Single-Cell Chromatin Accessibility Reveals Principles of Regulatory Variation
Nature, 2015 · paper · Biomedicine
Develops single-cell ATAC-seq to map accessible DNA in individual cells. Pooled single-cell maps resemble profiles from millions of cells, and variation in accessibility is linked to specific transcription factors, cell type and the three-dimensional organization of the genome.
7 findings
Substance Figure 3 legend (d) reads "Variability associated with trans-factor motifs across 7 cell types", but the panel has eight columns (H1 ESC, GM, K562, EML, TF1, mESC, HL60, BJ) and the abstract says "across 8 cell types".
- p. 19 Figure 3 legend, panel (d)
Variability associated with trans-factor motifs across 7 cell types
- p. 1 Abstract
cell-type specific accessibility variance across 8 cell types
- p. 19 Figure 3 caption (d)
Variability associated with trans-factor motifs across 7 cell
- p. 19 Figure 3 legend (d)
Variability associated with trans-factor motifs across 7 cell types.
Judge's note: Viewed page 19 image and counted eight labelled columns in the Fig. 3d heat map: H1 ESC, GM, K562, EML, TF1, mESC, HL60, BJ. The abstract (main.txt line 23) says 8 cell types. Main text lines 104-105 list V6.5 mouse ESCs, EML1, TF-1 and HL-60 alongside the three tier-1 lines, and BJ fibroblasts appear at line 214, giving eight. I found no statement that excludes or merges a cell type for panel d. The legend's count of 7 is wrong as printed.
Second check: another model agreed
Report a problem with this finding
Reference or label Front-page Supplementary Materials list reads "Extended Data Figures 1 to 7", but the manuscript contains Extended Data Figures 1 to 10 and the text cites Extended Data Figs. 8, 9 and 10.
- p. 1 Front-page footnote, Supplementary Materials list
Extended Data Figures 1 to 7
- p. 14 Extended Data section
Extended Data Figure 10. Measurements of individual peaks within single-cells
Judge's note: Page 1 image shows "Extended Data Figures 1 to 7". Extended Data Figure legends 1-10 are present on pages 6-14 (ED Fig. 10 on p14), and the main text cites ED Fig. 8a, 8b,c, 9a,b, 9c-i and 10. No statement in the files explains a count of 7.
Second check: another model agreed
Report a problem with this finding
Reference or label Text says "no substantial change in variability of GATA1 binding sites that co-occur with JUN or CEBPB (Extended Data Fig. 6i)", but panel 6i and its legend show "CJUN and/or GATA2 and CEBPB and/or GATA2".
- p. 4 Main text, paragraph 1
find no substantial change in variability of GATA1 binding sites that co-occur with JUN or CEBPB (Extended Data Fig. 6i)
- p. 11 Extended Data Figure 6 legend and panel i
(i) CJUN and/or GATA2 and CEBPB and/or GATA2 (co-)
- p. 4 Main text, first paragraph
variability of GATA1 binding sites that co-occur with JUN or CEBPB (Extended Data Fig. 6i)
- p. 11 Extended Data Figure 6 legend and panel i
(i) CJUN and/or GATA2 and CEBPB and/or GATA2 (co-) occurring ChIP-seq sites
Only the authors can say what is right here.
Judge's note: On the page 11 image, the Venn diagrams in panel i are labelled CJUN/GATA2 and CEBPB/GATA2, and the legend agrees. The page 4 text names GATA1. No GATA1/JUN or GATA1/CEBPB panel exists in the files. Either the text or the panel names the wrong factor; only the authors can say which.
Second check: another model agreed
Report a problem with this finding
Reference or label Extended Data Fig. 9 legend (a) reads "Interchromosomal chromosome 1 co-correlations of deviation scores within single cells", but the panel is a chr1-by-chr1 map titled "Intrachromosomal ..." and the main text computes co-correlation "within the same chromosome"; the legend should say "Intrachromosomal".
