A Common Variant in the FTO Gene is Associated with Body Mass Index and Predisposes to Childhood and Adult Obesity
2007 · paper · Biomedicine
Reports that a common variant in the FTO gene, found in a genome-wide search for type 2 diabetes genes, is associated with body mass index. It is described as predisposing to diabetes through BMI and to childhood and adult obesity.
3 findings
Reference or label "ENPP1 (5,8,9) and INSIG2 (9–12)" cites reference 9 ("Herbert A, et al. Science 2006;312:279") for ENPP1 as well as for INSIG2; reference 9 is the INSIG2 report.
- p. 2 Introduction, second paragraph
ENPP1 (5,8,9) and INSIG2 (9–12)
- p. 5 Reference 9
9. Herbert A, et al. Science 2006;312:279.
- p. 5 Reference 9
Herbert A, et al. Science 2006;312:279.
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that Herbert et al., Science 2006;312:279 is the report of a common variant near INSIG2 and not a study of ENPP1. On the page (p2) the sentence reads as quoted and the reference list (p5) gives reference 9 as Herbert et al. References 10–12 are three Science 2007;315:179b–d items, consistent with comments on that paper, which supports 9 belonging to the INSIG2 group only. The list has no titles, so the intended ENPP1 references can only be confirmed by the authors.
Second check: another model agreed
Report a problem with this finding
Reference or label Page 5: "We used reverse transcription PCR to assess the expression of FTO and KIAA1005 in a human tissue panel (18)" cites reference 18 ("Peters T, Ausmeier K, Dildrop R, Rüther U. Mamm Genome 2002;13:186"), which the preceding paragraph cites twice for the mouse fused-toe deletion; note 13 (materials and methods) may be intended.
- p. 5 Last text paragraph
in a human tissue panel (18)
- p. 6 References and Notes, entry 18
18. Peters T, Ausmeier K, Dildrop R, Rüther U. Mamm Genome 2002;13:186.
- p. 6 References and Notes, entry 13
13. Materials and methods are available as supporting material on Science Online.
- p. 5 Discussion, third paragraph, first sentence
tissue panel (18)
Only the authors can say what is right here.
Judge's note: Read page 5 and the reference list on page 6. The sentence describes the authors' own RT-PCR on a human tissue panel and cites (18). Reference 18 is a 2002 Mammalian Genome paper, cited twice in the paragraph above for the 1.6-Mb mouse deletion and the genes deleted in Ft mice. The paper's pointer for its own methods is note 13 ("Materials and methods are available as supporting material on Science Online"), used that way on page 3. The reference list gives no titles, so I cannot exclude that reference 18 is cited for primers or a panel method; only the authors can say which number is meant.
Second check: another model agreed
Report a problem with this finding
Wording Page 5: "We conclude that the 47-kb intron within the FTO gene is most likely to contain the predisposing variant(s)", although the same paragraph defines it as "a 47-kb region that encompasses parts of the first two introns as well as exon 2 of FTO".
- p. 5 First paragraph
a 47-kb region that encompasses parts of the first two introns as well
- p. 5 First paragraph, conclusion
We conclude that the 47-kb intron within the FTO gene
- p. 5 Discussion, first paragraph, concluding sentence
We conclude that the 47-kb intron within the FTO gene is most likely to contain the
- p. 5 Discussion, first paragraph, second sentence
additional SNPs within a 47-kb region that encompasses parts of the first two introns as well
Judge's note: Read the whole paragraph on the page 5 image. The 47-kb stretch is introduced as a region covering parts of the first two introns and exon 2, then called "the 47-kb region", "the 47-kb associated region" and, in the next paragraph but one, "the 47-kb interval". Only the concluding sentence calls it "the 47-kb intron". By the paper's own description it is not an intron, and no 47-kb intron is described anywhere else in the article. "Region" or "interval" is the evident fix.
