Critical Assessment of Metagenome Interpretation—a benchmark of metagenomics software

作者：Alexander Sczyrba, Peter Hofmann, Peter Belmann, David Koslicki, Stefan Janssen, Johannes Dröge, Ivan Gregor, Stephan Majda, Jessika Fiedler, Eik Dahms, Andreas Bremges, Adrian Fritz, Rubén Garrido‐Oter, Tue Sparholt Jørgensen, Nicole Shapiro, Philip D. Blood, Alexey A. Gurevich, Yang Bai, Dmitrij Turaev, Matthew Z. DeMaere, Rayan Chikhi, Niranjan Nagarajan, Christopher Quince, Fernando Meyer, Monika Balvočiūtė, Lars Hestbjerg Hansen, Søren Johannes Sørensen, Burton Kuan Hui Chia, Bertrand Denis, Jeff L. Froula, Zhong Wang, Robert W. Egan, Dongwan Don Kang, Jeffrey J. Cook, Charles Deltel, Michael Beckstette, Claire Lemaitre, Pierre Peterlongo, Guillaume Rizk, Dominique D. Lavenier, Yu‐Wei Wu, Steven W. Singer, Chirag Jain, Marc Strous, Heiner Klingenberg, Peter Meinicke, Michael D. Barton, Thomas Lingner, Hsin-Hung Lin, Yu-Chieh Liao, Genivaldo Gueiros Z. Silva, Daniel Cuevas, Robert Alan Edwards, Surya Saha, Vitor C. Piro, Bernhard Y. Renard, Mihai Pop, Hans‐Peter Klenk, Markus Göker, Nikos C. Kyrpides, Tanja Woyke, Julia A. Vorholt, Paul Schulze‐Lefert, Edward M. Rubin, Aaron E. Darling, Thomas Rattei, Alice Carolyn McHardy · 发表于：Nature Methods · 年份：2017 · DOI：10.1038/nmeth.4458 · 被引用次数：983 · 研究领域：Gut microbiota and health、Metabolomics and Mass Spectrometry Studies、Microbial Community Ecology and Physiology

The Critical Assessment of Metagenome Interpretation (CAMI) community initiative presents results from its first challenge, a rigorous benchmarking of software for metagenome assembly, binning and taxonomic profiling. Methods for assembly, taxonomic profiling and binning are key to interpreting metagenome data, but a lack of consensus about benchmarking complicates performance assessment. The Critical Assessment of Metagenome Interpretation (CAMI) challenge has engaged the global developer community to benchmark their programs on highly complex and realistic data sets, generated from ∼700 newly sequenced microorganisms and ∼600 novel viruses and plasmids and representing common experimental setups. Assembly and genome binning programs performed well for species represented by individual genomes but were substantially affected by the presence of related strains. Taxonomic profiling and binning programs were proficient at high taxonomic ranks, with a notable performance decrease below family level. Parameter settings markedly affected performance, underscoring their importance for program reproducibility. The CAMI results highlight current challenges but also provide a roadmap for software selection to answer specific research questions.