Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index.

IVT-seq reveals extreme bias in RNA sequencing

Open access

Published version of a preprint

Abstract

Background: RNA-seq is a powerful technique for identifying and quantifying transcription and splicing events, both known and novel. However, given its recent development and the proliferation of library construction methods, understanding the bias it introduces is incomplete but critical to realizing its value. Results: We present a method, in vitro transcription sequencing (IVT-seq), for identifying and assessing the technical biases in RNA-seq library generation and sequencing at scale. We created a pool of over 1,000 in vitro transcribed RNAs from a full-length human cDNA library and sequenced them with polyA and total RNA-seq, the most common protocols. Because each cDNA is full length, and we show in vitro transcription is incredibly processive, each base in each transcript should be equivalently represented. However, with common RNA-seq applications and platforms, we find 50% of transcripts have more than two-fold and 10% have more than 10-fold differences in within-transcript sequence coverage. We also find greater than 6% of transcripts have regions of dramatically unpredictable sequencing coverage between samples, confounding accurate determination of their expression. We use a combination of experimental and computational approaches to show rRNA depletion is responsible for the most significant variability in coverage, and several sequence determinants also strongly influence representation. Conclusions: These results show the utility of IVT-seq for promoting better understanding of bias introduced by RNA-seq. We find rRNA depletion is responsible for substantial, unappreciated biases in coverage introduced during library preparation. These biases suggest exon-level expression analysis may be inadvisable, and we recommend caution when interpreting RNA-seq results.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/gb-2014-15-6-r86

