doi: 10.1111/age.12294
Transcriptome analysis of adiposity in domestic ducks by transcriptomic comparison with their wild counterparts L. Chen*, J. Luo†, J. X. Li‡, J. J. Li*, D. Q. Wang*, Y. Tian* and L. Z. Lu* *Institute of Animal Sciences and Veterinary Medicine, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China. †Institute of Digital Agricultural Research, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China. ‡State Key Laboratory for Agro-biotechnology, China Agricultural University, Beijing 100193, China.
Summary
Excessive adiposity is a major problem in the duck industry, but its molecular mechanisms remain unknown. Genetic comparisons between domestic and wild animals have contributed to the exploration of genetic mechanisms responsible for many phenotypic traits. Significant differences in body fat mass have been detected between domestic and wild ducks. In this study, we used the Peking duck and Anas platyrhynchos as the domestic breed and wild counterpart respectively and performed a transcriptomic comparison of abdominal fat between the two breeds to comprehensively analyze the transcriptome basis of adiposity in ducks. We obtained approximately 350 million clean reads; assembled 61 250 transcripts, including 23 699 novel ones; and identified alternative 50 splice sites, alternative 30 splice sites, skipped exons and retained intron as the main alternative splicing events. A differential expression analysis between the two breeds showed that 753 genes exhibited differential expression. In Peking ducks, some lipid metabolism-related genes (IGF2, FABP5, BMP7, etc.) and oncogenes (RRM2, AURKA, CYR61, etc.) were upregulated, whereas genes related to tumor suppression and immunity (TNFRSF19, TNFAIP6, IGSF21, NCF1, etc.) were downregulated, suggesting adiposity might closely associate with tumorigenesis in ducks. Furthermore, 280 576 single-nucleotide variations were found differentiated between the two breeds, including 8641 non-synonymous ones, and some of the non-synonymous ones were found enriched in genes involved in lipidassociated and immune-associated pathways, suggesting abdominal fat of the duck undertakes both a metabolic function and immune-related function. These datasets enlarge our genetic information of ducks and provide valuable resources for analyzing mechanisms underlying adiposity in ducks. Keywords abdominal fat, Anas platyrhynchos, RNA-seq, single-nucleotide variations
Introduction The domestic duck is an economically important waterfowl that provides meat, eggs and feathers to humans. Anas platyrhynchos is believed to be the major ancestor of domestic ducks (He et al. 2008; Li et al. 2010). After domestication and breeding, numerous domestic duck breeds were developed (Chen et al. 2004). Compared with their wild ancestors, these domestic breeds have undergone remarkable phenotypic changes in morphology, physiology Address for correspondence L. Z. Lu, Institute of Animal Sciences and Veterinary Medicine, Zhejiang Academy of Agricultural Sciences, NO. 198 Shiqiao Road, Hangzhou 310021, China. E-mail: [email protected] Accepted for publication 26 February 2015
and behavior, such as reduced fearfulness, earlier sexual maturation and rapid growth, but this has been accompanied by excessive adiposity. Excessive adiposity is a major problem in the poultry industry, as it reduces feed efficiency and carcass yield, leading to rejection by consumers (Zhang et al. 2013). Genetic mechanisms of adiposity have been extensively investigated in chickens (for example, see Bourneuf et al. 2006; Wang et al. 2010; Resnyk et al. 2013). Nevertheless, they remain poorly understood in ducks. Animal domestication is accompanied by many genetic changes; hence, this process provides a valuable model for investigating genes that shape phenotypic changes. Genetic comparisons between domestic animals and their wild counterparts have contributed to the exploration of the molecular mechanisms responsible for many phenotypic traits. For example, performing comparative whole-genome
© 2015 Stichting International Foundation for Animal Genetics, 46, 299–307
299
300
Chen et al. resequencing between domestic chickens representing eight populations and their major wild ancestor (Red Junglefowl), Rubin et al. (2010) have identified selective sweeps shared by populations with the same traits, one involving a selective sweep in the TSHR gene, which was found in all domestic chickens and may be related to the absence of seasonal reproduction in domestic animals. Using QTL analysis and crossing the Red Junglefowl with a domestic white-egg layer (White Leghorn chickens), Kerje et al. (2003) have identified a number of QTL that determine phenotypic differences between the two breeds. One particular region on chromosome 1 was shown to have effects on growth, egg production and feed consumption. Similar studies were conducted on wheat, maize, beans, rice and pigs (Andersson & Georges 2004; Pickersgill 2007). In addition, numerous genes differentially expressed between wild and domestic animals have been identified (Cockett et al. 1996; McPherron et al. 1997; Albert et al. 2012). All these successful works provide a new perspective for genetic studies on adiposity in ducks. In this study, a transcriptomic comparison between domestic ducks and their wild counterparts was conducted to comprehensively exploit the genetic basis of adiposity in domestic ducks. For this, Peking ducks and Anas platyrhynchos were selected as domestic and wild ducks respectively, and abdominal fat samples from the two breeds were chosen for analysis. By RNA-seq, we obtained the transcriptome data of duck adipose tissue and analyzed differentially expressed genes and differentiated single-nucleotide variations (SNVs) between the two breeds. Taken collectively, the first transcriptomic analysis of the duck adipose tissue enlarges our genetic information about ducks and provides valuable resources for further studies on adiposity in ducks.
