The Transcriptional Landscape of the Yeast Genome Defined by RNA Sequencing
Science, 2008 · paper · Biomedicine
Applies RNA-Seq, sequencing cDNA fragments on the Illumina platform, to map the yeast transcriptome. It finds 74.5% of unique genome sequence transcribed, maps UTR boundaries and introns, and reports RNA-Seq at least as accurate as microarrays.
7 findings
Substance The text reads "29,912,517 total reads, 15,870,540 (56%) were mapped to unique regions", but 15,870,540/29,912,517 is 53.1%.
- p. 3 Results, first line
29,912,517 total reads, 15,870,540 (56%) were mapped to unique regions
- p. 3 Results, first line
29,912,517 total reads, 15,870,540 (56%) were mapped
- p. 3 Mapping section, p3 first lines
15,870,540 (56%) were mapped to unique regions
Judge's note: Checked p2 and p3. The total equals the two library sizes given on p2 (14,125,182 + 15,787,335), and the paper states no other denominator. 15,870,540 is larger than either library alone, so a per-library denominator cannot give 56% either.
Second check: another model agreed
Report a problem with this finding
Substance The text reads "Comparison of these results with 1025 boundaries that we mapped using 5′ RACE ... 786 genes (77.9%; Fig. 3A, top left panel)", but the cited panel shows "786/1009 77.9%", and 786/1025 is 76.7%.
- p. 4 Results, 5′ boundary paragraph
Comparison of these results with 1025 boundaries
- p. 4 Results, 5′ boundary paragraph
786 genes (77.9%; Fig. 3A, top left panel)
- p. 12 Fig. 3A top-left panel label
786/1009 77.9%
- p. 4 Results, 5′ end mapping paragraph
Comparison of these results with 1025 boundaries that we mapped using 5′ RACE
Judge's note: The Fig. 3A top-left panel on p12 reads '+/- 50bp 786/1009 77.9%'. 77.9% matches 1009, not 1025. The text gives no reason for 1025 boundaries to shrink to 1009, so the text denominator disagrees with both its own percentage and the figure.
Second check: another model agreed
Report a problem with this finding
Reference or label The text cites "(Fig. 2C and 2D)" for the 74.5% expressed figure, but the published Fig. 2 has only panels A and B; the legend's descriptions of B, C and D do not match the panels shown.
- p. 3 Extensive Expression section
74.5% of the genome is expressed, at least at a low level (Fig. 2C and 2D)
Only the authors can say what is right here.
Judge's note: Fig. 2 on p11 shows only panel A (chromosome map) and panel B (pie chart). The legend still describes B as 'chromosome VI', C as 'Histogram of transcribed bases' and D as 'A summary of the transcription level'. The 74.5% data appear to come from the pie labelled B. The figure, its legend and the text pointers do not agree, and only the authors can say which layout is intended.
Second check: another model agreed
Report a problem with this finding
Reference or label The Fig. 2 legend describes panels A–D ('B) Distribution of transcribed regions on chromosome VI. C) Histogram of transcribed bases. D) A summary…'), but the figure has only panel A (genome map) and panel B (expression pie chart). The text on p.3 cites 'Fig. 2C and 2D'.
- p. 11 Figure 2 and caption
B) Distribution of transcribed regions on chromosome VI. C) Histogram of transcribed bases. D) A summary of the transcription level
- p. 11 Figure 2 caption
B) Distribution of transcribed regions on chromosome VI. C) Histogram of transcribed bases. D) A summary of the transcription level of the transcriptome.
- p. 3 Extensive Expression section
is expressed, at least at a low level (Fig. 2C and 2D)
Judge's note: p.11 image: only panels A and B are present. B is a pie chart of no/low/medium/high/repeat-masked fractions, which matches the legend's D description rather than its B description. Panels C and D do not exist, so the text pointers on p.3 point at absent panels.
Second check: another model agreed
Report a problem with this finding
Reference or label Figure 3 contains panel E (chrXVI, YPL075W (GCR1) with SGD_Intron), which the text cites as 'Fig. 3E', but the legend on pp.12–13 stops at panel D and never describes E.
- p. 12 Figure 3 panel E
YPL075W (GCR1)
- p. 13 Figure 3 caption end
information were not used in this case due to low scores.
