Real-time transcriptomic profiling in distinct experimental conditions
- Institute of Pharmaceutical and Biomedical Sciences, Johannes Gutenberg-University Mainz, Germany
- Institute of Human Genetics, University Medical Center of the Johannes Gutenberg University Mainz, Germany
Figures
Figure 1
Streamlined pipeline for the rapid experimental setup and utilization of NanopoReaTA.
Sample preparation involved Trizol RNA isolation (~1 hr) following library preparation, which included the synthesis of dscDNA library for the desired RNA sample (~2 hr). Samples were barcoded, and adapter ligated (~1 hr). Sequencing was performed using a PromethION R10 flow cell or MinION R9 flow cell (for HEK293 and HeLa), and real-time data analysis and visualization occurred alongside ongoing sequencing. For this study, figures were exported at 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr psi, providing insights into the dynamic transcriptional changes of long-read RNA-seq between distinct conditions.
Figure 2 with 10 supplements
Real-time transcriptomic analysis between HEK293 and HeLa using NanopoReaTA.
(A) Experimental strategy. RNA was isolated from HEK293 and HeLa cells, and the dscDNA library was prepared, which included sample barcoding and adapter ligation. Samples were loaded and sequenced using a PromethION R10 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data was collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–D) Differential gene expression 1 hr post-sequencing initiation between HEK293 and HeLa samples. Selected data plots showing sample-to-sample distance plot (B), Principal component analysis (PCA) (C), and volcano plot (D) 1 hr post-sequencing initiation. (E–G) Differential gene expression 24 hr post-sequencing initiation between HEK293 and HeLa samples. Selected data plots showing sample-to-sample distance plot (E), PCA (F), and volcano plot (G) 24 hr post-sequencing initiation. (H) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points. (I) Validation of identified differentially expressed genes through Harmonizome database (Rouillard et al., 2016). For this analysis, we utilized ‘HPA Cell Line Gene Expression Profiles’ (Uhlén et al., 2015). Among the DEGs (identified at the 24 hr time point), 19 genes were found to be enriched in HEK293 and depleted in HeLa, while 46 genes showed enrichment in HeLa and depletion in HEK293. One gene (PFN1) was enriched in HeLa and enriched in HEK293 Harmonizome database.
-
Figure 2—source data 1
General sequencing overview of HEK293 and HeLa (2 and 10 replicates per condition).
- https://cdn.elifesciences.org/articles/98768/elife-98768-fig2-data1-v1.xlsx
Figure 2—figure supplement 1
Experimental strategy of HEK293 versus HeLa and sequencing overview.
(A) RNA was isolated from HEK293 and HeLa cells dscDNA library was prepared which included samples barcoding and adapter ligation. Samples were loaded and sequenced using a PromethION R10 flow cell or MinION R9 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data was collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–E) The box plot represents total read generated (B), total mapped reads (C), total gene counts (D), and total transcript count (E) per condition. The data for 2 replicates and 10 replicates per condition are shown in left and right, respectively.
Figure 2—figure supplement 2
PromethION (10 rep) - General overview in HEK293 vs HeLa.
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plot shows the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 2—figure supplement 3
PromethION (2 rep) - general overview in HEK293 vs HeLa.
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 2—figure supplement 4
MinION - general overview HEK293 vs HeLa.
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes.
Figure 2—figure supplement 5
PromethION (10 rep) - gene expression analysis in HEK293 vs HeLa.
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization.
Figure 2—figure supplement 6
PromethION - gene expression analysis in HEK293 vs HeLa.
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization.
Figure 2—figure supplement 7
MinION – gene and transcript expression in HEK293 vs HeLa.
(A-D) DGE analysis. (A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples. (B) The Euclidean distance between the gene expression patterns of all samples to each other is plotted using heatmap. Normalized gene counts were used for distance computation. (C) The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (D) The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. (E–H) DTE analysis. (E) PCA of the top 500 transcripts with the highest variance across all samples. (F) The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using heatmap. Normalized transcript counts were used for distance computation. (G) The differential transcript expression analysis is shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per transcript observed from the Wald-Test integrated in DESeq2 (H) The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per gene.
