Limits...
The maternal and early embryonic transcriptome of the milkweed bug Oncopeltus fasciatus.

Ewen-Campen B, Shaner N, Panfilio KA, Suzuki Y, Roth S, Extavour CG - BMC Genomics (2011)

Bottom Line: We identified 10,775 unique genes, including members of all major conserved metazoan signaling pathways and genes involved in several major categories of early developmental processes.We also specifically address the effects of cDNA normalization on gene discovery in de novo transcriptome analyses.Our sequencing, assembly and annotation framework provide a simple and effective way to achieve high-throughput gene discovery for organisms lacking a sequenced genome.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Organismic and Evolutionary Biology, Harvard University, 16 Divinity Avenue, Cambridge, MA 02138, USA.

ABSTRACT

Background: Most evolutionary developmental biology ("evo-devo") studies of emerging model organisms focus on small numbers of candidate genes cloned individually using degenerate PCR. However, newly available sequencing technologies such as 454 pyrosequencing have recently begun to allow for massive gene discovery in animals without sequenced genomes. Within insects, although large volumes of sequence data are available for holometabolous insects, developmental studies of basally branching hemimetabolous insects typically suffer from low rates of gene discovery.

Results: We used 454 pyrosequencing to sequence over 500 million bases of cDNA from the ovaries and embryos of the milkweed bug Oncopeltus fasciatus, which lacks a sequenced genome. This indirectly developing insect occupies an important phylogenetic position, branching basal to Diptera (including fruit flies) and Hymenoptera (including honeybees), and is an experimentally tractable model for short-germ development. 2,087,410 reads from both normalized and non-normalized cDNA assembled into 21,097 sequences (isotigs) and 112,531 singletons. The assembled sequences fell into 16,617 unique gene models, and included predictions of splicing isoforms, which we examined experimentally. Discovery of new genes plateaued after assembly of ~1.5 million reads, suggesting that we have sequenced nearly all transcripts present in the cDNA sampled. Many transcripts have been assembled at close to full length, and there is a net gain of sequence data for over half of the pre-existing O. fasciatus accessions for developmental genes in GenBank. We identified 10,775 unique genes, including members of all major conserved metazoan signaling pathways and genes involved in several major categories of early developmental processes. We also specifically address the effects of cDNA normalization on gene discovery in de novo transcriptome analyses.

Conclusions: Our sequencing, assembly and annotation framework provide a simple and effective way to achieve high-throughput gene discovery for organisms lacking a sequenced genome. These data will have applications to the study of the evolution of arthropod genes and genetic pathways, and to the wider evolution, development and genomics communities working with emerging model organisms.[The sequence data from this study have been submitted to GenBank under study accession number SRP002610 (http://www.ncbi.nlm.nih.gov/sra?term=SRP002610). Custom scripts generated are available at http://www.extavourlab.com/protocols/index.html. Seven Additional files are available.].

Show MeSH

Related in: MedlinePlus

Normalization decreases coverage of highly abundant genes, but does not change the GO term distribution of contigs. In both samples, most contigs are composed of <102 reads. However, the non-normalized sample (A) contains contigs with many more reads per contigs than the normalized sample (B). In other words, normalization preferentially decreases the number of reads of those contigs with the most reads. (C) GO term distributions do not differ dramatically between pyrosequenced libraries of N versus NN cDNA. However, see Additional file 6 for exceptions. Column heights reflect the percentage of annotated sequences in each assembly that mapped to a given GO term. Note that the GO terms shown represent the results of mapping the N and NN reads against the complete assembly, rather than those obtained via independent assemblies of N and NN reads.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3040728&req=5

Figure 8: Normalization decreases coverage of highly abundant genes, but does not change the GO term distribution of contigs. In both samples, most contigs are composed of <102 reads. However, the non-normalized sample (A) contains contigs with many more reads per contigs than the normalized sample (B). In other words, normalization preferentially decreases the number of reads of those contigs with the most reads. (C) GO term distributions do not differ dramatically between pyrosequenced libraries of N versus NN cDNA. However, see Additional file 6 for exceptions. Column heights reflect the percentage of annotated sequences in each assembly that mapped to a given GO term. Note that the GO terms shown represent the results of mapping the N and NN reads against the complete assembly, rather than those obtained via independent assemblies of N and NN reads.

Mentions: First, to test whether our normalization protocol successfully reduced the presence of highly abundant transcripts, we created separate assemblies from the N and NN cDNA samples (equalizing the total number of bases to reduce the contribution of additional sequence found in the NN sample). The N assembly contained a greater number of isotigs that were shorter on average than those in the NN assembly (Figure 2B). Additionally, more singletons were generated in the N assembly relative to the NN assembly (Table 2). Further, similar to the results obtained by Bellin and colleagues [27], we observed the predicted decrease in the maximum number of reads per contig in the N assembly compared to the NN assembly (Figure 8A, B), demonstrating that the normalization procedure successfully reduced the sequencing of highly abundant transcripts. These statistics, which could be interpreted to suggest that the N reads generated an inferior assembly, may result from the shorter average length of reads in the N sample (Figure 2A). Indeed, Newbler rejected 7.9% (30,780) of the N reads as too short, compared to only 1% (3,935) of the NN reads. However, these assembly statistics could also indicate greater heterogeneity in the N sample, which would suggest that normalization might increase the number of new genes identified.


