:92–5.
Article CAS PubMed PubMed Central Google Scholar
Durand NC, Shamim MS, Machol I, Rao SS, Huntley MH, Lander ES, et al. Juicer provides a one-click system for analyzing loop-resolution Hi-C experiments. Cell Syst. 2016;3(1):95–8.
Article CAS PubMed PubMed Central Google Scholar
Manni M, Berkeley MR, Seppey M, Simao FA, Zdobnov EM. BUSCO Update: novel and streamlined workflows along with broader and deeper phylogenetic coverage for scoring of eukaryotic, prokaryotic, and viral genomes. Mol Biol Evol. 2021;38(10):4647–54.
Article CAS PubMed PubMed Central Google Scholar
Benson G. Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res. 1999;27(2):573–80.
Article CAS PubMed PubMed Central Google Scholar
Xu Z, Wang H. LTR_FINDER: an efficient tool for the prediction of full-length LTR retrotransposons. Nucleic Acids Res. 2007;35(Web Server issue):W265-8.
Article PubMed PubMed Central Google Scholar
Ellinghaus D, Kurtz S, Willhoeft U. LTRharvest, an efficient and flexible software for de novo detection of LTR retrotransposons. BMC Bioinformatics. 2008;9:18.
Article PubMed PubMed Central Google Scholar
Ou S, Jiang N. LTR_retriever: a highly accurate and sensitive program for identification of long terminal repeat retrotransposons. Plant Physiol. 2018;176(2):1410–22.
Article CAS PubMed Google Scholar
Lowe TM, Eddy SR. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 1997;25(5):955–64.
Article CAS PubMed PubMed Central Google Scholar
Nawrocki EP, Eddy SR. Infernal 1.1: 100-fold faster RNA homology searches. Bioinformatics. 2013;29(22):2933–5.
Article CAS PubMed PubMed Central Google Scholar
Kalvari I, Nawrocki EP, Argasinska J, Quinones-Olvera N, Finn RD, Bateman A, et al. Non-coding RNA analysis using the Rfam Database. Curr Protoc Bioinformatics. 2018;62(1): e51.
Article PubMed PubMed Central Google Scholar
Haas BJ, Delcher AL, Mount SM, Wortman JR, Smith RK Jr, Hannick LI, et al. Improving the Arabidopsis genome annotation using maximal transcript alignment assemblies. Nucleic Acids Res. 2003;31:5654–66.
Article CAS PubMed PubMed Central Google Scholar
Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, et al. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc. 2013;8(8):1494–512.
Article CAS PubMed Google Scholar
Stanke M, Keller O, Gunduz I, Hayes A, Waack S, Morgenstern B. AUGUSTUS: ab initio prediction of alternative transcripts. Nucleic Acids Res. 2006;34(Web Server issue):W435–9.
Article CAS PubMed PubMed Central Google Scholar
Gremme G, Brendel V, Sparks ME, Kurtz S. Engineering a software tool for gene structure prediction in higher organisms. Inform Software Tech. 2005;47(15):965–78.
Article Google Scholar
Kim D, Langmead B, Salzberg SL. HISAT: A fast spliced aligner with low memory requirements. Nat Methods. 2015;12(4):357–60.
Article CAS PubMed PubMed Central Google Scholar
Pertea M, Kim D, Pertea GM, Salzberg SL. Transcript-level expression analysis of RNA-seq experiments with HISAT. StringTie and Ballgown Nat Protoc. 2016;11(9):1650–67.
Article CAS PubMed Google Scholar
Haas BJ, Salzberg SL, Wei Z, Pertea M. Automated eukaryotic gene structure annotation using EVidenceModeler and the Program to Assemble Spliced Alignments. Genome Biol. 2008;9(1):R7.
Article PubMed PubMed Central Google Scholar
Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25(14):1754–60.
Article CAS PubMed PubMed Central Google Scholar
Emms DM, Kelly S. OrthoFinder: phylogenetic orthology inference for comparative genomics. Genome Biol. 2019;20(1):238.
Article PubMed PubMed Central Google Scholar
Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32(5):1792–7.
Article CAS PubMed PubMed Central Google Scholar
Jia KH, Wang ZX, Wang L, Li GY, Zhang W, Wang XL, et al. SubPhaser: a robust allopolyploid subgenome phasing method based on subgenome-specific k-mers. New Phytol. 2022;235(2):801–9.
Article CAS PubMed Google Scholar
Marcais G, Kingsford C. A fast, lock-free approach for efficient parallel counting of occurrences of k-mers. Bioinformatics. 2011;27(6):764–70.
Article CAS PubMed PubMed Central Google Scholar
Philip J, David B, Chang HY, Matthew F, Li W, Craig MA, et al. InterProScan 5: genome-scale protein function classification. Bioinformatics. 2014;9:1236–40.