Second check: another model agreed
Report a problem with this finding
13 minor (typos and the like; most checked by the judge only)
Reference or label "GAD2 (4–7), ENPP1 (5,8,9)": the GAD2 range 4–7 includes reference 5 ("Meyre D, et al. Nat Genet 2005;37:863"), which is also cited for ENPP1 and is the ENPP1 report.
- p. 2 Introduction, second paragraph
GAD2 (4–7), ENPP1 (5,8,9)
- p. 5 Reference 5
5. Meyre D, et al. Nat Genet 2005;37:863.
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that Meyre et al., Nat Genet 2005;37:863 is the ENPP1 obesity association paper and not a GAD2 study. The page (p2) and reference list (p5) read as quoted; reference 5 is cited under both genes in one sentence. The list carries no titles, so whether the range should read "4,6,7" or a different Meyre paper was intended is for the authors. Related to C004 (the same sentence's citation numbering) but a different reference.
Second check: another model agreed
Report a problem with this finding
Wording The abstract's "replicated in 13 cohorts with 38,759 participants" cannot be reconstructed from the article: Table 1 sums to the "30,081 participants" in the text, the children are "10,172", and the 13 replication cohorts (all but the initial WTCCC cases, N 1913) sum to 38,340.
- p. 2 Abstract
replicated in 13 cohorts with 38,759 participants
- p. 3 Results, paragraph 3
a total of 30,081 participants
- p. 3 Results, paragraph 2
10,172 white European children from two studies
- p. 3 Second paragraph
19,424 white European adults from seven general population-based
Only the authors can say what is right here.
Judge's note: Read the abstract (p2), Results (p3), Table 1 (p9 upright) and Table 2 in the text layer (p10 image not opened). Table 1 N values sum to 30,081 and the population studies to 19,424, both as printed. The 10,172 children equal ALSPAC age 7 (5969) plus NFBC1966 age 14 (4203). Dropping the initial WTCCC cases leaves 13 cohorts with 38,340 participants, 419 short of 38,759. No statement in the article explains the figure. Table S1 is cited for the cohorts but the supplement is not among the files, so it may hold different counts; hence a query, not a correction. The third reviewer note (10,172 vs 7477+4320) is not a defect: those are the birth-weight Ns in Table 2, a different analysis set from the BMI Ns.
Second check: another model agreed
Report a problem with this finding
Wording The genome-wide scan is described as covering "490,032 autosomal single-nucleotide polymorphisms", but the Bonferroni correction is for "the number of tests performed in the initial genome-wide scan (∼400,000)"; the article does not explain the difference.
- p. 2 Introduction, third paragraph
490,032 autosomal
- p. 3 Results, paragraph 3
(∼400,000), the association remained significant (P = 1.2 × 10−29)
- p. 3 Results, Bonferroni
number of tests performed in the initial genome-wide scan (∼400,000)
Judge's note: Both passages read on the page images (p2, p3). The corrected P value follows the 400,000 figure: 3 × 10−35 × 400,000 = 1.2 × 10−29, as printed. The article gives no reason why fewer tests than SNPs were counted (for example a post-quality-control SNP count). The "∼" marks the figure as approximate and a quality-control explanation is plausible, so this is a minor unexplained inconsistency; the supplement is not among the files.
Second check: another model agreed
Report a problem with this finding
Wording "All BMI-associated SNPs (P ranging from 1 × 10−4 to 1 × 10−5)" excludes the P value printed for the representative SNP rs9939609 in the same analysis ("8 × 10−6" in Table 1 and on p4), and Fig. 1B plots several SNPs above −log10 P = 5.
- p. 2 Introduction, last line of page, continuing on page 3
All BMI-associated SNPs (P ranging from 1
- p. 3 First line
× 10−4 to 1 × 10−5) are highly correlated with each other
- p. 9 Table 1, row UK cases (WTCCC), column P
8 × 10−6
- p. 4 Population stratification paragraph
8 × 10−6 unadjusted
Only the authors can say what is right here.