Figure 2—figure supplement 8
Transcript expression in HEK293 vs HeLa (10 rep).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components, PC1 and PC2 - that explain the most variance in the dataset are plotted against each other. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts. (E) Differential transcript usage volcano plot. The log2FoldChange values of the transcripts from DEXSeq (Anders et al., 2012) are plotted against the –log10 adjusted p-value.
Figure 2—figure supplement 9
Transcript expression in HEK293 vs HeLa.
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components, PC1 and PC2 - that explain the most variance in the dataset are plotted against each other. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 2—figure supplement 10
Comparative gene expression analysis between HEK293 and HeLa cells across PromethION and MinION sequencing platforms.
(A–B) Venn diagrams showing genes upregulated in HEK293 compared to HeLa (A) and in HeLa compared to HEK293 (B) across PromethION (10-replicate and 2-replicate setups) and MinION (2-replicate setup). Overlaps indicate genes consistently detected across platforms, highlighting shared and unique DEGs. (C) Heatmaps showing the relative expression levels of selected DEGs enriched in HEK293 or HeLa according to the Harmonizome database (Rouillard et al., 2016). The Harmonizome database contains various datasets, including the ‘HPA Cell Line Gene Expression Profiles’ which evaluates differential gene expression across distinct cell lines (Uhlén et al., 2015). We examined whether the DEGs aligned with the observations in the database. (D) Venn diagram comparing DEGs identified in HEK293 and HeLa against the Harmonizome database. Overlapping regions validate the observed transcriptional profiles, confirming consistency with known cell line gene expression patterns.
Figure 3 with 10 supplements
Real-time transcriptomic analysis between rRNA-depleted and rRNA-enriched transcripts using NanopoReaTA.
(A) Experimental strategy. RNA was isolated from HEK293 and selective purification of distinct RNA populations was performed using the Ribominus Eukaryote kit. The dscDNA library was prepared, which included sample barcoding and adapter ligation. Samples were loaded and sequenced using a PromethION R10 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data was collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–H) Differential gene expression between RiboM and TotalR from HEK293. Selected data plots showing PCA and volcano plots 1 hr (B–C) and 24 hr (D–E) post sequencing initiation. (F) Normalized gene counts for selected genes 24 hr. Normalized gene counts are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 was used for normalization. (G) Number of identified genes with >0 aligned reads after each iteration per condition. (H) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points. (I–O) Differential gene expression between RiboP and TotalR. Similar analyses to B-H were conducted for comparisons at 1 hr (I–J) and 24 hr (K–L) as well as normalized gene counts (M), gene expression variability (N) and Venn diagram (O). (P–U) Differential gene expression compared between RiboM and RiboP. Similar analyses to B-H were conducted for comparisons at 1 hr (P–Q) and 24 hr (R–S) as well as normalized gene counts (T), gene expression variability (V), and Venn diagram (U).
-
Figure 3—source data 1
General sequencing overview of HEK293 Total RNA, RiboPlus and Ribominus samples (2 replicates per condition).
- https://cdn.elifesciences.org/articles/98768/elife-98768-fig3-data1-v1.xlsx
Figure 3—figure supplement 1
Selective purification of ribosomal-depleted (RiboMinus) and ribosomal-enriched (RiboPlus) transcripts and sequencing overview.
(A) Selective purification of distinct RNA populations was performed using the Ribominus Eukaryote kit for RNA-seq (#Ambion, A10837-08) according to the manufacturer’s ‘standard protocol’ instructions. Total RNA (containing rRNA and other RNA transcripts) is incubated with 5’ biotin label probes complementary to rRNA RNA transcripts (28 S, 18 S, 5.8 S, and 5 S). Probes hybridize to rRNA and are separated using magnetic bead separation. Non-rRNA is purified using ethanol precipitation, whereas bead-bound rRNA is purified using Trizol RNA isolation. (B) 1 µg of Total RNA (HEK293) and RiboPlus and 150 ng of Ribominus were loaded on 1% TBE agarose gel stained with ethidium bromide (in duplicates). Both total RNA and RiboPlus (rRNA-enriched) contained bands corresponding to 28 S, 18 S, and 5 S rRNA, whereas RiboMinus contained a band corresponding to lower fragment length. (C–F) The box plot represents total read generated (C), total mapped reads (D), total gene counts (E), and total transcript count (F) per condition (from two replicates), per time point.