Material and methods Animal preparation and tissue collection Peking ducks and Anas platyrhynchos were used in this study, representing domestic ducks and their wild counterparts respectively. Peking duck is a famous domestic breed characterized by its high rate of growth and excessive adiposity. Anas platyrhynchos is believed to be the major ancestor of domestic ducks and is distinguished by its low percentage of abdominal fat. Anas platyrhynchos ducklings were supplied by a local duck farm (Aoji Duck Farm, Ningbo) that has hunting permission granted by the local government. Male Peking ducks and Anas platyrhynchos were kept together from duckling to adulthood, fed a commercial corn–soybean-based diet ad libitum and fasted for 8 h before being slaughtered. Eight-week-old birds were slaughtered according to the national and institutional guidelines for the ethical conduct and treatment of experimental animals. Abdominal fat samples were collected,
immediately frozen in liquid nitrogen and then stored at 80 °C.
RNA extraction and sequencing Six birds representing three biological replicates per breed (Peking ducks and Anas platyrhynchos) were used in this experiment. Total RNA was extracted from approximately 300 mg of frozen abdominal adipose tissue using the RNeasy Lipid Tissue Mini Kit (QIAGEN), following the manufacturer’s instructions, quantified using the NanoDrop1000 spectrophotometer (Thermo Fisher Scientific) and then evaluated for purity and integrity using the Bioanalyzer-2100 (Agilent Technologies). Total RNA from each individual was used to construct the library as previously described (Yang et al. 2012). Subsequently, paired-end sequencing of each library was performed using the Illumina HiSeq 2000 platform. All the sequence data were submitted to the National Center for Biotechnology Information (NCBI) database Short Read Archive (accession number SRP042174).
Histological analysis Abdominal fat tissues from the six birds were fixed overnight in 4% paraformaldehyde, dehydrated with ethanol, embedded in paraffin and then sectioned (5 mm). Hematoxylin and eosin staining was performed using standard protocols, and samples were evaluated via light microscopy.
Sequencing data analysis After removing sequencing adaptors and low-quality reads, high-quality reads were mapped against the duck genome assembly BGI_duck_1.0 using TOPHAT v2.0.9 software (Trapnell et al. 2009). Assembly of mapped reads into transcripts, transcript annotation, and their expression level calculation was processed using CUFFLINKS v2.1.1 (Trapnell et al. 2010) with default parameters for the reference annotation based on the transcript assembly method. After the completion of transcript assembly, transcripts.gtf files from CUFFLINKS and duck reference annotations were used as input to CUFFCOMPARE, which can classify transcripts as known or novel transcripts. Finally, the all.combined.gtf file from CUFFCOMPARE and mapping files from TOPHAT were used by CUFFDIFF for differential analysis. Both CUFFCOMPARE and CUFFDIFF were run with default parameters. SNVs and small insertions and deletions (indels) were discovered using SAMTOOLS 0.1.19 (Li et al. 2009) with minimum and maximum depths of four and 100 respectively. SNV filtering was performed with BEDTOOLS 2.17.0. TOPHAT was used to analyze the alternative splicing as previously described (Trapnell et al. 2009). A heatmap showing the top 50 differentially expressed genes (DEGs) was generated based on log10-transformed RPKM (reads per kilobase per million) values using MEV4.8.1.