The maternal and early embryonic transcriptome of the milkweed bug Oncopeltus fasciatus.

Ewen-Campen B, Shaner N, Panfilio KA, Suzuki Y, Roth S, Extavour CG - BMC Genomics (2011)

Normalization decreases coverage of highly abundant genes, but does not change the GO term distribution of contigs. In both samples, most contigs are composed of <102 reads. However, the non-normalized sample (A) contains contigs with many more reads per contigs than the normalized sample (B). In other words, normalization preferentially decreases the number of reads of those contigs with the most reads. (C) GO term distributions do not differ dramatically between pyrosequenced libraries of N versus NN cDNA. However, see Additional file 6 for exceptions. Column heights reflect the percentage of annotated sequences in each assembly that mapped to a given GO term. Note that the GO terms shown represent the results of mapping the N and NN reads against the complete assembly, rather than those obtained via independent assemblies of N and NN reads.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3040728&req=5

Figure 8: Normalization decreases coverage of highly abundant genes, but does not change the GO term distribution of contigs. In both samples, most contigs are composed of <102 reads. However, the non-normalized sample (A) contains contigs with many more reads per contigs than the normalized sample (B). In other words, normalization preferentially decreases the number of reads of those contigs with the most reads. (C) GO term distributions do not differ dramatically between pyrosequenced libraries of N versus NN cDNA. However, see Additional file 6 for exceptions. Column heights reflect the percentage of annotated sequences in each assembly that mapped to a given GO term. Note that the GO terms shown represent the results of mapping the N and NN reads against the complete assembly, rather than those obtained via independent assemblies of N and NN reads.
Mentions: First, to test whether our normalization protocol successfully reduced the presence of highly abundant transcripts, we created separate assemblies from the N and NN cDNA samples (equalizing the total number of bases to reduce the contribution of additional sequence found in the NN sample). The N assembly contained a greater number of isotigs that were shorter on average than those in the NN assembly (Figure 2B). Additionally, more singletons were generated in the N assembly relative to the NN assembly (Table 2). Further, similar to the results obtained by Bellin and colleagues [27], we observed the predicted decrease in the maximum number of reads per contig in the N assembly compared to the NN assembly (Figure 8A, B), demonstrating that the normalization procedure successfully reduced the sequencing of highly abundant transcripts. These statistics, which could be interpreted to suggest that the N reads generated an inferior assembly, may result from the shorter average length of reads in the N sample (Figure 2A). Indeed, Newbler rejected 7.9% (30,780) of the N reads as too short, compared to only 1% (3,935) of the NN reads. However, these assembly statistics could also indicate greater heterogeneity in the N sample, which would suggest that normalization might increase the number of new genes identified.

Bottom Line: We identified 10,775 unique genes, including members of all major conserved metazoan signaling pathways and genes involved in several major categories of early developmental processes.We also specifically address the effects of cDNA normalization on gene discovery in de novo transcriptome analyses.Our sequencing, assembly and annotation framework provide a simple and effective way to achieve high-throughput gene discovery for organisms lacking a sequenced genome.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Organismic and Evolutionary Biology, Harvard University, 16 Divinity Avenue, Cambridge, MA 02138, USA.

ABSTRACT

Background: Most evolutionary developmental biology ("evo-devo") studies of emerging model organisms focus on small numbers of candidate genes cloned individually using degenerate PCR. However, newly available sequencing technologies such as 454 pyrosequencing have recently begun to allow for massive gene discovery in animals without sequenced genomes. Within insects, although large volumes of sequence data are available for holometabolous insects, developmental studies of basally branching hemimetabolous insects typically suffer from low rates of gene discovery.

Results: We used 454 pyrosequencing to sequence over 500 million bases of cDNA from the ovaries and embryos of the milkweed bug Oncopeltus fasciatus, which lacks a sequenced genome. This indirectly developing insect occupies an important phylogenetic position, branching basal to Diptera (including fruit flies) and Hymenoptera (including honeybees), and is an experimentally tractable model for short-germ development. 2,087,410 reads from both normalized and non-normalized cDNA assembled into 21,097 sequences (isotigs) and 112,531 singletons. The assembled sequences fell into 16,617 unique gene models, and included predictions of splicing isoforms, which we examined experimentally. Discovery of new genes plateaued after assembly of ~1.5 million reads, suggesting that we have sequenced nearly all transcripts present in the cDNA sampled. Many transcripts have been assembled at close to full length, and there is a net gain of sequence data for over half of the pre-existing O. fasciatus accessions for developmental genes in GenBank. We identified 10,775 unique genes, including members of all major conserved metazoan signaling pathways and genes involved in several major categories of early developmental processes. We also specifically address the effects of cDNA normalization on gene discovery in de novo transcriptome analyses.

Conclusions: Our sequencing, assembly and annotation framework provide a simple and effective way to achieve high-throughput gene discovery for organisms lacking a sequenced genome. These data will have applications to the study of the evolution of arthropod genes and genetic pathways, and to the wider evolution, development and genomics communities working with emerging model organisms.[The sequence data from this study have been submitted to GenBank under study accession number SRP002610 (http://www.ncbi.nlm.nih.gov/sra?term=SRP002610). Custom scripts generated are available at http://www.extavourlab.com/protocols/index.html. Seven Additional files are available.].

Show MeSH
Related in: MedlinePlus