Figure 3—figure supplement 2
General overview in RiboMinus vs total RNA.
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 3—figure supplement 3
Gene expression analysis in RiboMinus vs total RNA.
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two Principal Components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 3—figure supplement 4
Transcript expression in RiboMinus vs total RNA.
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components, PC1 and PC2 - that explain the most variance in the dataset are plotted against each other. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 3—figure supplement 5
General overview in RiboPlus vs total RNA.
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 3—figure supplement 6
Gene expression analysis in RiboPlus vs total RNA.
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 3—figure supplement 7
Transcript expression in RiboPlus vs total RNA.
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 3—figure supplement 8
General overview in RiboPlus vs RiboMinus.
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 3—figure supplement 9
Gene expression analysis in RiboPlus vs RiboMinus.
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 3—figure supplement 10
Transcript expression in RiboPlus vs RiboMinus.
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 4 with 4 supplements
Real-time transcriptomic analysis between heat-shock and non-heat shock treated cells using NanopoReaTA.
(A) Experimental strategy. RNA was isolated from HEK293 treated with heat shock on non-heat shock treatment (n=6), and selective purification of distinct RNA populations was performed using the Ribominus Eukaryote kit. The dscDNA library was prepared, which included sample barcoding and adapter ligation. Samples were loaded and sequenced using a PromethION R10 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data was collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–D) Differential gene expression 1 hr post-sequencing initiation between HEK293 and HeLa samples. Selected data plots showing PCA (B), volcano plot (C), and normalized counts for selected genes (D) 1 hr post-sequencing initiation. (E–G) Differential gene expression 24 hr post-sequencing initiation between HEK293 and HeLa samples. Selected data plots showing PCA (E), volcano plot (F), and normalized counts for selected genes (G) 24 hr post-sequencing initiation. (H) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points. (I) Gene Ontology (GO) enrichment analysis of upregulated gene in HS compared to NHs conditions. (J) Volcano plot depicting differential transcript usage between HS and NHS conditions.
-
Figure 4—source data 1
General sequencing overview of heat-shock versus non-heat shock samples (6 replicates per condition).
- https://cdn.elifesciences.org/articles/98768/elife-98768-fig4-data1-v1.xlsx
Figure 4—figure supplement 1
Experimental strategy of heat-shock experiments and sequencing overview.
(A) RNA was isolated from HEK293 and HeLa cells (n=6) dscDNA library was prepared which included samples barcoding and adapter ligation. Samples were loaded and sequenced using a PromethION R10 flow cell or MinION R9 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data was collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–E) The box plot represents total read generated (B), total mapped reads (C), total gene counts (D), and total transcript count (E) per condition (from two replicates), per time point.
Figure 4—figure supplement 2
General overview in HS vs NHS.
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 4—figure supplement 3
Gene expression analysis in HS vs NHS.
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 4—figure supplement 4
Transcript expression in HS vs NHS.
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 5 with 17 supplements
Real-time transcriptomic analysis between new1Δ-pEV(HIS3) vs WT-pEV(HIS3) using NanopoReaTA.
(A) Experimental strategy. Following library preparation (n=3), samples were loaded and sequenced using a PromethION R10 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data were collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–C) PCA plot showing the sample separation between new1Δ-pEV(HIS3) vs WT-pEV(HIS3) 1 hr post-sequencing (B) and 24 hr post-sequencing (C). (D–E) Volcano plots showing the differentially expressed genes between new1Δ-pEV(HIS3) vs WT-pEV(HIS3) 1 hr post-sequencing initiation (D) and 24 hr post-sequencing initiation (E). (F) Normalized gene count for selected genes 24 hr post-sequencing initiation. Normalized gene counts are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 was used for normalization. (G) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
-
Figure 5—source data 1
General sequencing overview of Yeast setup 1.
- https://cdn.elifesciences.org/articles/98768/elife-98768-fig5-data1-v1.xlsx
Figure 5—figure supplement 1
Sequencing overview yeast setup 1.