Google Scholar
Matthias B, Chang HY, Sara C, Tiago G, Swaathi K, Alex M, et al. The InterPro protein families and domains database: 20 years on. Nucleic Acids Res. 2020;D1:D1.
Google Scholar
Aramaki T, Blanc-Mathieu R, Endo H, Ohkubo K, Kanehisa M, Goto S, et al. KofamKOALA: KEGG Ortholog assignment based on profile HMM and adaptive score threshold. Bioinformatics. 2020;36(7):2251–2.
Article CAS PubMed Google Scholar
Servant N, Varoquaux N, Lajoie BR, Viara E, Chen CJ, Vert JP, et al. HiC-Pro: an optimized and flexible pipeline for Hi-C data processing. Genome Biol. 2015;16:259.
Article PubMed PubMed Central Google Scholar
Price MN, Dehal PS, Arkin AP. FastTree 2-approximately maximum-likelihood trees for large alignments. PLoS ONE. 2010;5(3): e9490.
Article PubMed PubMed Central Google Scholar
Sanderson MJ. r8s: inferring absolute rates of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics. 2003;19:301–2.
Article CAS PubMed Google Scholar
Kumar S, Stecher G, Suleski M, Hedges SB. TimeTree: a resource for timelines, timetrees, and divergence times. Mol Biol Evol. 2017;34(7):1812–9.
Article CAS PubMed Google Scholar
Mira VH, Thomas GWC, Lugo-Martinez J, Hahn MW. Estimating gene gain and loss rates in the presence of error in genome assembly and annotation using CAFE 3. Mol Biol Evol. 2013;30(8):1987–97.
Article Google Scholar
Wang Y, Tang H, Debarry JD, Tan X, Li J, Wang X, et al. MCScanX: a toolkit for detection and evolutionary analysis of gene synteny and collinearity. Nucleic Acids Res. 2012;40(7): e49.
Article CAS PubMed PubMed Central Google Scholar
Tang H, Bowers JE, Wang X, Ming R, Alam M, Paterson AH. Synteny and collinearity in plant genomes. Science. 2008;320:486–8.
Article CAS PubMed Google Scholar
Pan S, Cheng S, Zhang B, Mu D, Ni P, Zhang G, et al. Genome sequence and analysis of the tuber crop potato. Nature. 2011;475(7355):189–95.
Article PubMed Google Scholar
Wang D, Zhang Y, Zhang Z, Zhu J, Yu J. KaKs_Calculator 2.0: a toolkit incorporating gamma-series methods and sliding window strategies. Genom Proteom Bioinf. 2010;8(1):77–80.
Article CAS Google Scholar
Pham SK, Pevzner PA. DRIMM-Synteny: decomposing genomes into evolutionary conserved segments. Bioinformatics. 2010;26(20):2509–16.
Article CAS PubMed Google Scholar
Gao S, Yang X, Sun J, Zhao X, Wang B, Ye K, et al. IAGS: inferring ancestor genome structure under a wide range of evolutionary scenarios. Mol Biol Evol. 2022;39(3):msac041.
Article CAS PubMed PubMed Central Google Scholar
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550.
Article PubMed PubMed Central Google Scholar
Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, von Haeseler A, et al. IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era. Mol Biol Evol. 2020;37(5):1530–4.
Article CAS PubMed PubMed Central Google Scholar
Letunic I, Bork P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 2021;49(W1):W293–6.
Article CAS PubMed PubMed Central Google Scholar
Marcais G, Delcher AL, Phillippy AM, Coston R, Salzberg SL, Zimin A. MUMmer4: A fast and versatile genome alignment system. PLoS Comput Biol. 2018;14(1): e1005944.
Article PubMed PubMed Central Google Scholar
Goel M, Sun H, Jiao WB, Schneeberger K. SyRI: finding genomic rearrangements and local sequence differences from whole-genome assemblies. Genome Biol. 2019;20(1):277.
Article PubMed PubMed Central Google Scholar
Goel M, Schneeberger K, Robinson P. plotsr: visualizing structural similarities and rearrangements between multiple genomes. Bioinformatics. 2022;38(10):2922–6.
Article CAS PubMed PubMed Central Google Scholar
Chen T, Chen X, Zhang S, Zhu J, Tang B, Wang A, et al. The Genome Sequence Archive Family: toward explosive data growth and diverse data types. Genom Proteom Bioinf. 2021;19(4):578–83.
Article Google Scholar
CNCB-NGDC Members, Partners. Database Resources of the National Genomics Data Center, China National Center for Bioinformation in 2022. Nucleic Acids Res. 2022;50(D1):D27–38.
Article Google Scholar
Xing L. The genome assembly and annotation data of Pennisetum giganteum. Figshare; 2022. https://doi.org/10.6084/m9.figshare.23118794.