- p. 5 Intron paragraph
(Fig. 3E and fig. S5)
- p. 12 Figure 3 legend
D) 3′-UTR determined by
Judge's note: p.12 image shows panel E. The legend continues on p.13 with only the end of the D text and the abbreviations. Two reviewers said panel E does not exist; that is wrong. The defect is the missing legend entry.
Second check: another model agreed
Report a problem with this finding
Reference or label The Fig. 3C legend reads 'end tags for gene YDR460W, YDR004W, and YDR461-C, or YDR004W that is also determined by cDNA sequencing', but the panel is labelled YDR460W, YDR461W and YDR461C-A, with the Miura 3′UTR over YDR461W.
- p. 12 Figure 3 caption, panel C
end tags for gene YDR460W, YDR004W, and YDR461-C, or YDR004W that is also
- p. 12 Figure 3C panel labels
YDR461C-A
- p. 12 Figure 3 legend C
YDR460W, YDR004W, and YDR461-C, or YDR004W
- p. 12 Fig. 3C labels
YDR461W
Judge's note: p.12 image: the panel C labels are legible as YDR460W, YDR461W and YDR461C-A. The legend repeats YDR004W, which is not in the panel, and gives the malformed 'YDR461-C'.
Second check: another model agreed
Report a problem with this finding
Reference or label The Fig. 4 legend refers to 'a previously well annotated region on chrVI (A) and a relatively poor annotated region on the same chromosome (B)', but the figure has a single unlabelled panel (chrVI about 232k–237k). The legend's 'cDNA transcripts in red … blue' also does not match the blue and orange tracks shown.
- p. 14 Figure 4 caption
a previously well annotated region on chrVI (A) and a relatively poor annotated region on the same chromosome (B)
- p. 14 Figure 4 legend
previously well annotated region on chrVI (A) and a relatively poor annotated region on the same chromosome (B)
Judge's note: p.14 image: one panel only, with no A/B letters. The Miura cDNA track shows blue and orange bars, not red.
Second check: another model agreed
Report a problem with this finding
16 minor (typos and the like; most checked by the judge only)
Reference or label The text cites "4,666 of the 5099 annotated ORFs (91.5%) in the Saccharomyces Genome Database (SGD) (Fig. 2)", but Fig. 2 shows per-base genome coverage; the per-ORF distribution for 5,099 ORFs is in Fig. 6D.
- p. 3 Extensive Expression section, second paragraph
(91.5%) in the Saccharomyces Genome Database (SGD) (Fig. 2)
- p. 16 Fig. 6D caption
D) Transcription level distribution for 5,099 ORFs by RNA-Seq.
- p. 3 Results, transcriptional activity
4,666 of the 5099 annotated ORFs (91.5%) in the Saccharomyces Genome Database (SGD) (Fig. 2)
- p. 16 Fig. 6D legend
Transcription level distribution for 5,099 ORFs by RNA-Seq.
Only the authors can say what is right here.
Judge's note: The published Fig. 2 (p11) contains only a per-base chromosome map and a per-base pie chart, with no ORF counts. The Fig. 6D legend (p16) reads 'Transcription level distribution for 5,099 ORFs by RNA-Seq'. The missing Fig. 2D ('summary of the transcription level') might once have held this content, so this is a query rather than a definite wrong pointer. The arithmetic itself is consistent.
Second check: another model agreed
Report a problem with this finding
Reference or label The Fig. 5E x-axis is labelled 'Covered by Miura et al. (%)' and the legend says 'in percentages', but the tick labels run 0–5000.
- p. 15 Fig. 5E x-axis
Covered by Miura et al. (%)
Only the authors can say what is right here.
Judge's note: p.15 image: panel E ticks read 0, 1000 … 5000, with a y-axis of #Regions up to 250. A percentage axis cannot exceed 100, so either the unit label or the scale is wrong. The values may be in bp; only the authors can say which.
Second check: another model agreed
Report a problem with this finding
Wording The abstract says "most (74.5%) of the unique sequence of the yeast genome is transcribed", but the Results ("a total of 74.5% of the genome is expressed") and the Discussion ("overall much (74.5%) of the yeast genome is transcribed") give the same figure as a share of the whole genome. The Fig. 2B pie shows 70% of the genome expressed and 6% repeat-masked.