(A) For Yeast setup 1, WT strain (BY4741, MATa, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0) and new1Δ::KanMX, where the NEW1 gene was replaced with the KanMX cassette which contains the Kanamycin resistance gene (KanR). These strains were transformed with either an empty vector with the HIS3 selection marker (pEV(HIS3)) or an overexpression vector for C-terminally FLAG-tagged New1 with the HIS3 selection marker (pNew1(HIS3)). (B–E) The box plots represent total read generated (B), total mapped reads (C), total gene counts (D) and total transcript count (E) per condition, per time point (three biological replicates per condition). Abbreviations: WTpEV - WTpEV(HIS3), WTpN - WTpNew1(HIS3), dNpEV - new1ΔpEV(HIS3), dNpN - new1ΔpNew1(HIS3).
Figure 5—figure supplement 2
Real-time transcriptomic analysis in yeast setup 1 samples using NanopoReaTA.
(A) Experimental strategy of yeast setup 1. RNA was isolated from selected yeast strains and dscDNA library was prepared which included samples barcoding and adapter ligation. Samples were loaded and sequenced using a PromethION R10 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data was collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–G) Differential gene expression in new1Δ-pEV(HIS3) versus WT-pEV(HIS3). Selected data plots showing PCA (B–C) and volcano plots (D–E) 1 hr and 24 hr post sequencing initiation. (F) Normalized gene count for selected genes 24 hr post sequencing initiation. Normalized gene counts are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 was used for normalization. (G) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points. (H–M) Differential gene expression in WT-pNew1(HIS3) versus WT-pEV(HIS3). Similar analyses to B-G were conducted for PCA (H–I) and volcano plots (J–K) as well as normalized gene counts (L) and Venn diagram (M). (N–S) Differential gene expression in new1Δ-pNew1(HIS3) versus new1Δ-pEV(HIS3). Similar analyses to B-G were conducted for PCA (N–O) and volcano plots (P–Q) as well as normalized gene counts (R) and Venn diagram (S). (T–Y) Differential gene expression in new1Δ-pNew1(HIS3) versus WT-pNew1(HIS3). Similar analyses to B-G were conducted for PCA (T–V) and volcano plots (U–W) as well as normalized gene counts (X) and Venn diagram (Y).
Figure 5—figure supplement 3
General overview in new1Δ-pEV(HIS3) vs WT-pEV(HIS3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 5—figure supplement 4
Gene expression in new1Δ-pEV(HIS3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points. (G) Overlap of up- and down-regulated DEGs between this study and Kasari et al., 2019.
Figure 5—figure supplement 5
Transcript expression new1Δ-pEV(HIS3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 5—figure supplement 6
General overview in WT-pNew1(HIS3) vs WT-pEV(HIS3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 5—figure supplement 7
Gene expression in WT-pNew1(HIS3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 5—figure supplement 8
Gene expression in WT-pNew1(HIS3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 5—figure supplement 9
General overview in new1Δ-pNew1(HIS3) vs new1Δ-pEV(HIS3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 5—figure supplement 10
Gene expression in new1Δ-pNew1(HIS3) vs new1Δ-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 5—figure supplement 11
Transcript expression in new1Δ-pNew1(HIS3) vs new1Δ-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 5—figure supplement 12
General overview in new1Δ-pNew1(HIS3) vs WT-pEV(HIS3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 5—figure supplement 13
Gene expression in new1Δ-pNew1(HIS3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) 5-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 5—figure supplement 14
Transcript expression in new1Δ-pNew1(HIS3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 5—figure supplement 15
General overview in new1Δ-pNew1(HIS3) vs WT-pNew1(HIS3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 5—figure supplement 16
Gene expression in new1Δ-pNew1(HIS3) vs WT-pNew1(HIS3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 5—figure supplement 17
Transcript expression in new1Δ-pNew1(HIS3) vs WT-pNew1(HIS3).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 6 with 15 supplements
Real-time transcriptomic analysis between rkr1Δ-pEV(URA3) vs WT-pEV(HIS3) using NanopoReaTA.
(A) Experimental strategy. Following library preparation (n=3), samples were loaded and sequenced using a PromethION R10 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data was collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–C) PCA plot showing the sample separation between rkr1Δ-pEV(URA3) vs WT-pEV(HIS3) 1 hr post-sequencing initiation (B) and 24 hr post-sequencing initiation (C). (D–E) Volcano plots showing the differentially expressed genes between rkr1Δ-pEV(URA3) vs WT-pEV(HIS3) 1 hr post-sequencing initiation (D) and 24 hr post-sequencing initiation (E). (F) Normalized gene count for selected genes 24 hr post sequencing initiation. Normalized gene counts are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 was used for normalization. (G) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
-
Figure 6—source data 1
General sequencing overview of Yeast setup 2.