Download references

Acknowledgements

We thank Margaret Biswas, PhD, from Liwen Bianji (Edanz) (www.liwenbianji.cn/) for editing the English text of a draft of this manuscript. We thank the anonymous reviewers for their critical comments and suggestions.

Funding

This work is supported by the Natural Science Foundation of Hebei Province (C2021201048) and the National Natural Science Foundation of China (32100500).

Author information

Longsheng Xing and Meijia Wang contributed equally to this work.

Authors and Affiliations

College of Life Sciences, Institute of Life Sciences and Green Development, Hebei University, Baoding, 071000, China
Longsheng Xing, Meijia Wang, Qiang He, Hongyu Zhang, Hanfei Liang, Qinghong Zhou, Yu Liu, Ze Liu, Yu Wang, Cailian Du, Yao Xiao, Jianan Liu, Wei Li, Guixia Liu & Huilong Du
Hebei Basic Science Center for Biotic Interaction, Baoding, 071000, China
Longsheng Xing, Qiang He, Wei Li, Guixia Liu & Huilong Du

Authors

Longsheng Xing
View author publications
You can also search for this author in PubMed Google Scholar
Meijia Wang
View author publications
You can also search for this author in PubMed Google Scholar
Qiang He
View author publications
You can also search for this author in PubMed Google Scholar
Hongyu Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Hanfei Liang
View author publications
You can also search for this author in PubMed Google Scholar
Qinghong Zhou
View author publications
You can also search for this author in PubMed Google Scholar
Yu Liu
View author publications
You can also search for this author in PubMed Google Scholar
Ze Liu
View author publications
You can also search for this author in PubMed Google Scholar
Yu Wang
View author publications
You can also search for this author in PubMed Google Scholar
Cailian Du
View author publications
You can also search for this author in PubMed Google Scholar
Yao Xiao
View author publications
You can also search for this author in PubMed Google Scholar
Jianan Liu
View author publications
You can also search for this author in PubMed Google Scholar
Wei Li
View author publications
You can also search for this author in PubMed Google Scholar
Guixia Liu
View author publications
You can also search for this author in PubMed Google Scholar
Huilong Du
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

H.D. and G.L. conceived and supervised the project; H.D. and L.X. designed the paper; L.X., M.W., and H.D. wrote the paper; M.W., Q.H., H.Z., H.L., Y.L., and Q.Z. sequenced and processed the raw data; M.W., L.X., W.L., Q.H., Z.L., and H.L. assembled and annotated the genome; L.X., M.W., and Y.W. performed the phylogenetic and genome evolution analysis; M.W., L.X., and C.D. analyzed the gene families; M.W., L.X., Y.X., J.L., and W.L. conducted the transcriptome analysis. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Guixia Liu or Huilong Du.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

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

Supplementary Information

Additional file 1: Table S1.

