Limits...
SC3-seq: a method for highly parallel and quantitative measurement of single-cell gene expression.

Nakamura T, Yabuta Y, Okamoto I, Aramaki S, Yokobayashi S, Kurimoto K, Sekiguchi K, Nakagawa M, Yamamoto T, Saitou M - Nucleic Acids Res. (2015)

Bottom Line: We here present single-cell mRNA 3-prime end sequencing (SC3-seq), a practical methodology based on PCR amplification followed by 3-prime-end enrichment for highly quantitative, parallel and cost-effective measurement of gene expression in single cells.Furthermore, the SC3-seq reveals the heterogeneity in human-induced pluripotent stem cells (hiPSCs) cultured under on-feeder as well as feeder-free conditions, demonstrating a more homogeneous property of the feeder-free hiPSCs.We propose that SC3-seq might be used as a powerful strategy for single-cell transcriptome analysis in a broad range of investigations in biomedical sciences.

View Article: PubMed Central - PubMed

Affiliation: Department of Anatomy and Cell Biology, Graduate School of Medicine, Kyoto University, Yoshida-Konoe-cho, Sakyo-ku, Kyoto 606-8501, Japan JST, ERATO, Yoshida-Konoe-cho, Sakyo-ku, Kyoto 606-8501, Japan.

Show MeSH

Related in: MedlinePlus

Coverage and accuracy of the SC3-seq. (A) Coverage of the SC3-seq from 10 pg of total RNAs as a function of the expression level [log2 (RPM+1)] in 100 ng of total RNAs. The black squares represent the means of coverage in single-sample analysis, with bars representing SDs. The results of multiple-sample analyses under the definitions of detection where transcripts are detected in ≧1–8 of the eight amplified samples are represented by squares with the indicated color codes. Copy numbers per 10 pg of total RNAs estimated by the SC3-seq reads of the ERCC spike-in RNAs in 100 ng of RNAs are indicated by dashed lines (red: 1000 copies; orange: 100 copies; pale blue: 10 copies; blue: 1 copy). (B) Accuracy of the SC3-seq from 10 pg of total RNAs as a function of the expression level [log2 (RPM+1)]. The representation code is the same as in (A). (C) The plot of the number of genes detected [log2 (RPM+1) ≧ 4, fold changes of gene expression levels ≦2 in comparison to those determined by the full sequence reads (Supplementary Table S1)] by the SC3-seq from 100 ng, 10 ng, 1 ng, 100 pg and 10 pg of mESC total RNAs (color code indicated) as a function of the sequence reads. (D) The plot of the percentage of the detection [fold changes of gene expression levels ≦2 in comparison to those determined by the full sequence reads (Supplementary Table S1)] by the SC3-seq from 10 pg of total RNAs as a function of the sequence reads, categorized by expression level ranges in 100 ng of total RNAs (color code indicated).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4482058&req=5

Figure 3: Coverage and accuracy of the SC3-seq. (A) Coverage of the SC3-seq from 10 pg of total RNAs as a function of the expression level [log2 (RPM+1)] in 100 ng of total RNAs. The black squares represent the means of coverage in single-sample analysis, with bars representing SDs. The results of multiple-sample analyses under the definitions of detection where transcripts are detected in ≧1–8 of the eight amplified samples are represented by squares with the indicated color codes. Copy numbers per 10 pg of total RNAs estimated by the SC3-seq reads of the ERCC spike-in RNAs in 100 ng of RNAs are indicated by dashed lines (red: 1000 copies; orange: 100 copies; pale blue: 10 copies; blue: 1 copy). (B) Accuracy of the SC3-seq from 10 pg of total RNAs as a function of the expression level [log2 (RPM+1)]. The representation code is the same as in (A). (C) The plot of the number of genes detected [log2 (RPM+1) ≧ 4, fold changes of gene expression levels ≦2 in comparison to those determined by the full sequence reads (Supplementary Table S1)] by the SC3-seq from 100 ng, 10 ng, 1 ng, 100 pg and 10 pg of mESC total RNAs (color code indicated) as a function of the sequence reads. (D) The plot of the percentage of the detection [fold changes of gene expression levels ≦2 in comparison to those determined by the full sequence reads (Supplementary Table S1)] by the SC3-seq from 10 pg of total RNAs as a function of the sequence reads, categorized by expression level ranges in 100 ng of total RNAs (color code indicated).