Cites

26 of 26 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Construction of IVT-seq libraries. (A) Preparation of a pool of 1,062 human cDNA plasmids. Contents of three 384-well plates containing MGC plasmids were pooled together. Pool was amplified via transformation in Escherichia coli, and resulting clones were purified and re-pooled. (B) Generation of IVT transcripts. Pool of MGC plasmids was linearized and used as a template for an in vitro transcription reaction. Enzymes and unincorporated nucleotides were purified, leaving pool of polyA transcripts. (C) Creation of IVT-seq libraries. Listed quantities of IVT RNA were mixed with mouse liver total RNA to create six pools with final RNA quantities of 1 μg. Ribosomal RNA was depleted from these pools using the Ribo-Zero Gold kit. IVT RNA and mouse RNA are now present in pools at the listed ratios, following depletion of rRNA from mouse total RNA. These pools were used to generate RNA-seq libraries using Illumina’s TruSeq kit/protocol. This entire process was performed in duplicate. Replicate libraries were pooled separately and sequenced in separate HiSeq 2000 lanes (two lanes total). IVT, in vitro transcribed; MGC, Mammalian Gene Collection.
Figure 1 Construction of IVT-seq libraries. (A) Preparation of a pool of 1,062 human cDNA plasmids. Contents of three 384-well plates containing MGC plasmids were pooled together. Pool was amplified via transformation in Escherichia coli, and resulting clones were purified and re-pooled. (B) Generation of IVT transcripts. Pool of MGC plasmids was linearized and used as a template for an in vitro transcription reaction. Enzymes and unincorporated nucleotides were purified, leaving pool of polyA transcripts. (C) Creation of IVT-seq libraries. Listed quantities of IVT RNA were mixed with mouse liver total RNA to create six pools with final RNA quantities of 1 μg. Ribosomal RNA was depleted from these pools using the Ribo-Zero Gold kit. IVT RNA and mouse RNA are now present in pools at the listed ratios, following depletion of rRNA from mouse total RNA. These pools were used to generate RNA-seq libraries using Illumina’s TruSeq kit/protocol. This entire process was performed in duplicate. Replicate libraries were pooled separately and sequenced in separate HiSeq 2000 lanes (two lanes total). IVT, in vitro transcribed; MGC, Mammalian Gene Collection.CC BY · PubMed Central
Figure 2 Within-transcript variations in RNA-seq coverage. (A) Simulated RNA-seq coverage for a representative IVT transcript (BC015891). RNA-seq coverage plot (black) is displayed according to the gene model (green), as it is mapped to the reference genome. Blocks correspond to exons and lines indicate introns. The chevrons within the intronic lines indicate the direction of transcription. Numbers on y-axis refer to RNA-seq read-depth at a given nucleotide position. (B) The actual RNA-seq coverage plot for BC015891 in the IVT-only sample. Representative coverage plots for the IVT transcripts (C) BC009037 and (D) BC016283 are displayed according to the same conventions used above. All transcripts are displayed in the 5ʹ to 3ʹ direction.
Figure 2 Within-transcript variations in RNA-seq coverage. (A) Simulated RNA-seq coverage for a representative IVT transcript (BC015891). RNA-seq coverage plot (black) is displayed according to the gene model (green), as it is mapped to the reference genome. Blocks correspond to exons and lines indicate introns. The chevrons within the intronic lines indicate the direction of transcription. Numbers on y-axis refer to RNA-seq read-depth at a given nucleotide position. (B) The actual RNA-seq coverage plot for BC015891 in the IVT-only sample. Representative coverage plots for the IVT transcripts (C) BC009037 and (D) BC016283 are displayed according to the same conventions used above. All transcripts are displayed in the 5ʹ to 3ʹ direction.CC BY · PubMed Central
Figure 3 Between-sample variations in RNA-seq coverage. RNA-seq coverage plots across all samples for exons 4 to 11 of the IVT transcript BC003355. The black rectangles identify exon six, which shows extreme variability in coverage relative to the rest of the transcript when viewed across all of the samples. The ratio of IVT RNA to mouse RNA is listed to the left of each sample’s coverage plots. Coverage plots (red for first replicate; blue for second replicate) are displayed according to the gene model (black), as it is mapped to the reference genome. Blocks in the gene model correspond to exons and lines indicate introns. The chevrons within the intronic lines indicated the direction of transcription. Numbers on y-axes refer to RNA-seq read-depth at a given nucleotide position.
Figure 3 Between-sample variations in RNA-seq coverage. RNA-seq coverage plots across all samples for exons 4 to 11 of the IVT transcript BC003355. The black rectangles identify exon six, which shows extreme variability in coverage relative to the rest of the transcript when viewed across all of the samples. The ratio of IVT RNA to mouse RNA is listed to the left of each sample’s coverage plots. Coverage plots (red for first replicate; blue for second replicate) are displayed according to the gene model (black), as it is mapped to the reference genome. Blocks in the gene model correspond to exons and lines indicate introns. The chevrons within the intronic lines indicated the direction of transcription. Numbers on y-axes refer to RNA-seq read-depth at a given nucleotide position.CC BY · PubMed Central
Figure 4 Sources of bias in RNA-seq coverage. (A) RNA-seq coverage plots for IVT transcript BC003355 from simulated (black), plasmid (blue), no selection (green), rRNA-depleted (red), and polyA (orange) data. The gene model is displayed in black, below all of the coverage plots. Blocks correspond to exons and lines indicate introns. The chevrons within the intronic lines indicate the direction of transcription. (B) Distributions for coefficients of variation across data displayed above, with the addition of the QM-simulated data (gray). Note that while the graph is cutoff at a coefficient of variation of 1.3, the tails for the rRNA-depleted and polyA distributions extend out to 2.13 and 2.7, respectively. (C) Effect sizes for the differences in distribution of coefficients of variation between sequencing libraries and simulated data. Effect sizes are calculated as the per-transcript ratios of coefficients of variation between a given library and the simulated dataset. QM, quantity matched.
Figure 4 Sources of bias in RNA-seq coverage. (A) RNA-seq coverage plots for IVT transcript BC003355 from simulated (black), plasmid (blue), no selection (green), rRNA-depleted (red), and polyA (orange) data. The gene model is displayed in black, below all of the coverage plots. Blocks correspond to exons and lines indicate introns. The chevrons within the intronic lines indicate the direction of transcription. (B) Distributions for coefficients of variation across data displayed above, with the addition of the QM-simulated data (gray). Note that while the graph is cutoff at a coefficient of variation of 1.3, the tails for the rRNA-depleted and polyA distributions extend out to 2.13 and 2.7, respectively. (C) Effect sizes for the differences in distribution of coefficients of variation between sequencing libraries and simulated data. Effect sizes are calculated as the per-transcript ratios of coefficients of variation between a given library and the simulated dataset. QM, quantity matched.CC BY · PubMed Central
Figure 5 Effects of sequence characteristics on coverage variability. Distributions of (A) hexamer entropy, (B) GC-content, and (C) rRNA sequence similarity for the 100 transcripts with the highest and lowest coefficients of variation for transcript coverage from the plasmid, no selection, rRNA-depleted, and polyA libraries. Asterisks indicate the significance of a Wilcoxon signed-rank test comparing values for the listed sequence characteristics between each pair of groups from the same sample. *P <0.05; **P <0.01; ***P <0.001.
Figure 5 Effects of sequence characteristics on coverage variability. Distributions of (A) hexamer entropy, (B) GC-content, and (C) rRNA sequence similarity for the 100 transcripts with the highest and lowest coefficients of variation for transcript coverage from the plasmid, no selection, rRNA-depleted, and polyA libraries. Asterisks indicate the significance of a Wilcoxon signed-rank test comparing values for the listed sequence characteristics between each pair of groups from the same sample. *P <0.05; **P <0.01; ***P <0.001.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 26 references