- https://cdn.elifesciences.org/articles/98768/elife-98768-fig6-data1-v1.xlsx
Figure 6—figure supplement 1
Sequencing overview yeast setup 2.
(A) For yeast setup 2, WT strain (BY4741, MATa, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0), rkr1Δ::HphMX strain, where the rkr1 gene was replaced with the Hygromycin B resistance gene (HygR) and double KO jlp2Δ::KanMX rkr1Δ::HphMX, where the jlp2 gene was replaced with KanR. These strains were transformed with either an empty vector with the URA3 selection marker (pEV(URA3)) or an overexpression vector for C-terminally HA-tagged jlp2 with the URA3 selection marker (pJlp2(URA3)). WT was transformed with an empty vector with the His3 selection marker (pEV(HIS3)). (B) Total read generated per condition per time point. (B–E) The box plot represents total read generated (B), total mapped reads (C), total gene counts (D), and total transcript count (E) per condition, per time point (three biological replicates per condition). Abbreviations: WTpEV - WTpEV(HIS3), dRpEV - rkr1ΔpEV(URA3), dRpj - rkr1Δ pJlp2(URA3), dRjpEV - jlp2Δrkr1ΔpEV(URA3),dRjpj - jlp2Δrkr1ΔpJlp2(URA3).
Figure 6—figure supplement 2
Real-time transcriptomic analysis in yeast setup 2 samples using NanopoReaTA.
(A) Experimental strategy of yeast setup 2. RNA was isolated from selected yeast strains and the dscDNA library was prepared, which included sample barcoding and adapter ligation. Samples were loaded and sequenced using a PromethION R10 flow cell. NanopoReaTA was activated shortly after sequencing initiation and data was collected 1 hr, 2 hr, 5 hr, 10 hr, and 24 hr post-sequencing initiation. (B–G) Differential gene expression in rkr1Δ-pEV(URA3) versus WT-pEV(HIS3). Selected data plots showing PCA (B–C) and volcano plots (D–E) 1 hr and 24 hr post-sequencing initiation. (F) Normalized gene counts for selected genes 24 hr post-sequencing initiation. Normalized gene counts are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 was used for normalization. (G) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points. (H–M) Differential gene expression in rkr1Δ-pJlp2(URA3) versus rkr1Δ-pEV(URA3). Similar analyses to B-G were conducted for PCA (H–I) and volcano plots (J–K) as well as normalized gene counts (L) and Venn diagram (M). (N–S) Differential gene expression in rkr1Δ, jlp2Δ-pJlp2(URA3) versus rkr1Δ, jlp2Δ-pEV(URA3). Similar analyses to B-G were conducted for PCA (N–O) and volcano plots (P–Q) as well as normalized gene counts (R) and Venn diagram (S). (T–Y) Differential gene expression in rkr1Δ, jlp2Δ-pEV(URA3) versus WT-pEV(HIS3). Similar analyses to B-G were conducted for PCA (T–V) and volcano plots (U–W) as well as normalized gene counts (X) and Venn diagram (Y).
Figure 6—figure supplement 3
General overview in rkr1Δ-pEV(URA3) vs WT-pEV(HIS3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N). Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 6—figure supplement 4
Gene expression in rkr1Δ-pEV(URA3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) Five-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 6—figure supplement 5
Transcript expression in rkr1Δ-pEV(URA3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 6—figure supplement 6
General overview in rkr1Δ-pJlp2(URA3) vs rkr1Δ-pEV(URA3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 6—figure supplement 7
Gene expression in rkr1Δ-pJlp2(URA3) vs rkr1Δ-pEV(URA3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) 5-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 6—figure supplement 8
Transcript expression in rkr1Δ-pJlp2(URA3) vs rkr1Δ-pEV(URA3).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 6—figure supplement 9
General overview in rkr1Δjlp2Δ-pJlp2(URA3) vs rkr1Δjlp2Δ-pEV(URA3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 6—figure supplement 10
Gene expression in rkr1Δjlp2Δ-pJlp2(URA3) vs rkr1Δjlp2Δ-pEV(URA3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) 5-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 6—figure supplement 11
Transcript expression inrkr1Δjlp2Δ-pJlp2(URA3) vs rkr1Δjlp2Δ-pEV(URA3).