Mentions: To examine the performance of the SC3-seq further, we next evaluated the coverage [the number of genes detected in 10-pg RNAs (log2 (RPM+1) ≧1)/the number of genes detected in 100-ng RNAs (log2 (RPM+1) ≧1)] and accuracy [the number of genes detected in 10-pg RNAs (log2 (RPM+1) ≧1) that are detected in 100-ng RNAs (log2 (RPM+1) ≧1)] of the SC3-seq from 10-pg RNAs by using the data from eight replicates. The truly expressed genes were defined as those that were detected [log2 (RPM+1) ≧1] in both samples prepared by SC3-seq from 100 ng of RNAs. Coverage of the single amplified samples as a function of the expression level was plotted (black squares in Figure 3A). As expected from the previous reports (7) and the data shown above (Figure 2), coverage was dependent on the expression level, but a vast majority of the truly expressed genes (cumulative percentage, 94.1%) that are expressed more than 10 copies per 10 pg were successfully detected (Figure 3A). The accuracy of the single amplified samples was plotted similarly, and we found that 99.7% (cumulative percentage) of the genes detected were truly expressed in the expression level range of more than 10 copies per 10-pg RNAs (Figure 3B). When we performed multiple sample analyses (eight samples were analyzed), coverage was improved under the definitions of detection where ≧1 to ≧5 of the eight amplified samples exhibited reads (≧10 copies per 10-pg RNAs), whereas accuracy was essentially nearly 100% under all detection definitions (Figure 3A and B). These findings indicate that a single sample prepared by SC3-seq from single-cell level RNAs exhibits excellent coverage and accuracy, and multiple sample analyses further improve both the coverage and accuracy.


SC3-seq: a method for highly parallel and quantitative measurement of single-cell gene expression.

Nakamura T, Yabuta Y, Okamoto I, Aramaki S, Yokobayashi S, Kurimoto K, Sekiguchi K, Nakagawa M, Yamamoto T, Saitou M - Nucleic Acids Res. (2015)

Coverage and accuracy of the SC3-seq. (A) Coverage of the SC3-seq from 10 pg of total RNAs as a function of the expression level [log2 (RPM+1)] in 100 ng of total RNAs. The black squares represent the means of coverage in single-sample analysis, with bars representing SDs. The results of multiple-sample analyses under the definitions of detection where transcripts are detected in ≧1–8 of the eight amplified samples are represented by squares with the indicated color codes. Copy numbers per 10 pg of total RNAs estimated by the SC3-seq reads of the ERCC spike-in RNAs in 100 ng of RNAs are indicated by dashed lines (red: 1000 copies; orange: 100 copies; pale blue: 10 copies; blue: 1 copy). (B) Accuracy of the SC3-seq from 10 pg of total RNAs as a function of the expression level [log2 (RPM+1)]. The representation code is the same as in (A). (C) The plot of the number of genes detected [log2 (RPM+1) ≧ 4, fold changes of gene expression levels ≦2 in comparison to those determined by the full sequence reads (Supplementary Table S1)] by the SC3-seq from 100 ng, 10 ng, 1 ng, 100 pg and 10 pg of mESC total RNAs (color code indicated) as a function of the sequence reads. (D) The plot of the percentage of the detection [fold changes of gene expression levels ≦2 in comparison to those determined by the full sequence reads (Supplementary Table S1)] by the SC3-seq from 10 pg of total RNAs as a function of the sequence reads, categorized by expression level ranges in 100 ng of total RNAs (color code indicated).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4482058&req=5