Summary of DNA sequencing data for Pennisetum giganteum. Table S2. Genome size estimation result of P. giganteum based on k-mer frequency analysis. Table S3. Statistics of the P. giganteum genome assembly. Table S4. The number of contigs and gaps in chromosomes of P. giganteum. Table S5. Statistics of chromosome-anchored contigs in subgenomes of P. giganteum. Table S6. BUSCO analysis of P. giganteum genome assembly. Table S7. The mapping rate of RNA-seq data to the P. giganteum genome. Table S8. Summary of homozygous SNPs detected in genome resequencing data of P. giganteum. Table S9. Statistics of repeat contents in the A and B subgenomes of P. giganteum. Table S10. Overview of protein-coding genes and transposable elements on chromosomes of A and B subgenomes. Table S11. Summary of intact LTR-RTs identified in the P. giganteum genome. Table S12. Statistics of gene annotation in P. giganteum. Table S13. Functional annotation of the predicted P. giganteum genes. Table S14. Mapping rates of pearl millet short reads on chromosomes of P. giganteum. Table S15. Summary of subgenome-specifically enriched 13-mers in P. giganteum. Table S16. Orthologous groups identified in P. giganteum and other plant species. Table S17. Distribution of different categories of gene families in P. giganteum and other plant species. Table S18. GO enrichment analysis of expanded genes of subgenome A and B in P. giganteum. Table S19. KEGG enrichment analysis of expanded genes of subgenome A and B in P. giganteum. Table S20. GO enrichment analysis of conserved genes of subgenome A in P. giganteum. Table S21. GO enrichment analysis of conserved genes of subgenome B in P. giganteum. Table S22. GO enrichment analysis of unique genes P. giganteum A and B compared with C. americanus. Table S23. GO enrichment analysis of unique genes in P. giganteum A and B compared with S. viridis. Table S24. GO enrichment analysis of unique genes in P. giganteum A and B compared with S. bicolor. Table S25. GO enrichment analysis of unique genes in P. giganteum A and B. Table S26. Statistics of number and length of SVs between different plant species. Table S27. Summary of highly conversed genes of A and B subgenome in P. giganteum relative to C. americanus and S. viridis. Table S28. Summary of the ratio of 3D chromatin A/B compartments between the syntenic gene pairs of A and B subgenomes. Table S29. Comparison of gene and exon in different tissues of P. giganteum. Table S30. The number of highly expressed genes in subgenomes of P. giganteum. Table S31. Summary of differentially expressed genes in inter-tissue comparison groups. Table S32. Summary of expressed homoeologous genes in different tissues of the AB subgenome. Table S33. Summary of homoeolog-specific genes in different tissues of the AB subgenome. Table S34. Summary of cellulose synthase related gene families in P. giganteum and S. viridis. Table S35. Summary of expression levels of cellulose synthesis related genes in different tissues of P. giganteum. Table S36. Distribution of cellulose synthase genes on chromosomes of P. giganteum. Table S37. Comparison of expression levels of homoeologous gene pairs between the A and B subgenomes in P. giganteum. Table S38. Summary of differentially expressed genes in seven stages of P. giganteum. Table S39. GO enrichment analysis of DEGs in leaves at seven stages in P. giganteum. Table S40. GO enrichment analysis of DEG in stems at seven stages in P. giganteum.Table S41. Statistics of all and full-length LTRs in A5-B14 syntenic regions. Table S42. Homoeolog expression dominance in P. giganteum and three other polyploid species.

Additional file 2: Figure S1.