(A) Principal component analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 6—figure supplement 12
General overview in rkr1Δjlp2Δ-pEV(URA3) vs WT-pEV(HIS3).
(A–B) Read length overview. The distribution of read lengths derived from generated fastq files is plotted per sample (A) and per condition (B). All reads of length over the 99% quantile of all lengths are removed from these visualizations. (C–D) Gene expression variability. The number of identified genes (>0 reads counted) is plotted after each iteration per sample (C) and per condition (D). When the detection of additional genes finishes, the lines reach a plateau. (E–F) Change in gene composition. The change in gene composition (CGC) plot shows the sum of absolute differences of relative gene (%gi) counts per individual gene (g) in iteration (i) to the relative gene counts in iteration (i-1) for each respective sample (E) and condition (F). CGC = sum(|(%g(i-1))-(%g(i))|). (G) Process time. The bar plots show the time in seconds all preprocessing steps needed per iteration. One iteration process at a maximum of 30 files from all samples. This plot updates automatically as a new process finishes. (H–N) Final general overview. Read length (H–I), Gene expression variability (J–K), Change in gene composition (L–M), and processing time (N) for the 24 hr time point. All plots are organized according to their respective time points of collection.
Figure 6—figure supplement 13
Gene expression in rkr1Δjlp2Δ-pEV(URA3) vs WT-pEV(HIS3).
(A) Principal component analysis (PCA) of the top 500 genes with the highest variance across all samples, the two principal components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample distance plot. The Euclidean distance between the gene expression patterns of all samples to each other is plotted using heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential gene expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2 (Love et al., 2014). The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed genes is plotted using Heatmap by coloring the number of reads per gene. The legend on the right side describes which colors correspond to highly and lowly expressed genes. (E) Normalized gene count for selected genes. Normalized gene counts from FeatureCounts (Liao et al., 2014) are visualized for selected genes per condition using boxplots. The median-of-ratios normalization method from DESeq2 (Love et al., 2014) was used for normalization. (F) 5-way Venn diagram showing the differentially expressed gene overlaps between the distinct collected time points.
Figure 6—figure supplement 14
Transcript expression in rkr1Δjlp2Δ-pEV(URA3) vs WT-pEV(HIS3).
(A) Principal component Analysis (PCA) of the top 500 transcripts with the highest variance across all samples, the two Principal Components. Each dot corresponds to one sample and is colored by the respective condition. (B) Sample-to-sample plot. The Euclidean distance between the transcript expression patterns of all samples to each other is plotted using a heatmap. Normalized gene counts were used for distance computation. (C) Volcano plots. The differential transcript expression analysis results from DESeq2 are shown by plotting the log2FoldChange between the conditions of interest against the –log10 adjusted p-value per gene observed from the Wald-Test integrated in DESeq2. The top 10 significant genes are labeled by their symbol. (D) Heatmaps. The expression of the top 20 differentially expressed transcripts is plotted using Heatmap by coloring the number of reads per transcript. The legend on the right side describes which colors correspond to highly and lowly expressed transcripts.
Figure 6—figure supplement 15
IGV coverage snapshots of selected gene loci in yeast setup 1 and 2.
(A–F) Integrative Genome Viewer (IGV) snapshots of reads mapping to AmpR, HygR, and KanR (A), NEW1 (YPL226W) (B), RKR1 (YMR247C) (C), JLP2 (YMR132C) (D), HIS3 (YOR202W) (E), and URA3 (YEL021W) (F).
Appendix 1—figure 1
Comparison of # genes detected in different experimental conditions.
(A) PromethION 10 replicate experimental setup. (B) PromethION 2 replicate experimental setup (C) MinION 2 replicate experimental setup (D) TotalR/Ribominus/Riboplus comparison. The plots present the number of identified genes after each iteration per sample (left) and per condition (right).
Appendix 2—figure 1
IGV coverage snapshots of selected gene loci in yeast setup 1 and 2.