Judge's note: Read the sentence across p2 and p3, Table 1 (p9 upright: UK cases (WTCCC), P = 8 × 10−6), p4 ("P = 9 × 10−6 adjusted; 8 × 10−6 unadjusted") and Fig. 1B (p7), where the top cluster points lie between about 5.0 and 5.3 on the −log10 P axis, i.e. below 1 × 10−5. The stated range therefore does not cover the strongest associations. The range may be an order-of-magnitude approximation, so this is a wording-level inconsistency and a query for the authors.
Second check: another model agreed
Report a problem with this finding
Wording "the association of genetic variation at FTO with BMI results from longitudinal changes in fat mass", although fat mass was measured by DEXA at one age only ("available at age 9").
- p. 4 Results, fat mass paragraph, final sentence
results from longitudinal changes in fat mass
- p. 4 same paragraph
dual-energy x-ray absorptiometry (DEXA)–derived measures of fat mass and lean mass were available at age 9
Judge's note: Read on the p4 image. The same paragraph says DEXA measures of fat and lean mass were available at age 9, and the Fig. 2 legend (text layer, p8) describes panels C and D as 9-year-old children. No repeated fat-mass measurement is reported, so "longitudinal" is not supported for fat mass as worded; BMI was followed across ages 7 to 11, which may be what was meant. A wording point.
Second check: another model agreed
Report a problem with this finding
Wording Table 2 caption: "P values represent the change in log BMI per A allele. BMI presented as geometric means and back-transformed 95% confidence intervals." The table also gives P values (0.21, 0.42) and means (e.g. "3438 (3422, 3455)") for birth weight, for which the caption defines neither the P value, the kind of mean, nor the unit.
- p. 10 Table 2 caption
P values represent the change in log BMI per A allele. BMI presented as geometric means
- p. 10 Table 2, Birth rows
3438 (3422, 3455)
- p. 10 Table 2 caption
P values represent the change in log BMI per A allele.
Judge's note: Read the upright page 10 image. The caption's first sentence says the table covers BMI and birth weight, and the last two rows under "Birth" hold values around 3400 to 3500 under the heading "Mean trait value (95% CI) by genotype". The caption's description of the P values and of the means refers to BMI only, and no unit is given for the birth rows (grams can be inferred). The footnotes add exclusions for the birth-weight analysis but do not define its P values. A labelling gap in the caption, not a numerical error.
Second check: another model agreed
Report a problem with this finding
Typo Unbalanced parentheses in three statistics on p4: "95% CI = 0.07 to 0.11); P = 4 × 10−17; ∼1.2 kg in adults)", "P = 6 × 10−10), equivalent to a 14% difference across the three genotype groups)" and "P = 0.03), which is equivalent to a 1% increase across the three genotype groups)".
- p. 4 Second paragraph (weight)
95% CI = 0.07 to 0.11); P = 4 × 10−17; ∼1.2 kg in adults)
- p. 4 Second paragraph (fat mass)
P = 6 × 10−10), equivalent to a 14% difference across the three genotype groups) (Fig.
- p. 4 Second paragraph (lean mass)
P = 0.03), which is equivalent to a 1% increase across the three
- p. 2 Abstract
weighed about 3 kilograms more
Judge's note: All three extra closing parentheses are printed on the p4 page image; each passage has one opening and two closing parentheses. Cosmetic. The second reviewer note in this cluster (abstract "about 3 kilograms" for homozygotes vs "∼1.2 kg" per allele, i.e. 2 × 1.2 = 2.4 kg) is not confirmed: the abstract figure compares AA with TT directly, the per-allele figure is an approximate conversion from a pooled Z-score, and Table 1 shows AA groups above the additive line in several cohorts. The weight tables S3B/S3C that would settle it are not among the files.
