MetaCHIP: community-level horizontal gene transfer identification through the combination of best-match and phylogenetic approaches

Weizhi Song,Shan Zhang,Bernd Wemheuer,Torsten Thomas,Kerrin Steensen

doi:10.1186/s40168-019-0649-y

Weizhi Song, Shan Zhang + Show 3 more

Open Access

https://doi.org/10.1186/s40168-019-0649-y

Copy DOI

Abstract

BackgroundMetagenomic datasets provide an opportunity to study horizontal gene transfer (HGT) on the level of a microbial community. However, current HGT detection methods cannot be applied to community-level datasets or require reference genomes. Here, we present MetaCHIP, a pipeline for reference-independent HGT identification at the community level.ResultsAssessment of MetaCHIP’s performance on simulated datasets revealed that it can predict HGTs with various degrees of genetic divergence from metagenomic datasets. The results also indicated that the detection of very recent gene transfers (i.e. those with low levels of genetic divergence) from metagenomics datasets is largely affected by the read assembly step. Comparison of MetaCHIP with a previous analysis on soil bacteria showed a high level of consistency for the prediction of recent HGTs and revealed a large number of additional non-recent gene transfers, which can provide new biological and ecological insight. Assessment of MetaCHIP’s performance on real metagenomic datasets confirmed the role of HGT in the spread of genes related to antibiotic resistance in the human gut microbiome. Further testing also showed that functions related to energy production and conversion as well as carbohydrate transport and metabolism are frequently transferred among free-living microorganisms.ConclusionMetaCHIP provides an opportunity to study HGTs among members of a microbial community and therefore has several applications in the field of microbial ecology and evolution. MetaCHIP is implemented in Python and freely available at https://github.com/songweizhi/MetaCHIP.

Highlights

Metagenomic datasets provide an opportunity to study horizontal gene transfer (HGT) on the level of a microbial community
The results showed a high degree of congruence between the Single-copy gene (SCG) protein trees and the tree based on 16S rRNA gene sequences for genome bins with completeness higher than 40% (Fig. 3)
Detection of between-class HGTs became unsuccessful at 30% divergence, while at the genus level, the detection threshold was reached at around 20% divergence

Summary

Introduction

Metagenomic datasets provide an opportunity to study horizontal gene transfer (HGT) on the level of a microbial community. The reconstructed genome bins have provided new insights into the biochemistry, physiology and adaptation of previously uncharacterized microbial groups [4,5,6,7,8]. They offer the opportunity to HGT, the transmission of genetic information between organisms, is thought to be an important driver of microbial evolution and adaptation, including the development of antibiotic resistance and virulence [9, 10]. Explicit phylogenetic approaches are employed by Ranger-DTL [15] and AnGST [16], which (2019) 7:36 predict HGTs through the reconciliation of gene trees with corresponding species trees

Methods

Results

Conclusion