Top, IGV snapshots of reads mapping to AmpR, HygR, and KanR. Bottom, IGV snapshots of reads mapping to HIS3 (YOR202W).
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Cell line (Homo sapiens) | HEK293 | ATCC | CRL-1573; RRID:CVCL_0045 | |
| Cell line (H. sapiens) | HeLa | ATCC | CCL-2; RRID:CVCL_0030 | |
| Strain (Saccharomyces cerevisiae) | BY4741 | Brachmann et al., 1998 | https://doi.org/10.1002/(SICI)1097-0061(19980130)14:2<115::AID-YEA204>3.0.CO;2–2 Available as well from: Euroscarf: Y00000; ATCC: 4040002 | Genotype: MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 |
| Strain (S. cerevisiae) | WTpEV | This study | Genotype: BY4741+pEV(HIS3); available from M.L. Winz | |
| Strain (S. cerevisiae) | WTpN | This study | Genotype: BY4741+pNew1-Flag(HIS3); available from M.L. Winz | |
| Strain (S. cerevisiae) | dNpEV | Müller et al., 2026 | https://doi.org/10.1093/nar/gkag047 | Genotype: new1Δ::KanMX +pEV(HIS3); available from M.L. Winz |
| Strain (S. cerevisiae) | dNpN | Müller et al., 2026 | https://doi.org/10.1093/nar/gkag047 | Genotype: new1Δ::KanMX +pNew1-Flag(HIS3); available from M.L. Winz |
| Strain (S. cerevisiae) | dRpEV | This study | Genotype: rkr1Δ::HphMX +pEV(URA3); available from M.L. Winz | |
| Strain (S. cerevisiae) | dRpJ | This study | Genotype: rkr1Δ::HphMX +pJLP2 HA(URA3); available from M.L. Winz | |
| Strain (S. cerevisiae) | dRJpEV | This study | Genotype: jlp2Δ::KanMX rkr1Δ::HphMX +pEV(URA3); available from M.L. Winz | |
| Strain (S. cerevisiae) | dRJpJ | Iyer et al., 2025 | https://doi.org/10.1101/2025.09.04.673968 | Genotype: jlp2Δ::KanMX rkr1Δ::HphMX +pJLP2 HA(URA3); available from M.L. Winz; Note that RKR1 is designated by its alias LTN1 in the cited reference. |
| Recombinant DNA reagent | pEV(HIS3) | Müller et al., 2026 | https://doi.org/10.1093/nar/gkag047 | Shuttle vector plasmid: empty vector, HIS3 selection marker, available from M.L. Winz |
| Recombinant DNA reagent | pNew1(HIS3) | Müller et al., 2026 | https://doi.org/10.1093/nar/gkag047 | Shuttle vector plasmid: encoding C-terminal FLAG-tagged New1, HIS3 selection marker, available from M.L. Winz |
| Recombinant DNA reagent | pEV(URA3) | This study | N/A | Shuttle vector plasmid: empty vector, URA3 selection marker, available from M.L. Winz |
| Recombinant DNA reagent | pJlp2(URA3) | Iyer et al., 2025 | https://doi.org/10.1101/2025.09.04.673968 | Shuttle vector plasmid: encoding C-terminal HA-tagged Jlp2, URA3 selection cassette, available from M.L. Winz |
| Chemical compound, drug | TRIzol | Thermo Fisher Scientific | Cat#: 15596018 | |
| Chemical compound, drug | Chloroform | Roth | Cat#: 7331.2 | |
| Chemical compound, drug | Isopropanol | Roth | Cat#: 6752.2 | |
| Chemical compound, drug | Chloroform:isoamyl alcohol (24:1) | Roth | Cat#: X984.1 | |
| Chemical compound, drug | Ethanol | Roth | Cat#: 9065.3 | |
| Commercial assay or kit | Ribominus Eukaryote Kit for RNA-seq | Thermo Fisher / Ambion | Cat#: A10837-08 | |
| Commercial assay or kit | Maxima H Minus Double-Stranded cDNA Synthesis Kit | Thermo Scientific | Cat#: K2561 | |
| Commercial assay or kit | RNaseOUT | Invitrogen | Cat#:10777019 | |
| Commercial assay or kit | NEBNext Ultra II End Repair / dA-Tailing Module | New England Biolabs | Cat#: E7546 | |