© 2015 Stichting International Foundation for Animal Genetics, 46, 299–307
RNA-seq of duck abdominal fats
GO enrichment and pathway analysis Differentially expressed genes were mapped to the Gene Ontology (GO) terms that were downloaded from the GO database (http://www.geneontology.org/), and hypergeometric test P-values were calculated to identify the GO terms that were overrepresented in the DEGs. GO terms with P < 0.05 were considered significantly enriched. DEGs together with genes containing non-synonymous SNVs were also included in the pathway enrichment analysis using KEGG Mapper.
two breeds exhibited different abdominal fat percentages, that is, 1.20% in Peking ducks and 0.68% in Anas platyrhynchos. Then, histological analysis of these abdominal fat tissues showed that the size of adipocytes from Peking ducks was larger than that of adipocytes from Anas platyrhynchos (Fig. 1b); consistent with this finding was the decrease in cell number (Fig. 1c). These results suggest that Peking ducks accumulate more abdominal fat than do their wild counterparts.
Global view of transcriptome sequencing Real-time PCR analysis Real-time PCR analysis was performed using the 7900 Sequence Detection System (Applied Biosystems) under the following conditions: 95 °C for 10 min, followed by 40 cycles of 95 °C for 30 s and 60 °C for 1 min. The primer sequences for real-time PCR analysis were designed using PRIMER EXPRESS (v3.0) (Applied Biosystems) and are listed in Table S1. Five birds from each breed were used in this experiment, and the relative expression of each gene was normalized to the internal control gene EEF1A1 (ENSAPLG00000009349) using the 2DDCt method (Livaka & Schmittgen 2001).
Results Differences in abdominal fat between Peking ducks and Anas platyrhynchos To identify differences in abdominal fat between Peking ducks and Anas platyrhynchos, abdominal fat percentages were calculated in the two breeds. As shown in Fig. 1a, the
To globally understand the transcriptome changes underlying abdominal fat variation between Peking ducks and Anas platyrhynchos, transcriptome sequencing of abdominal fats from the two breeds was performed using the Illumina HiSeq 2000 platform with three biological replicates per breed. After removing sequencing adaptors and low-quality reads, approximately 350 million clean reads were obtained. These clean reads were then used for mapping to the duck reference genome. According to the mapping results, approximately 64% of the clean reads were successfully mapped to the duck genome. Of the mapped reads, 98% were uniquely aligned (Table 1). All of the mapped reads were further assembled into transcripts using CUFFLINKS, generating 61 250 transcripts, including 23 699 novel ones. Reads per kilobase per million of mapped reads is the normalized quantitative method for gene expression in RNA-seq (Mortazavi et al. 2008). To overview the gene expression in each sample, RPKM was calculated for each gene. Based on the RPKM value, 18 000 to 18 866 genes
(c) 1.4
**
160
1.2
Number of adipocyte
Abdominal fat percentage (%)
(a)
1 0.8 0.6 0.4 0.2
140 120 100 80 60
*
40 20 0
0 PD
PD
AP
AP
(b) Figure 1 Differences in abdominal fat between Peking duck (PD) and Anas platyrhynchos (AP). (a) Difference in abdominal fat percentage between the two breeds. Data are presented as mean SD (n = 5). (b) Representative sections of hematoxylin-and-eosinstained abdominal fats from PD and AP at 4009 magnification. (c) Difference in adipocyte number between the two breeds. Data are presented as mean SD (n = 3). *P < 0.05; **P < 0.01. © 2015 Stichting International Foundation for Animal Genetics, 46, 299–307
PD
AP
301
Chen et al. Table 1 Overview of RNA sequencing, mapping and assembly.
Sample
Clean reads
Reads mapped on the genome (%)
Reads uniquely mapped on the genome (%)
AP3 AP4 AP5 PD3 PD4 PD5
61 453 324 37 003 864 49 083 112 109 355 142 44 991 866 48 853 976
66.49 62.49 65.18 62.60 66.29 63.18
98.00 97.92 98.06 97.68 98.11 97.92
Transcripts
Novel transcripts
61 250
23 699
Note: PD3, PD4 and PD5 represent three biological replicates of Peking ducks; AP3, AP4 and AP5 represent three biological replicates of their wild counterparts (Anas platyrhynchos). 10 000
RPKM
Transcriptome analysis of adiposity in domestic ducks by transcriptomic comparison with their wild counterparts L. Chen*, J. Luo†, J. X. Li‡, J. J. Li*, D. Q. Wang*, Y. Tian* and L. Z. Lu* *Institute of Animal Sciences and Veterinary Medicine, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China. †Institute of Digital Agricultural Research, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China. ‡State Key Laboratory for Agro-biotechnology, China Agricultural University, Beijing 100193, China.