- p. 1 Abstract
most (74.5%) of the unique sequence of the yeast genome is transcribed
- p. 3 Extensive Expression section
a total of 74.5% of the genome is expressed
- p. 8 Discussion
overall much (74.5%) of the yeast genome is
- p. 8 Discussion, novel transcribed regions paragraph
overall much (74.5%) of the yeast genome is transcribed
Only the authors can say what is right here.
Judge's note: Checked p1, p3, p8 and the Fig. 2B pie on p11. The pie shows 24% no expression, 44% low, 19% medium, 7% high and 6% repeat-masked. Low + medium + high is 70% of the whole genome; 70/94 = 74.5% is the share of non-repeat sequence, which matches the abstract's wording. The Results and Discussion drop the 'unique sequence' qualifier. 'Genome' may be loose shorthand, so the authors should state which denominator is meant.
Second check: another model agreed
Report a problem with this finding
Wording Conclusion claims the method can 'define for the first time UTRs', but p.3 says 'A cDNA sequencing study did reveal 5′ UTRs of many genes … (7)', and Fig. 3 compares against 3′ UTRs from Miura et al.
- p. 8 Conclusion
define for the first time UTRs and many novel transcribed regions
- p. 3 Results, gene boundaries intro
A cDNA sequencing study did reveal 5′ UTRs of many genes
Only the authors can say what is right here.
Judge's note: Checked p.8 (conclusion) and p.3 (gene-boundaries paragraph). The unqualified priority claim conflicts with the paper's own account of ref. 7. The authors may have meant 'genome-wide'. This is a wording overstatement.
Second check: another model agreed
Report a problem with this finding
Wording The text gives "DNA binding proteins (P < 0.0027, FDR adjusted)", while the Fig. 5B label appears to read "FDR p < 2.07 e-3"; the second term agrees in both places (0.0045 vs 4.49e-3).
- p. 5 uORF section
DNA binding proteins (P < 0.0027, FDR adjusted)
- p. 15 Fig. 5B label
FDR p < 2.07 e-3
- p. 5 Results, uORF GO analysis
(P < 0.0027, FDR adjusted)
Only the authors can say what is right here.
Judge's note: In the p15 image the Fig. 5B label reads most plausibly as '2.07 e-3', but the print is very small and the middle digit cannot be read with full certainty. It is the same FDR-adjusted GO enrichment printed in two places, and the values look like a digit transposition. The authors should confirm which value is right.
Second check: another model agreed
Report a problem with this finding
Wording The text says "this work provides the first experimental sequence confirmation of nearly all yeast introns" right after reporting that the algorithm "readily detected 240 of 306 known introns" (78%).
- p. 5 Results, introns paragraph
the first experimental sequence confirmation of nearly all yeast introns
- p. 5 Results, introns paragraph
readily detected 240 of 306 known introns
Judge's note: On p5, 66 known introns were not validated, and the next paragraph says only 30 of them belong to expressed genes. 78% is more naturally 'most' than 'nearly all'. The authors may mean introns in expressed genes, but the text does not say so.
Second check: another model agreed
Report a problem with this finding
Wording Results support 'better than … standard expression microarrays (R=0.72; Fig. 6C. (15)' by comparing R=0.98 (RNA-Seq vs qPCR, 34 genes) with R=0.72 (RNA-Seq vs Holstege microarray, 4,422 ORFs). The parenthesis is left unclosed, and the Discussion goes on to say RNA-Seq was 'significantly better' with no test reported.
- p. 6 Discussion
revealed that RNA-Seq was significantly better for quantification of RNA levels than standard gene expression microarrays
- p. 6 Results
(R=0.72; Fig. 6C.
Only the authors can say what is right here.
Judge's note: Checked p.6 and Fig. 6 on p.16. Fig. 6A is corr=0.9775 (qPCR vs RNA-Seq, 34 ORFs). Fig. 6C is corr=0.7232 (Holstege et al. vs RNA-Seq, 4,422 ORFs). The two correlations compare different pairs of methods on different gene sets, and the paper reports no significance test, so 'significantly better' cannot be traced to a test in the paper. The image also shows the bracket opened at '(R=0.72; Fig. 6C.' is never closed. Wording and ambiguity only.