| Commercial assay or kit | Native Barcoding Expansion 1–12 | Oxford Nanopore Technologies | Cat#: EXP-NBD104 | |
| Commercial assay or kit | RNase I | Thermo Fisher Scientific | Cat#: EN0601 | |
| Commercial assay or kit | Quick T4 DNA Ligase | New England Biolabs Inc. | Cat#: E6056 | |
| Commercial assay or kit | AMPure XP beads | Beckman Coulter | Cat#: A63881 | |
| Commercial assay or kit | Qubit Fluorometric Quantitation | Thermo Fisher Scientific | Cat# Q33238 | |
| Commercial assay or kit | Flow Cell Wash Kit | Oxford Nanopore Technologies | Cat#: EXP-WSH003 | |
| Commercial assay or kit | PromethION Flow cell (R10.4.1) | Oxford Nanopore Technologies | Cat#: FLO-PRO114M | |
| Commercial assay or kit | MinION Flow Cell (R9.4.1) | Oxford Nanopore Technologies | Cat#: FLO-MIN106 | |
| Software, algorithm | Guppy basecaller | Oxford Nanopore Technologies | v3.6.1 | Used in high-accuracy (hac) mode for PromethION and super-accuracy mode for MinION |
| Software, algorithm | NanopoReaTA | Wierczeiko et al., 2023 | https://doi.org/10.1093/bioinformatics/btad492 | Real-time transcriptomic analysis pipeline; All associated software/algorithm is reported in this publication. See more information in https://github.com/AnWiercze/NanopoReaTA (Wierczeiko, 2024) |
Appendix 1—table 1
HEK293 versus HeLa experimental setup (10 replicates per condition).
Total amount of cDNA converted from 2 μg RNA and utilized in PromethION and MinION sequencing.
| HEK293 | HeLa | cDNA taken for sequencing | ||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| R1 | R2 | R3 | R4 | R5 | R6 | R7 | R8 | R9 | R10 | R1 | R2 | R3 | R4 | R5 | R6 | R7 | R8 | R9 | R10 | |||
| Total amount of cDNA generated (ng) | PromethION | 516 | 496 | 480 | 516 | 508 | 512 | 464 | 480 | 456 | 532 | 348 | 404 | 424 | 420 | 508 | 492 | 440 | 468 | 488 | 432 | 300 ng (both HEK293 and HeLa) |
Appendix 1—table 2
HEK293 versus HeLa experimental setup (two replicates per condition).
Total amount of cDNA converted from 2 μg RNA and utilized in PromethION and MinION sequencing.
| HEK293 | HeLa | cDNA taken for sequencing | ||||
|---|---|---|---|---|---|---|
| R1 | R2 | R1 | R2 | |||
| Total amount of cDNA generated (ng) | PromethION | 1,116 (bc1) | 1,172 (bc2) | 492 (bc3) | 440 (bc4) | 400 ng (both HEK293 and HeLa) |
| MinION | 888 (bc4) | 824 (bc5) | 452 (bc6) | 564 (bc7) | All cDNA per replicate | |
Appendix 1—table 3
rRNA-depleted and rRNA-enriched transcripts setup.
Total amount of cDNA converted from 2 μg RNA (500 ng for RiboM) and utilized in PromethION sequencing.
| Total R | Ribominus | Riboplus | ||||
|---|---|---|---|---|---|---|
| R1 | R2 | R1 | R2 | R1 | R2 | |
| Total amount of cDNA generated (ng) | 1,116 (bc1) | 1,172 (bc2) | 132 (bc5) | 163 (bc6) | 364 (bc7) | 2,120 (bc8) |
| cDNA taken for sequencing | 400 ng | 400 ng | 132 ng | 163 ng | 364 ng | 400 ng |
Additional files
-
MDAR checklist
- https://cdn.elifesciences.org/articles/98768/elife-98768-mdarchecklist1-v1.pdf
-
Supplementary file 1
Step-by-step guideline to NanopoReaTA.
- https://cdn.elifesciences.org/articles/98768/elife-98768-supp1-v1.docx
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Real-time transcriptomic profiling in distinct experimental conditions
eLife 13:RP98768.
https://doi.org/10.7554/eLife.98768.3