Summary
Excessive adiposity is a major problem in the duck industry, but its molecular mechanisms remain unknown. Genetic comparisons between domestic and wild animals have contributed to the exploration of genetic mechanisms responsible for many phenotypic traits. Significant differences in body fat mass have been detected between domestic and wild ducks. In this study, we used the Peking duck and Anas platyrhynchos as the domestic breed and wild counterpart respectively and performed a transcriptomic comparison of abdominal fat between the two breeds to comprehensively analyze the transcriptome basis of adiposity in ducks. We obtained approximately 350 million clean reads; assembled 61 250 transcripts, including 23 699 novel ones; and identified alternative 50 splice sites, alternative 30 splice sites, skipped exons and retained intron as the main alternative splicing events. A differential expression analysis between the two breeds showed that 753 genes exhibited differential expression. In Peking ducks, some lipid metabolism-related genes (IGF2, FABP5, BMP7, etc.) and oncogenes (RRM2, AURKA, CYR61, etc.) were upregulated, whereas genes related to tumor suppression and immunity (TNFRSF19, TNFAIP6, IGSF21, NCF1, etc.) were downregulated, suggesting adiposity might closely associate with tumorigenesis in ducks. Furthermore, 280 576 single-nucleotide variations were found differentiated between the two breeds, including 8641 non-synonymous ones, and some of the non-synonymous ones were found enriched in genes involved in lipidassociated and immune-associated pathways, suggesting abdominal fat of the duck undertakes both a metabolic function and immune-related function. These datasets enlarge our genetic information of ducks and provide valuable resources for analyzing mechanisms underlying adiposity in ducks. Keywords abdominal fat, Anas platyrhynchos, RNA-seq, single-nucleotide variations
Introduction The domestic duck is an economically important waterfowl that provides meat, eggs and feathers to humans. Anas platyrhynchos is believed to be the major ancestor of domestic ducks (He et al. 2008; Li et al. 2010). After domestication and breeding, numerous domestic duck breeds were developed (Chen et al. 2004). Compared with their wild ancestors, these domestic breeds have undergone remarkable phenotypic changes in morphology, physiology Address for correspondence L. Z. Lu, Institute of Animal Sciences and Veterinary Medicine, Zhejiang Academy of Agricultural Sciences, NO. 198 Shiqiao Road, Hangzhou 310021, China. E-mail: [email protected] Accepted for publication 26 February 2015
and behavior, such as reduced fearfulness, earlier sexual maturation and rapid growth, but this has been accompanied by excessive adiposity. Excessive adiposity is a major problem in the poultry industry, as it reduces feed efficiency and carcass yield, leading to rejection by consumers (Zhang et al. 2013). Genetic mechanisms of adiposity have been extensively investigated in chickens (for example, see Bourneuf et al. 2006; Wang et al. 2010; Resnyk et al. 2013). Nevertheless, they remain poorly understood in ducks. Animal domestication is accompanied by many genetic changes; hence, this process provides a valuable model for investigating genes that shape phenotypic changes. Genetic comparisons between domestic animals and their wild counterparts have contributed to the exploration of the molecular mechanisms responsible for many phenotypic traits. For example, performing comparative whole-genome
© 2015 Stichting International Foundation for Animal Genetics, 46, 299–307
299
300
Chen et al. resequencing between domestic chickens representing eight populations and their major wild ancestor (Red Junglefowl), Rubin et al. (2010) have identified selective sweeps shared by populations with the same traits, one involving a selective sweep in the TSHR gene, which was found in all domestic chickens and may be related to the absence of seasonal reproduction in domestic animals. Using QTL analysis and crossing the Red Junglefowl with a domestic white-egg layer (White Leghorn chickens), Kerje et al. (2003) have identified a number of QTL that determine phenotypic differences between the two breeds. One particular region on chromosome 1 was shown to have effects on growth, egg production and feed consumption. Similar studies were conducted on wheat, maize, beans, rice and pigs (Andersson & Georges 2004; Pickersgill 2007). In addition, numerous genes differentially expressed between wild and domestic animals have been identified (Cockett et al. 1996; McPherron et al. 1997; Albert et al. 2012). All these successful works provide a new perspective for genetic studies on adiposity in ducks. In this study, a transcriptomic comparison between domestic ducks and their wild counterparts was conducted to comprehensively exploit the genetic basis of adiposity in domestic ducks. For this, Peking ducks and Anas platyrhynchos were selected as domestic and wild ducks respectively, and abdominal fat samples from the two breeds were chosen for analysis. By RNA-seq, we obtained the transcriptome data of duck adipose tissue and analyzed differentially expressed genes and differentiated single-nucleotide variations (SNVs) between the two breeds. Taken collectively, the first transcriptomic analysis of the duck adipose tissue enlarges our genetic information about ducks and provides valuable resources for further studies on adiposity in ducks.