Interaction frequency distribution of Hi-C links among chromosomes. The Hi-C version of Pennisetum giganteum genome assembly was generated by 3D-DNA pipeline. Figure S2. K-mer frequency distribution. The sharp peak depthwas 40 in the distribution. Figure S3. Mapping rates of C. americanus short reads on chromosomes of P. giganteum. Figure S4. Unsupervised hierarchical clustering of differential 15-mer sequences for each chromosome in P. giganteum. The clustering result confirmed that the genome is successfully phased into two subgenomes based on distinct patterns of differential k-mers. Figure S5. Phylogenetic trees of homologous chromosomes among the two subgenomes of P. giganteum, C. americanus, and S. viridis based on single-copy orthologs. All phylogenetic trees displayed a similar topology and validated that subgenome A of P. giganteum showed a closer phylogenetic relationship with C. americanus than did subgenome B. Figure S6. The syntenic blocks across two subgenomes of P. giganteum, C. americanus, S. viridis and S. bicolor. The circled numbers below the ideograms marked the big syntenic blocksthat were selected for constructing phylogenetic trees based on syntenic gene pairs. Figure S7. Phylogenetic analysis based on homologous genes within syntenic blocks across two subgenomes of P. giganteum, C. americanus, S. viridis and S. bicolor. The chromosome numberings refer to the block division of Figure S6. Figure S8. Bar plot showing the numbers of genes encoding proteins containing the Pfam domain PF00195 that are involved in flavonoid biosynthesis with detectable expression in the six tissues of P. giganteum. P-values were calculated using the Fisher’s exact test. Figure S9. GO and KEGG enrichment analysis of conserved genes between subgenomes A and B. GO enrichment analysis of conserved genes in subgenome A. GO enrichment analysis of conserved genes in subgenome B. KEGG enrichment analysis of conserved genes in subgenome A. KEGG enrichment analysis of conserved genes in subgenome B. Figure S10. GO enrichment analysis of unique genes between P. giganteum A, B and three related species. GO enrichment analysis of unique genes in P. giganteum A. GO enrichment analysis of unique genes in P. giganteum B. GO enrichment analysis of unique genes in P. giganteum A. GO enrichment analysis of unique genes in P. giganteum B. GO enrichment analysis of unique genes in P. giganteum A. GO enrichment analysis of unique genes in P. giganteum B. Figure S11. GO enrichment analysis of unique genes between P. giganteum A and B. GO enrichment analysis of unique genes in P. giganteum A. GO enrichment analysis of unique genes in P. giganteum B. Figure S12. Boxplot showing the average expression of three classes of genes in two subgenomes of P. giganteum, including conserved genes, unique genes and other genes. **, significant; ****, extremely significant; ns, not significant. Figure S13. Classification of gene duplicates in the A and B subgenomes of P. giganteum and P. purpureum. The origins of duplicated genes were categorized into five types: WGD/segmental duplication, tandem duplication, proximal duplication, dispersed duplication, and singleton. P-values were calculated using the Fisher’s exact test. Figure S14. Genome-wide syntenic analysis between C. americanus and two subgenomes of P. giganteum. Synteny between P. giganteum A and C. americanus. Synteny between P. giganteum B and C. americanus. Pairwise whole-genome alignments across the A and B subgenomes of P. giganteum and C. americanus. Figure S15. Comparison of conserved gene numbers between two subgenomes of P. giganteum and C. americanus, S. viridis, respectively. Figure S16. Chromosome rearrangements and validation by Hi-C interaction evidence. The verification of inverted fragments involved in chromosome rearrangement was performed using Hi-C heatmap between two subgenomes of P. giganteum. Collinearity and chromosome rearrangements between the A and B subgenomes of P. giganteum. Figure S17. Inference of centromere positions based on the distribution of three kinds of genomic features along the 14 P. giganteum chromosomes. Orange curves indicate the density of long terminal repeat retrotransposonson each chromosome of P. giganteum with a window size of 1 Mb. Green curves indicate the distribution protein-coding genes on each chromosome of P. giganteum. Blue curves indicate the distribution of the most abundant 20-k-mers along each chromosome. Red dashed lines corresponds to the putative centromeric region of each P. giganteum chromosome. Figure S18. Distribution of A/B compartment status conservation or switching in P. giganteum. Stacked bar chart showing the distribution of different types of chromatin A/B compartment status for homologous chromosome pairs in P. giganteum. A-to-A and B-to-B represent conserved chromatin compartment status between syntenic gene pairs, while A-to-B and B-to-A denote the switching of chromatin compartment status between syntenic gene pairs. Comparison of A/B compartment status switching between non-rearranged and rearranged genomic regions. Statistical significance was determined using one-tailed Fisher’s exact test. Figure S19. Preliminary analysis of six tissues for differential expression analysis in P. giganteum. Correlation analysis for biological replicates of six tissues in P. giganteum. PCA clustering of all RNA-seq samples in P. giganteum. Figure S20. Preliminary analysis of six tissues for differential expression analysis of subgenome A and B in P. giganteum. Distribution of normalized expression levels in six tissues of P. giganteum. Correlation analysis for biological replicates of six tissues in P. giganteum. PCA clustering of all RNA-seq samples in P. giganteum A and B . Figure S21. GO enrichment of differentially expressed genesbetween paired tissues in P. giganteum. Enriched biological processes for DEGs between paired tissues in P. giganteum. Enriched molecular functions for DEGs between paired tissues in P. giganteum. Enriched cellular components for DEGs between paired tissues in P. giganteum. Figure S22. Identification of DEGs and GO enrichment analysis. Upset plot showing DEGs among six tissues in the A subgenome of P. giganteum. Upset plot showing DEGs among six tissues in the B subgenome of P. giganteum. GO enrichment of DEGs of six tissues in the A and B subgenomes of P. giganteum. Figure S23. Identification of homoeologous differentially expressed genesand clustering analysis. Upset plot of HDEGs among six tissues of the A subgenome in P. giganteum. Upset plot of HDEGs among six species of the B subgenome in P. giganteum. Heatmap showing the expression pattern of consistent group of homoeolog-specific expressed genesof six tissues in the A and B subgenomes of P. giganteum. Figure S24. Comparison of gene level characteristics between the A and B subgenomes of P. giganteum. Figure S25. Preliminary analysis of seven stages for differential expression analysis in P. giganteum. Correlation analysis for biological replicates of leaf samples at seven stages in P. giganteum. PCA clustering of all leaf RNA-seq samples in P. giganteum. Figure S26. Expression patterns of differentially expressed genes in leaves at different growth stages. The heatmap shows the log2-based FPKM+0.01 at seven growth stages. Clustering heatmap of differentially expressed genes in leaves at different growth stages. Figure S27. Preliminary analysis of stem RNA-seq samples at seven stages for differential expression analysis in P. giganteum. Correlation analysis for biological replicates of stem samples at seven stages in P. giganteum. PCA clustering of all stem RNA-seq samples in P. giganteum. Figure S28. Clustering heatmap of differentially expressed genes in stems at different growth stages. Expression of differentially expressed genes in stems at different growth stages. The heatmap shows the log2-based FPKM+0.01 at seven growth stages. Figure S29. Expression of cellulose synthase-related genes in leaves at different growth stages. The heatmap shows the log2-based FPKM+0.01 at seven growth stages. Figure S30. Expression of cellulose synthase-related genes in stems at different growth stages. The heatmap shows the log2-based FPKM+0.01 at seven growth stages. Gene expression heatmap for cellulose synthase genes that were expressed at least one time point in stems. Figure S31. Distribution of intact LTRs in P. giganteum and comparison of LTRs in the syntenic region of chromosomes A5 and B14. Boxplot showing the distribution of LTR Assembly Indexscores on each chromosome of P. giganteum. The gray dotted line represents the average LAI score across the whole genome. The density plot of insertion time of Gypsy and Copia retrotransposons in C. americanus and subgenomes A and B of P. giganteum. The Gypsy-type LTR peak positions of the three species were very close, while the Copia-type LTR peak positions differed greatly between C. americanus and P. giganteum, thus resulting in the two peaks of LTR insertion in P. giganteum. The distribution of repeat sequences in syntenic regions of A5and B14. Intact LTRs are shown as green blocks and genes are indicated as blue blocks. The distribution of repeat sequences in a microsynteny block of A5and B14. Intact LTRs are shown as red blocks and other types of TEs are indicated as blue blocks. Figure S32. Comparison of differentially expressed genesand balanced genes between subgenomes across different polyploid species. Distribution of DEGs and balanced genes between subgenomes in different studies. Distribution of DEGs and balanced genes between subgenomes in different studies using the same method as used by us.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article