- p. 13 Extended Data Figure 9 legend, panel (a)
Interchromosomal chromosome 1 co-correlations of deviation scores
- p. 13 Extended Data Figure 9, panel a title
Intrachromosomal
- p. 5 Main text, paragraph 2
co-correlation of each window across all others within the same chromosome
- p. 13 Extended Data Figure 9, panel a title
Intrachromosomal devitation correlation
Judge's note: Viewed page 13 and the legend crop. Panel a has chr1 (bp) on its axis (918,679 to 249,040,706) and its embedded title begins "Intrachromosomal". The legend prints "Interchromosomal". Main text page 5 (main.txt line 224) says co-correlation of each window is taken across all others within the same chromosome. No passage explains a between-chromosome reading. The legend label is wrong; the data are unaffected.
Second check: another model agreed
Report a problem with this finding
Reference or label Extended Data Fig. 3 legend: "Recovery of typical promoters shown in (a) within single-cells" should point to panel (b), where the typical promoters are boxed.
- p. 9 Extended Data Figure 3 legend (continued)
Recovery of typical promoters shown in (a) within single-cells
- p. 8 Extended Data Figure 3 legend and panel b
(b) Accessibility across all annotated promoters in GM12878 cells. Typical promoters used for subsequent analysis are
- p. 8 Extended Data Figure 3 legend
Typical promoters used for subsequent analysis are
Judge's note: On the page 8 image, panel a is the histogram of all peaks (mean 0.10 fragments) with no promoter annotation. Panel b carries the 'Typical promoter' bracket with dotted lines, as the preceding legend sentence says. The '(a)' pointer on p9 is wrong.
Second check: another model agreed
Report a problem with this finding
Reference or label Extended Data Fig. 6g legend: "principal component (PC) analysis of heatmap from Fig. 2d", but Fig. 2d is a histogram of GATA1 deviations; the heat map of single-cell deviations is Extended Data Fig. 6e.
- p. 11 Extended Data Figure 6 legend, panel (g)
principal component (PC) analysis of heatmap from Fig. 2d
- p. 18 Figure 2 legend, panel (d)
(d) Distribution of normalized deviations from expected accessibility signal for GATA1 sites in individual cells, histogram
- p. 18 Figure 2 legend
(d) Distribution of normalized deviations from expected accessibility signal for GATA1 sites
- p. 18 Figure 2 legend (d)
Distribution of normalized deviations from expected accessibility signal for GATA1 sites in individual cells, histogram
Judge's note: On the page 18 image, Fig. 2d is a histogram with a density curve of GATA1 deviations, and Figure 2 contains no heat map of single-cell deviations. The bi-clustered heat map of K562 single-cell deviations is ED Fig. 6e on p11. The pointer is wrong and the intended target is evidently ED Fig. 6e.
Second check: another model agreed
Report a problem with this finding
Reference or label Extended Data Fig. 6j legend: "calculated changes in co-occurring ChIP-seq sites shown in Figure 2e" should refer to Figure 2g; Fig. 2e is the GATA1/GATA2 immunostaining.
- p. 11 Extended Data Figure 6 legend, panel (j)
seq sites shown in Figure 2e
- p. 18 Figure 2 legend, panels (e) and (g)
(e) Immunostaining of GATA1 (green) and GATA2 (red)
- p. 11 Extended Data Figure 6 legend, panel (j)
calculated changes in co-occurring ChIP-seq sites shown in Figure 2e
- p. 11 Extended Data Figure 6 legend (j)
co-occurring ChIP-seq sites shown in Figure 2e
Judge's note: On the page 18 image, Fig. 2e is the immunostaining panel and Fig. 2g is the factor-by-factor triangular synergy matrix, with the same factor layout as ED Fig. 6j on p11. The main text on p4 pairs 'Fig. 2g and Extended Data Fig. 6j'.
Second check: another model agreed
Report a problem with this finding
8 minor (typos and the like; most checked by the judge only)
Reference or label References 27 (Calo & Wysocka) and 28 (Mikkelsen et al.) are in the reference list but are not cited anywhere in the article; the call-outs go from "histone modifications26" to "chromatin domains29".
- p. 5 Main text
Specific chromatin states and histone modifications26
- p. 16 Reference list
27. Calo E, Wysocka J. Modification of Enhancer Chromatin
Only the authors can say what is right here.
Judge's note: I read the full text and legends: citations 1-26, 29 and 30 appear, and 27 and 28 do not. The ED Fig. 8 legend cites only Ernst et al.26. The reference list on p16 contains entries 27 and 28, both on chromatin modification, so the likely place is 'histone modifications26-28'. The supplementary methods are not among the files, so a citation there cannot be excluded.
