Metagenomic Analysis of the Human Distal Gut Microbiome

Steven R. Gill, Mihai Pop, Robert T. DeBoy, Paul B. Eckburg, Peter J. Turnbaugh, Buck S. Samuel, Jeffrey I. Gordon, David A. Relman, Claire M. Fraser-Liggett, Karen E. Nelson

2006 · paper · Biomedicine

Analyzes about 78 million base pairs of DNA and 2,062 16S rDNA sequences from the fecal samples of two healthy adults. Compared with average sequenced microbial genomes, the gut microbiome is enriched in metabolism of glycans, amino acids and xenobiotics, methanogenesis, and vitamin and isoprenoid biosynthesis.

2 findings

Reference or label The text cites "Baker et al. (10)", but reference 10 is "Tyson GW, et al. Nature 2004;428:25." There is no Baker entry in the reference list.

p. 2 Sequencing the microbiome, para 1
recent reports from Venter et al. (9) and Baker et al. (10) have
p. 6 References, entry 10
10. Tyson GW, et al. Nature 2004;428:25.
p. 2 Sequencing the microbiome, para 1
Venter et al. (9) and Baker et al. (10)
p. 6 Reference 10
Nature 2004;428:25

Judge's note: Both crops confirm it: p2 prints 'Baker et al. (10)' and p6 lists entry 10 as Tyson GW et al. The in-text first-author name does not match the cited entry, whereas 'Venter et al. (9)' matches ref 9. A separate reviewer point says the page should be 37, not 25. That rests on outside knowledge (the real page range of Tyson et al., Nature 428), cannot be checked from the files, and is not part of this confirmation.

Second check: another model agreed

Report a problem with this finding

Reference or label "Both human subjects showed similar patterns of enrichment for each COG (Fig. 2) and KEGG (Fig. 3) category", but Fig. 3 shows pooled data ("both samples compared with…"), with colours encoding the reference sets rather than the subjects.

p. 4 The metabolome section, para 1
Both human subjects showed similar patterns of enrichment for each COG (Fig. 2) and KEGG (Fig. 3) category
p. 11 Fig. 3 legend
both samples compared with all sequenced bacterial genomes in KEGG (blue)
p. 11 Fig. 3 legend
both samples compared with all sequenced bacterial genomes in KEGG (blue), the human genome (red), and all sequenced archaeal genomes in KEGG (yellow)

Only the authors can say what is right here.

Judge's note: In the Fig. 3 image and legend (p11), each KEGG category has one bar per reference set (blue bacteria, red human, yellow archaea), with no split by subject. Fig. 2 does split by subject (black/grey). Fig. 3 therefore cannot support the per-subject KEGG claim. The per-subject data may be in the supplement, which is not available here.

Second check: another model agreed

Report a problem with this finding

11 minor (typos and the like; most checked by the judge only)

Wording Abstract says the microbiome has "significantly enriched metabolism of glycans, amino acids, and xenobiotics", but in Fig. 3 all three 'Biodegradation of Xenobiotics' bars are below 1 and starred, which the legend defines as underrepresentation (odds ratio < 1, P < 0.05).

p. 1 Abstract
enriched metabolism of glycans, amino acids, and xenobiotics
p. 11 Fig. 3, 'Biodegradation of Xenobiotics' bars
Asterisks indicate enrichment (odds ratio > 1, P < 0.05) or underrepresentation (odds ratio < 1, P < 0.05)
p. 1 Abstract
significantly enriched metabolism of glycans, amino acids, and xenobiotics
p. 11 Fig. 3, 'Biodegradation of Xenobiotics' bars
Biodegradation of Xenobiotics

Only the authors can say what is right here.

Judge's note: On the Fig. 3 image (p11), the xenobiotics bars are about 0.3 (blue, vs bacteria), 0.8 (red, vs human) and 0.3 (yellow, vs archaea), each with an asterisk. The legend says an asterisk with odds ratio < 1 means significant underrepresentation. The main text (p6) supports xenobiotic enrichment only through specific COGs (β-glucosidase, β-glucuronidase) and named KEGG pathways in tables S3/S4, which are not among the files. The paper does not reconcile these with the category-level figure. Only the authors can say what the abstract claim rests on.

Second check: another model agreed

Report a problem with this finding

Wording Abstract: "we compared our human genome with the average content of previously sequenced microbial genomes". What the analysis actually compares is the gut microbiome, against bacterial and archaeal genomes and the human genome.

p. 1 Abstract
we compared our human genome with the average content of previously sequenced microbial genomes
p. 11 Fig. 3 y-axis
human gut microbiome/reference genomes

Only the authors can say what is right here.

Judge's note: The abstract image confirms the wording. The results (pp4-5) and Figs. 2-3 compare the microbiome with reference sets. The Fig. 3 y-axis reads 'human gut microbiome/reference genomes', and the next abstract sentence begins 'Our microbiome'. The paper never defines 'our human genome' to include the microbiome, so the sentence misdescribes the comparison, unless the authors meant the superorganism sense.