Cited by

See all 158

Metadata

Authors
Lahens, N. F., Kavakli, I. H., Zhang, R., Hayer, K., Black, M. B., Dueck, H., Pizarro, A., Kim, J., Irizarry, R., Thomas, R. S., Grant, G. R., & Hogenesch, J. B.
Affiliations
  • University of Pennsylvania (US)
  • The Hamner Institutes for Health Sciences (US)
  • Amazon (United States) (US)
  • Johns Hopkins University (US)
Corresponding author
Hogenesch, John B
Venue
Genome Biology
Published
2014-06-30
Volume
15
Issue
6
Pages
R86-R86
ISSN
  • 1465-6906 (ISSN-L)
  • 1474-760X
OpenAlex
W2098809361
Cited by
158
References
26
Citations by year
2026: 13 · 2025: 6 · 2024: 10 · 2023: 6 · 2022: 4 · 2021: 21 · 2020: 9 · 2019: 13 · 2018: 13 · 2017: 20 · 2016: 13 · 2015: 25 · 2014: 5
FWCI
8.60
Type
journal article
Language
en
Open access
Licence
Locations
  • Genome biology (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Digital Collections portal (Koç University) (submittedVersion)
Funders
  • Pennsylvania Department of Health
  • University of Pennsylvania (P30DK19525, UL1-TR-000003)
  • Institute for Translational Medicine and Therapeutics (UL1TR000003)
  • National Institutes of Health (5-R01-, 12-DARPA-1068, UL1TR000003)
  • Defense Advanced Research Projects Agency
  • Perelman School of Medicine, University of Pennsylvania (UL1TR000003)
  • National Heart, Lung, and Blood Institute (R01HL097800)
  • National Human Genome Research Institute (T32HG000046)
  • National Institute of Diabetes and Digestive and Kidney Diseases (P30 DK 19525, P30DK019525)
  • National Institute of Neurological Disorders and Stroke (R01NS054794, 2-R01-NS054794-06)
  • National Center for Research Resources
  • National Center for Advancing Translational Sciences
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene expression and cancer classification
Keywords
technical bias, rRNA depletion, RNA-seq library preparation
MeSH terms
Animals, Base Composition, Base Sequence, Humans, Male, Mice, Inbred C57BL, RNA, Ribosomal, Sequence Homology, Nucleic Acid, Transcription, Genetic, Gene Library, Artifacts, Sequence Analysis, RNA, In Vitro Techniques

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: IVT-seq reveals extreme bias in RNA sequencing. https://app.cochat.ai/papers/10.1186/gb-2014-15-6-r86
The paper
Lahens, N. F., Kavakli, I. H., Zhang, R., Hayer, K., Black, M. B., Dueck, H., Pizarro, A., Kim, J., Irizarry, R., Thomas, R. S., Grant, G. R., & Hogenesch, J. B. (2014). IVT-seq reveals extreme bias in RNA sequencing. Genome Biology. https://doi.org/10.1186/gb-2014-15-6-r86

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.