Xing, L., Wang, M., He, Q. et al. Differential subgenome expression underlies biomass accumulation in allotetraploid Pennisetum giganteum. BMC Biol 21, 161 (2023). https://doi.org/10.1186/s12915-023-01643-w

Download citation

Received: 11 January 2023
Accepted: 06 June 2023
Published: 21 July 2023
DOI: https://doi.org/10.1186/s12915-023-01643-w

Differential subgenome expression underlies biomass accumulation in allotetraploid Pennisetum giganteum

Abstract

Background

Results

Conclusions

Background

Results

High-quality genome assembly and separate A and B subgenomes of P. giganteum

Comparative genomics and gene family evolution analysis

Comparative genomic analysis between P. giganteum and its closely related species P. purpureum

Genome evolution analysis of the A and B subgenomes

Analysis of differentially expressed homoeologous genes between the A and B subgenomes

Strengthened cellulose synthesis and C4 photosynthesis in P. giganteum

Discussion

Conclusions

Methods

Plant materials, DNA and RNA sample collection and sequencing

Genome assembly and quality assessment

Genome annotation

Subgenome division of the P. giganteum genome

Functional annotation

Hi-C read mapping and normalization

Identification of A and B compartments

Variant calling and validation of accuracy rate of single base

Phylogenetic tree construction and evolutionary rate estimation

Gene family analysis

Whole-genome duplication (WGD) and synteny analysis

Ancestor genome structure inference among P. giganteum and other related grass species

Genome-wide differential expression analysis

Comparative genomic analysis of P. giganteum and P. purpureum

Availability of data and materials

Abbreviations

References

Acknowledgements

Funding

Author information

Authors and Affiliations

Contributions

Corresponding authors

Ethics declarations

Ethics approval and consent to participate

Consent for publication

Competing interests

Additional information

Publisher's Note

Supplementary Information

Additional file 1: Table S1.

Additional file 2: Figure S1.

Rights and permissions

About this article

Cite this article

Share this article

Keywords

BMC Biology

Contact us