The DOI matches its Crossref record.
Estimating coverage in metagenomic data sets and why it matters
Open accessAbstract
A ‘metagenome’ is the theoretical collection of genomes from all members of a given microbial community, and a ‘metagenomic data set’ is the subset captured in a given sequencing event. Although these terms are often used interchangeably and metagenomic data sets are regularly called metagenomes by synecdoche, their relationship is analogous to sample and population in statistics. The fraction of the metagenome represented in the metagenomic data set, termed coverage (not to be confused with the repetition of features, termed sequencing depth), is of key importance in assessing statistical significance of features sampled (taxa, genes and so on). However, quantitative computational methods to assess the level of coverage are limited, a problem we have recently attempted to solve. In extreme cases, where small data sets are used to characterize complex communities, misleading inferences can arise. For instance, random variation can be frequently mistaken for real differences in comparisons of metagenomic data sets with extreme differences in coverage. Further, insufficient coverage also reduces the detection limits and statistical power of the comparisons, hiding real, ecologically relevant trends and differences ( Figure 1 ). We demonstrate here how available solutions can determine the level of sequencing coverage obtained by metagenomic data sets and thus, guide their robust analysis and comparison. Figure 1 Effect of average coverage on detection of differentially abundant features. The abundance of nonredundant genes (assembled and clustered at 98% amino-acid identity) detected in the metagenomes of Lake Lanier (Atlanta, GA, USA; Sequence Read Archive Projects SRP028408, SRP005437-9; abundance estimates were based on read-mapping at 95% nucleotide identity) was compared between three summer and two winter samples, at different levels of subsampling (0.01–50% of the total data set) and the coverage was computed using Nonpareil (coverage axes). The main panel (bottom-left) shows the number of detected genes, represented by the color of the circles (see legend). The values between subsamples were estimated using bicubic interpolation. Note that the detection of genes is more strongly affected by the coverage in summer data sets owing to lower gene richness in the winter data sets. The additional panels correspond to the comparisons of the subsamples against the complete (not subsampled) data sets, which showed 64% and 75% coverage for winter and summer, respectively. The comparison between complete winter and summer data sets (top-right circle in main panel) was used as a reference for the definition of true/false positives (TP/FP) and true/false negatives (TN/FN). Sensitivity was defined as TP/(TP+FN), specificity as TN/(FP+TN) and accuracy of the test as (TP+TN)/(TP+TN+FN+FN). Sensitivity, specificity and accuracy were interpolated using cubic splines with smoothing parameter 0.6. Differential abundance was defined as adjusted P -value ⩽ 0.1 in the negative binomial test implemented in DESeq ( Anders and Huber, 2010 ). Note that sensitivity drops rapidly when coverage of any (or both) of the collections of data sets decreases, while specificity is typically high, except at extreme differences in coverage. In general, the accuracy was compromised (<90%) in data sets with >twofold difference of coverage. Full size image
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.
Integrity signals
Registry
Cites
Statistics
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.
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
Cited by
Normalization and microbial differential abundance strategies depend upon data characteristics
Weiss, S., Xu, Z. Z., Peddada, S., Amir, A., Bittinger, K., Gonzalez, A., Lozupone, C., Zaneveld, J. R., Vázquez-Baeza, Y., Birmingham, A., et al. (2017) · Microbiome · doi:10.1186/s40168-017-0237-y · Cited by 2,051
Diet shapes the gut microbiome of pigs during nursing and weaning
Frese, S. A., Parker, K., Calvert, C. C., & Mills, D. A. (2015) · Microbiome · doi:10.1186/s40168-015-0091-8 · Cited by 514
Evaluating the Information Content of Shallow Shotgun Metagenomics
Hillmann, B., Al-Ghalith, G. A., Shields-Cutler, R. R., Zhu, Q., Gohl, D. M., Beckman, K. B., Knight, R., & Knights, D. (2018) · mSystems · doi:10.1128/msystems.00069-18 · Cited by 474
“Another study explored the use of databasefree methods for metagenomic analysis and found that approximately 1 Gb of pairs would provide sufficient metagenomic coverage to compare metagenomic samples with such methods (15).”
BackgroundThe initial state of the human gut microbiome determines its reshaping by antibiotics
Raymond, F., Ouameur, A. A., Déraspe, M., Iqbal, N., Gingras, H., Dridi, B., Leprohon, P., Plante, P.-L., Giroux, R., Bérubé, È., et al. (2016) · The ISME Journal · doi:10.1038/ismej.2015.148 · Cited by 358
Nonpareil 3: Fast Estimation of Metagenomic Coverage and Sequence Diversity
Rodriguez-R, L. M., Gunturu, S., Tiedje, J. M., Cole, J. R., & Konstantinidis, K. T. (2018) · mSystems · doi:10.1128/msystems.00039-18 · Cited by 266
Self-citationRecovering complete and draft population genomes from metagenome datasets
Sangwan, N., Xia, F., & Gilbert, J. A. (2016) · Microbiome · doi:10.1186/s40168-016-0154-5 · Cited by 260
The balance of metagenomic elements shapes the skin microbiome in acne and health
Barnard, E., Shi, B., Kang, D., Craft, N., & Li, H. (2016) · Scientific Reports · doi:10.1038/srep39491 · Cited by 259
The influence of alfalfa‐switchgrass intercropping on microbial community structure and function
Cha, G., Meinhardt, K. A., Orellana, L. H., Hatt, J. K., Pannu, M. W., Stahl, D. A., & Konstantinidis, K. T. (2021) · Environmental Microbiology · doi:10.1111/1462-2920.15785 · Cited by 7
Self-citation
Metadata
- Authors
- Rodriguez-R, L. M., & Konstantinidis, K. T.
- Venue
- The ISME Journal
- Publisher
- Oxford University Press (OUP)
- Published
- 2014-05-13
- Volume
- 8
- Issue
- 11
- Pages
- 2349-2351
- ISSN
- 1751-7362 (ISSN-L)
- 1751-7370
- OpenAlex
- W2095498620
- Cited by
- 179
- References
- 0
- FWCI
- 4.14
- Type
- journal article
- Language
- en
- Open access
- bronze
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
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.