Material and methods Animal preparation and tissue collection Peking ducks and Anas platyrhynchos were used in this study, representing domestic ducks and their wild counterparts respectively. Peking duck is a famous domestic breed characterized by its high rate of growth and excessive adiposity. Anas platyrhynchos is believed to be the major ancestor of domestic ducks and is distinguished by its low percentage of abdominal fat. Anas platyrhynchos ducklings were supplied by a local duck farm (Aoji Duck Farm, Ningbo) that has hunting permission granted by the local government. Male Peking ducks and Anas platyrhynchos were kept together from duckling to adulthood, fed a commercial corn–soybean-based diet ad libitum and fasted for 8 h before being slaughtered. Eight-week-old birds were slaughtered according to the national and institutional guidelines for the ethical conduct and treatment of experimental animals. Abdominal fat samples were collected,
immediately frozen in liquid nitrogen and then stored at 80 °C.
RNA extraction and sequencing Six birds representing three biological replicates per breed (Peking ducks and Anas platyrhynchos) were used in this experiment. Total RNA was extracted from approximately 300 mg of frozen abdominal adipose tissue using the RNeasy Lipid Tissue Mini Kit (QIAGEN), following the manufacturer’s instructions, quantified using the NanoDrop1000 spectrophotometer (Thermo Fisher Scientific) and then evaluated for purity and integrity using the Bioanalyzer-2100 (Agilent Technologies). Total RNA from each individual was used to construct the library as previously described (Yang et al. 2012). Subsequently, paired-end sequencing of each library was performed using the Illumina HiSeq 2000 platform. All the sequence data were submitted to the National Center for Biotechnology Information (NCBI) database Short Read Archive (accession number SRP042174).
Histological analysis Abdominal fat tissues from the six birds were fixed overnight in 4% paraformaldehyde, dehydrated with ethanol, embedded in paraffin and then sectioned (5 mm). Hematoxylin and eosin staining was performed using standard protocols, and samples were evaluated via light microscopy.
Sequencing data analysis After removing sequencing adaptors and low-quality reads, high-quality reads were mapped against the duck genome assembly BGI_duck_1.0 using TOPHAT v2.0.9 software (Trapnell et al. 2009). Assembly of mapped reads into transcripts, transcript annotation, and their expression level calculation was processed using CUFFLINKS v2.1.1 (Trapnell et al. 2010) with default parameters for the reference annotation based on the transcript assembly method. After the completion of transcript assembly, transcripts.gtf files from CUFFLINKS and duck reference annotations were used as input to CUFFCOMPARE, which can classify transcripts as known or novel transcripts. Finally, the all.combined.gtf file from CUFFCOMPARE and mapping files from TOPHAT were used by CUFFDIFF for differential analysis. Both CUFFCOMPARE and CUFFDIFF were run with default parameters. SNVs and small insertions and deletions (indels) were discovered using SAMTOOLS 0.1.19 (Li et al. 2009) with minimum and maximum depths of four and 100 respectively. SNV filtering was performed with BEDTOOLS 2.17.0. TOPHAT was used to analyze the alternative splicing as previously described (Trapnell et al. 2009). A heatmap showing the top 50 differentially expressed genes (DEGs) was generated based on log10-transformed RPKM (reads per kilobase per million) values using MEV4.8.1.