Second check: another model agreed
Report a problem with this finding
Typo Author initials differ on the same page: "J.D.B., B.W., M.G., and D.R. developed the Fluidigm C1 microfluidic protocols" versus "D.R. and M.L.G. declare competing financial interests".
- p. 1 Author Contributions
J.D.B., B.W., M.G., and D.R. developed the Fluidigm C1 microfluidic protocols
- p. 1 Competing Interest
D.R. and M.L.G. declare competing financial interests
Judge's note: Seen on the page 1 image. The byline gives Michael L. Gonzales and no other author has initials M.G., so the person is unambiguous; only the abbreviation is inconsistent.
Second check: none (judge only)
Report a problem with this finding
Typo Affiliation 4 reads "Stanford University, Stanford, CA 94025, USA" while affiliations 1 and 2 read "Stanford, CA 94305, USA".
- p. 1 Affiliation 4
Stanford University, Stanford, CA 94025, USA
- p. 1 Affiliation 1
Stanford, CA 94305, USA
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: which ZIP code is correct for Stanford, CA (and whether 94025 belongs to a different city) is not in the files. The page 1 image confirms that the same city carries two different postal codes in affiliations 1/2 versus 4, which looks like a digit transposition.
Second check: none (judge only)
Report a problem with this finding
Typo The supplement is named inconsistently: "Supplemental Discussion" (p2), "Supplementary Discussion" (p6), and "see Supplemental Methods section 3.2" (Fig. 2 legend) versus the front-page list "Materials and Methods".
- p. 2 Main text
Extended Data Fig. 1 and Supplemental Discussion
- p. 6 Last paragraph
(Supplementary Discussion)
- p. 18 Figure 2 legend
see Supplemental Methods section 3.2
Judge's note: All three wordings were checked (p2 in the text layer; p6 and p18 on the page images). This is a naming and style inconsistency only; the target of each pointer is clear.
Second check: none (judge only)
Report a problem with this finding
Typo Misspelling: "Conversely, co-occurance with other factors" should read "co-occurrence".
- p. 5 Main text, final paragraph
Conversely, co-occurance with other factors
- p. 5 Discussion paragraph
co-occurance with other factors
- p. 5 Main text
histone modifications26 are also sometimes
Judge's note: The page 5 image shows "co-occurance"; it is not an extraction artefact. The point in one reviewer note about uncited references 27 and 28 is handled under C009.
Second check: none (judge only)
Report a problem with this finding
Typo Grammar: "these data provide exciting new hypothesis of regulatory mechanisms" should read "hypotheses" (or "an exciting new hypothesis").
- p. 6 Main text, top of page
these data provide exciting new hypothesis of regulatory mechanisms that give rise
- p. 6 Discussion, first lines of page
these data provide exciting new hypothesis of regulatory mechanisms
- p. 6 first paragraph
provide exciting new hypothesis of regulatory mechanisms
Judge's note: The page 6 image shows the singular 'hypothesis' with no article.
Second check: none (judge only)
Report a problem with this finding
Typo Extended Data Fig. 9a panel title reads "Intrachromosomal devitation correlation"; "devitation" should be "deviation".
- p. 13 Extended Data Figure 9, panel a title
Intrachromosomal devitation correlation
- p. 13 Extended Data Fig. 9a title
devitation correlation
Judge's note: Viewed page 13 image. The title embedded in panel a reads "Intrachromosomal devitation correlation", with an extra 't'. The lettering is small at this resolution, but the extra character is legible. The text is in the figure artwork, so it is absent from the text layer. Cosmetic misspelling.
Second check: none (judge only)
Report a problem with this finding
Typo Figure 1 legend labels panel c with a capital: "(C) Genome-wide accessibility patterns observed by scATAC-seq", while the other labels are "(a)", "(b)", "(d)" and the panel itself is marked "c".
- p. 17 Figure 1 legend
(C) Genome-wide accessibility patterns observed by scATAC-
- p. 17 Figure 1 legend
(C) Genome-wide accessibility patterns
Judge's note: Viewed page 17 image. The legend prints bold (a), (b), (C), (d); the figure panel is lettered lower-case c. Cosmetic inconsistency only.
Second check: none (judge only)