Figure 3: Coverage and accuracy of the SC3-seq. (A) Coverage of the SC3-seq from 10 pg of total RNAs as a function of the expression level [log2 (RPM+1)] in 100 ng of total RNAs. The black squares represent the means of coverage in single-sample analysis, with bars representing SDs. The results of multiple-sample analyses under the definitions of detection where transcripts are detected in ≧1–8 of the eight amplified samples are represented by squares with the indicated color codes. Copy numbers per 10 pg of total RNAs estimated by the SC3-seq reads of the ERCC spike-in RNAs in 100 ng of RNAs are indicated by dashed lines (red: 1000 copies; orange: 100 copies; pale blue: 10 copies; blue: 1 copy). (B) Accuracy of the SC3-seq from 10 pg of total RNAs as a function of the expression level [log2 (RPM+1)]. The representation code is the same as in (A). (C) The plot of the number of genes detected [log2 (RPM+1) ≧ 4, fold changes of gene expression levels ≦2 in comparison to those determined by the full sequence reads (Supplementary Table S1)] by the SC3-seq from 100 ng, 10 ng, 1 ng, 100 pg and 10 pg of mESC total RNAs (color code indicated) as a function of the sequence reads. (D) The plot of the percentage of the detection [fold changes of gene expression levels ≦2 in comparison to those determined by the full sequence reads (Supplementary Table S1)] by the SC3-seq from 10 pg of total RNAs as a function of the sequence reads, categorized by expression level ranges in 100 ng of total RNAs (color code indicated).
Mentions: To examine the performance of the SC3-seq further, we next evaluated the coverage [the number of genes detected in 10-pg RNAs (log2 (RPM+1) ≧1)/the number of genes detected in 100-ng RNAs (log2 (RPM+1) ≧1)] and accuracy [the number of genes detected in 10-pg RNAs (log2 (RPM+1) ≧1) that are detected in 100-ng RNAs (log2 (RPM+1) ≧1)] of the SC3-seq from 10-pg RNAs by using the data from eight replicates. The truly expressed genes were defined as those that were detected [log2 (RPM+1) ≧1] in both samples prepared by SC3-seq from 100 ng of RNAs. Coverage of the single amplified samples as a function of the expression level was plotted (black squares in Figure 3A). As expected from the previous reports (7) and the data shown above (Figure 2), coverage was dependent on the expression level, but a vast majority of the truly expressed genes (cumulative percentage, 94.1%) that are expressed more than 10 copies per 10 pg were successfully detected (Figure 3A). The accuracy of the single amplified samples was plotted similarly, and we found that 99.7% (cumulative percentage) of the genes detected were truly expressed in the expression level range of more than 10 copies per 10-pg RNAs (Figure 3B). When we performed multiple sample analyses (eight samples were analyzed), coverage was improved under the definitions of detection where ≧1 to ≧5 of the eight amplified samples exhibited reads (≧10 copies per 10-pg RNAs), whereas accuracy was essentially nearly 100% under all detection definitions (Figure 3A and B). These findings indicate that a single sample prepared by SC3-seq from single-cell level RNAs exhibits excellent coverage and accuracy, and multiple sample analyses further improve both the coverage and accuracy.

Bottom Line: We here present single-cell mRNA 3-prime end sequencing (SC3-seq), a practical methodology based on PCR amplification followed by 3-prime-end enrichment for highly quantitative, parallel and cost-effective measurement of gene expression in single cells.Furthermore, the SC3-seq reveals the heterogeneity in human-induced pluripotent stem cells (hiPSCs) cultured under on-feeder as well as feeder-free conditions, demonstrating a more homogeneous property of the feeder-free hiPSCs.We propose that SC3-seq might be used as a powerful strategy for single-cell transcriptome analysis in a broad range of investigations in biomedical sciences.

View Article: PubMed Central - PubMed

Affiliation: Department of Anatomy and Cell Biology, Graduate School of Medicine, Kyoto University, Yoshida-Konoe-cho, Sakyo-ku, Kyoto 606-8501, Japan JST, ERATO, Yoshida-Konoe-cho, Sakyo-ku, Kyoto 606-8501, Japan.

Show MeSH
Related in: MedlinePlus