© 2015 Stichting International Foundation for Animal Genetics, 46, 299–307
RNA-seq of duck abdominal fats
GO enrichment and pathway analysis Differentially expressed genes were mapped to the Gene Ontology (GO) terms that were downloaded from the GO database (http://www.geneontology.org/), and hypergeometric test P-values were calculated to identify the GO terms that were overrepresented in the DEGs. GO terms with P < 0.05 were considered significantly enriched. DEGs together with genes containing non-synonymous SNVs were also included in the pathway enrichment analysis using KEGG Mapper.
two breeds exhibited different abdominal fat percentages, that is, 1.20% in Peking ducks and 0.68% in Anas platyrhynchos. Then, histological analysis of these abdominal fat tissues showed that the size of adipocytes from Peking ducks was larger than that of adipocytes from Anas platyrhynchos (Fig. 1b); consistent with this finding was the decrease in cell number (Fig. 1c). These results suggest that Peking ducks accumulate more abdominal fat than do their wild counterparts.
Global view of transcriptome sequencing Real-time PCR analysis Real-time PCR analysis was performed using the 7900 Sequence Detection System (Applied Biosystems) under the following conditions: 95 °C for 10 min, followed by 40 cycles of 95 °C for 30 s and 60 °C for 1 min. The primer sequences for real-time PCR analysis were designed using PRIMER EXPRESS (v3.0) (Applied Biosystems) and are listed in Table S1. Five birds from each breed were used in this experiment, and the relative expression of each gene was normalized to the internal control gene EEF1A1 (ENSAPLG00000009349) using the 2DDCt method (Livaka & Schmittgen 2001).
Results Differences in abdominal fat between Peking ducks and Anas platyrhynchos To identify differences in abdominal fat between Peking ducks and Anas platyrhynchos, abdominal fat percentages were calculated in the two breeds. As shown in Fig. 1a, the
To globally understand the transcriptome changes underlying abdominal fat variation between Peking ducks and Anas platyrhynchos, transcriptome sequencing of abdominal fats from the two breeds was performed using the Illumina HiSeq 2000 platform with three biological replicates per breed. After removing sequencing adaptors and low-quality reads, approximately 350 million clean reads were obtained. These clean reads were then used for mapping to the duck reference genome. According to the mapping results, approximately 64% of the clean reads were successfully mapped to the duck genome. Of the mapped reads, 98% were uniquely aligned (Table 1). All of the mapped reads were further assembled into transcripts using CUFFLINKS, generating 61 250 transcripts, including 23 699 novel ones. Reads per kilobase per million of mapped reads is the normalized quantitative method for gene expression in RNA-seq (Mortazavi et al. 2008). To overview the gene expression in each sample, RPKM was calculated for each gene. Based on the RPKM value, 18 000 to 18 866 genes
(c) 1.4
**
160
1.2
Number of adipocyte
Abdominal fat percentage (%)
(a)
1 0.8 0.6 0.4 0.2
140 120 100 80 60
*
40 20 0
0 PD
PD
AP
AP
(b) Figure 1 Differences in abdominal fat between Peking duck (PD) and Anas platyrhynchos (AP). (a) Difference in abdominal fat percentage between the two breeds. Data are presented as mean SD (n = 5). (b) Representative sections of hematoxylin-and-eosinstained abdominal fats from PD and AP at 4009 magnification. (c) Difference in adipocyte number between the two breeds. Data are presented as mean SD (n = 3). *P < 0.05; **P < 0.01. © 2015 Stichting International Foundation for Animal Genetics, 46, 299–307
PD
AP
301
Chen et al. Table 1 Overview of RNA sequencing, mapping and assembly.
Sample
Clean reads
Reads mapped on the genome (%)
Reads uniquely mapped on the genome (%)
AP3 AP4 AP5 PD3 PD4 PD5
61 453 324 37 003 864 49 083 112 109 355 142 44 991 866 48 853 976
66.49 62.49 65.18 62.60 66.29 63.18
98.00 97.92 98.06 97.68 98.11 97.92
Transcripts
Novel transcripts
61 250
23 699
Note: PD3, PD4 and PD5 represent three biological replicates of Peking ducks; AP3, AP4 and AP5 represent three biological replicates of their wild counterparts (Anas platyrhynchos). 10 000
RPKM