Second check: another model agreed

Report a problem with this finding

Wording The text gives B. longum read identity as "(80 to 100% identity)", but Fig. 1A shows a few points below 80% (about 77-78% near 0.15, 1.58 and 1.9 Mb, and an isolated point near 72%).

p. 3 para 2
genome sequence (80 to 100% identity)
p. 8 Fig. 1A
Percent identity

Judge's note: In the Fig. 1A image (p8), a handful of small points fall below the 80% gridline. The range in the text is approximate. The discrepancy is small and visual, so this is an observation.

Second check: another model agreed

Report a problem with this finding

Wording The PCR library is said to show "a similar phylogenetic distribution among higher-order taxa", yet Actinobacteria are 10 of 72 phylotypes in the shotgun set but only 1 of 151 in the PCR set ("150 Firmicutes; 1 Actinobacteria").

p. 3 Identifying phylotypes
revealed a similar phylogenetic distribution among higher-order taxa
p. 3 Identifying phylotypes
(23% novel; 150 Firmicutes; 1 Actinobacteria)

Judge's note: From the p3 text: shotgun set 62 Firmicutes / 10 Actinobacteria phylotypes, PCR set 150/1. 'Similar' holds in the sense that both sets contain only Firmicutes and Actinobacteria and are dominated by Firmicutes, but the Actinobacteria share differs about 20-fold. This is a wording point.

Second check: another model agreed

Report a problem with this finding

Wording The Fig. 1 title reads "Comparison of random metagenome reads with completed genome of Bifidobacterium longum and Methanobrevibacter smithii", but panel B and the text use "the draft genome of M. smithii".

p. 8 Fig. 1 legend, first sentence
Comparison of random metagenome reads with completed genome of Bifidobacterium longum and Methanobrevibacter smithii
p. 8 Fig. 1 legend, panel B
along the draft genome of M. smithii
p. 3 para 3
a deep draft assembly of this
p. 8 Fig. 1 caption, first sentence
Comparison of random metagenome reads with completed genome of Bifidobacterium longum and Methanobrevibacter smithii.

Judge's note: The p8 legend says panel B uses a pseudomolecule of concatenated contigs from the M. smithii draft assembly, and p3 says 'this draft assembly' (ref 13). The word 'completed' fits only B. longum.

Second check: another model agreed

Report a problem with this finding

Wording The peptidoglycan odds ratio against all sequenced bacteria is described as "similar to all sequenced bacteria (1.83)", although the same paragraph defines an odds ratio of one as the same proportion and above one as enrichment. In the same sentence, 7.06 is called only "moderately enriched".

p. 4 Comparative functional analysis, end of para 1
similar to all sequenced bacteria (1.83), and moderately enriched relative to all sequenced Archaea (7.06)

Only the authors can say what is right here.

Judge's note: p4 text. 'Similar' is relative to 77.88 and to the expectation that peptidoglycan genes are bacterial, but under the paper's own definition 1.83 is enrichment. No P value or interval is given, and table S3 is not among the files. This is a wording point for the authors.

Second check: another model agreed

Report a problem with this finding

Wording "microbial β-glucosidases, β-rhamnosidases, and esterases (33)": dietary phenolic rhamnosides are usually cleaved by α-L-rhamnosidases.

p. 6 para 2
microbial β-glucosidases, β-rhamnosidases, and esterases (33)

Only the authors can say what is right here.

Judge's note: Rests on outside knowledge: that rhamnose in dietary flavonoid glycosides is α-L-linked and hydrolysed by α-L-rhamnosidases. The crop confirms 'β-rhamnosidases' in print. Ref 33 (Schneider et al.) cannot be checked here. The authors should confirm the intended enzyme.

Second check: another model agreed

Report a problem with this finding

Wording Fig. 1 panels A and B plot red and blue points, but neither the legend (pp8-9) nor a key in the figure defines the colours.

p. 8 Fig. 1 legend / panels A and B
Percent identity plot (PIP) of alignments of shotgun reads along the genome of B. longum strain NCC2705

Only the authors can say what is right here.

Judge's note: The p8 image shows red and blue points in both panels. The legend text on pp8-9 mentions subjects 7 and 8 but never assigns colours. Presumably the colours are the subjects, but the paper does not say so.

Second check: another model agreed

Report a problem with this finding

Typo Missing article: "Therefore, superorganismal view of our genetic landscape" should read "Therefore, a superorganismal view…".

p. 2 para 1
Therefore, superorganismal view of our genetic landscape

Judge's note: The p2 image confirms the printed text, so this is not an extraction loss.

Second check: none (judge only)

Report a problem with this finding

Typo Misspelling: "isopenteryl pyrophosphate (IPP)" should be "isopentenyl pyrophosphate".

p. 5 last para
isopenteryl pyrophosphate (IPP)

Judge's note: The p5 crop confirms 'isopenteryl' in print.

Second check: none (judge only)

Report a problem with this finding

Typo Misspelling: "the mevanolate pathway for IPP" should be "mevalonate pathway".