Second check: another model agreed
Report a problem with this finding
Typo The affiliation prints "Department of Molecular, Biophysics and Biochemistry" with a stray comma inside the department name.
- p. 1 Affiliation
Department of Molecular, Biophysics and Biochemistry
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the Yale department is named 'Molecular Biophysics and Biochemistry' with no comma. The p1 image shows the comma exactly as quoted. 'Department of Molecular' would not be a department name on its own, so the comma looks like a copy-edit slip.
Second check: none (judge only)
Report a problem with this finding
Typo The Introduction reads "through the presence of ORFs and subsequently though the identification of sequences conserved"; "though" should be "through".
- p. 2 Introduction
though the identification of sequences conserved
- p. 2 Introduction
subsequently though the identification of sequences conserved
- p. 2 Intro para 4
subsequently though the identification
Judge's note: The p2 image shows 'though' as quoted. The sentence only reads correctly with 'through'.
Second check: none (judge only)
Report a problem with this finding
Typo The Introduction reads "genes that contain new upstream open frames (uORFs)"; the word 'reading' is missing (the abstract has 'upstream open reading frames').
- p. 2 Introduction, last paragraph
upstream open frames (uORFs)
- p. 2 Introduction, final paragraph
new upstream open frames (uORFs)
- p. 2
identified genes that contain new upstream open frames (uORFs)
Judge's note: The p2 image shows the phrase as quoted. The abstract on p1 spells out 'upstream open reading frames (uORFs)'.
Second check: none (judge only)
Report a problem with this finding
Typo The text reads "Two of these cases affect the predict protein sequence"; it should be "predicted".
- p. 5 Intron paragraph
Two of these cases affect the predict protein sequence
- p. 5 Introns paragraph
affect the predict protein sequence
Judge's note: The p5 image shows the wording as quoted.
Second check: none (judge only)
Report a problem with this finding
Typo Misspelling: 'remarkable heterogeniety at the 3′ ends' should read 'heterogeneity'.
- p. 7 Discussion, first paragraph
heterogeniety at the 3′ ends
- p. 7 Discussion, first paragraph
remarkable heterogeniety at the 3′ ends
- p. 7 Discussion para 1
heterogeniety
Judge's note: The misspelling is visible in the p.7 page image, Discussion paragraph 1. The next line spells 'heterogeneity' correctly.
Second check: none (judge only)
Report a problem with this finding
Typo Missing verb: 'instances in which there an annotated intron but no evidence for splicing' should read 'there is an annotated intron'.
- p. 7 Discussion, intron sentence
which there an annotated intron but no evidence for splicing
- p. 7 Discussion
instances in which there an annotated intron
Judge's note: Read on the p.7 page image, Discussion paragraph 3.
Second check: none (judge only)
Report a problem with this finding
Typo Stray word: 'Pervasive occurrence of overlapping transcripts property may be a unique feature to S. cerevisiae' is ungrammatical; 'property' is left over.
- p. 7 Discussion, overlapping transcripts paragraph
Pervasive occurrence of overlapping transcripts property may be
- p. 7
Pervasive occurrence of overlapping transcripts property may be a unique feature
Judge's note: Read on the p.7 page image, Discussion paragraph 2.
Second check: none (judge only)
Report a problem with this finding
Typo Inconsistent term form: 'Thus, RNA Seq can define the presence of introns' (and 'the RNA Seq protocol', p.7) has no hyphen, whereas 'RNA-Seq' is used elsewhere.
- p. 7 Discussion
RNA Seq can define the presence of introns
Judge's note: The p.7 image shows 'RNA' at the end of one line and 'Seq' at the start of the next with no hyphen. The last line of the page has 'RNA Seq protocol' mid-line, so the missing hyphen is in the typeset text, not an extraction artefact. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo The Fig. 3 legend expands RACE as 'rapid implication of cDNA ends'. The correct expansion is 'rapid amplification of cDNA ends'.
- p. 13 Figure 3 caption end
RACE, rapid implication of cDNA ends
- p. 13 Figure 3 legend, abbreviations
rapid implication of cDNA ends
Judge's note: Visible on the p.13 image. This is a wrong word in an abbreviation expansion (standard term).
Second check: none (judge only)