ABSTRACT
Evaluating metagenomic software is key for optimizing metagenome interpretation and focus of the Initiative for the Critical Assessment of Metagenome Interpretation (CAMI). The CAMI II challenge engaged the community to assess methods on realistic and complex datasets with long- and short-read sequences, created computationally from around 1,700 new and known genomes, as well as 600 new plasmids and viruses. Here we analyze 5,002 results by 76 program versions. Substantial improvements were seen in assembly, some due to long-read data. Related strains still were challenging for assembly and genome recovery through binning, as was assembly quality for the latter. Profilers markedly matured, with taxon profilers and binners excelling at higher bacterial ranks, but underperforming for viruses and Archaea. Clinical pathogen detection results revealed a need to improve reproducibility. Runtime and memory usage analyses identified efficient programs, including top performers with other metrics. The results identify challenges and guide researchers in selecting methods for analyses.
Subject(s)
Metagenome , Metagenomics , Archaea/genetics , Metagenomics/methods , Reproducibility of Results , Sequence Analysis, DNA , SoftwareABSTRACT
Binning aims to recover microbial genomes from metagenomic data. For complex metagenomic communities, the available binning methods are far from satisfactory, which usually do not fully use different types of features and important biological knowledge. We developed a novel ensemble binner, MetaBinner, which generates component results with multiple types of features by k-means and uses single-copy gene information for initialization. It then employs a two-stage ensemble strategy based on single-copy genes to integrate the component results efficiently and effectively. Extensive experimental results on three large-scale simulated datasets and one real-world dataset demonstrate that MetaBinner outperforms the state-of-the-art binners significantly.