Second check: none (judge only)
Report a problem with this finding
Typo Stray closing square brackets with no opening bracket: "95% CI = 1.14 to 1.23], P = 2 × 10−17" and "95% CI = 1.26 to 1.52]; P = 4 × 10−11"; in the same sentence "95% CI 1.26 to 1.39" lacks the "=" used elsewhere.
- p. 3 Results, overweight/obesity paragraph
95% CI = 1.14 to 1.23], P = 2 × 10−17
- p. 3 Results, overweight/obesity paragraph
95% CI = 1.26 to 1.52]; P = 4 × 10−11
- p. 4 Results, weight
95% CI = 0.07 to 0.11); P = 4 × 10−17
Judge's note: Both brackets are printed on the p3 page image, so this is not an extraction artefact. Cosmetic; the numbers are unaffected. The unbalanced round parenthesis on p4 ("0.07 to 0.11); P = 4 × 10−17") is handled under C011.
Second check: none (judge only)
Report a problem with this finding
Typo Inconsistent interval format: "OR per-A allele at age 11 years = 1.35; CI = 1.14, 1.61; P = 6 × 10−4" omits "95%" and uses a comma where the rest of the text uses "95% CI = x to y"; the next interval ("CI = 1.16 to 1.39") also omits "95%".
- p. 4 Results, ALSPAC paragraph
OR per-A allele at age 11 years = 1.35; CI = 1.14, 1.61; P = 6 × 10−4
Judge's note: Seen on the p4 image as quoted. Style inconsistency only; the values are unaffected and the meaning is clear from context.
Second check: none (judge only)
Report a problem with this finding
Typo Page 4 prints "P= 2 × 10−7" with no space before the equals sign; every other P value in the paragraph is set "P = ...".
- p. 4 Results, ALSPAC paragraph
P= 2 × 10−7
Judge's note: Checked on the crop: "CI = 1.16 to 1.39; P= 2 × 10^-7" has no space between P and =, while the neighbouring values (P = 3 × 10^-5, P = 7 × 10^-9, P = 6 × 10^-4, P = 0.04) do. Cosmetic spacing slip only; the value itself is not in question.
Second check: none (judge only)
Report a problem with this finding
Typo Page 5, final sentence: "Further work with both knockout and overexpression models of FTO and KIAA1005 are likely to provide the most fruitful approach"; the subject "Further work" is singular, so the verb should be "is likely".
- p. 5 Discussion, final sentence
overexpression models of FTO and KIAA1005 are likely to provide
Judge's note: Read the closing sentence on the page 5 image. The verb agrees with the nearer plural noun "models" instead of the singular subject "Further work". Grammatical slip; the meaning is unaffected.
Second check: none (judge only)
Report a problem with this finding
Typo Fig. 1 legend (page 7) prints "Associations of SNPs in the FTO/KIA1005 region of chromosome 16"; the gene is KIAA1005 in panel D of the same figure and throughout the text.
- p. 7 Fig. 1 legend
FTO/KIA1005 region
- p. 5 Fig. 1 panel D label
KIAA1005
- p. 7 Fig. 1 legend
FTO/KIA1005 region of chromosome 16
- p. 7 Fig. 1 caption
FTO/KIA1005
Judge's note: The page 7 image shows "FTO/KIA1005" in the legend and "KIAA1005" as the label in panel D directly above it. Page 5 uses KIAA1005 six times. The legend drops an A; not an extraction artefact.
Second check: none (judge only)
Report a problem with this finding
Typo Fig. 2 legend (page 8) ends "Error bars represent 95% confidence intervals" with no full stop.
- p. 8 Fig. 2 legend
Error bars represent 95% confidence intervals
Judge's note: The page 8 image shows the legend ending at "confidence intervals" with no terminal punctuation; the Fig. 1 legend on page 7 ends with a full stop ("Gene positions."). Purely cosmetic.
Second check: none (judge only)