Genome-centric analysis of short and long read metagenomes reveals uncharacterized microbiome diversity in Southeast Asians

Abstract

Despite extensive efforts to address it, the vastness of uncharacterized ‘dark matter’ microbial genetic diversity can impact short-read sequencing based metagenomic studies. Population-specific biases in genomic reference databases can further compound this problem. Leveraging advances in hybrid assembly (using short and long reads) and Hi-C technologies in a cross-sectional survey, we deeply characterized 109 gut microbiomes from three ethnicities in Singapore to comprehensively reconstruct 4497 medium and high-quality metagenome assembled genomes, 1708 of which were missing in short-read only analysis and with >28× N50 improvement. Species-level clustering identified 70 (>10% of total) novel gut species out of 685, improved reference genomes for 363 species (53% of total), and discovered 3413 strains unique to these populations. Among the top 10 most abundant gut bacteria in our study, one of the species and >80% of strains were unrepresented in existing databases. Annotation of biosynthetic gene clusters (BGCs) uncovered more than 27,000 BGCs with a large fraction (36–88%) unrepresented in current databases, and with several unique clusters predicted to produce bacteriocins that could significantly alter microbiome community structure. These results reveal significant uncharacterized gut microbial diversity in Southeast Asian populations and highlight the utility of hybrid metagenomic references for bioprospecting and disease-focused studies.

Simple, reference-independent analyses help optimize hybrid assembly of microbial community metagenomes

“…We further showed that genome recovery was also greatly impacted by these processes, which combined with marker gene beta diversity, demonstrates that the exact number of iterations for these processes has a significant impact on reconstruction, and thus also interpretation, of the microbial communities. Fortunately, results here followed similar patterns with each other, and also in many aspects mirrored published studies comparing different aspects of hybrid assembly across multiple tools and over several iterations 1,2,4,7,10,[12][13][14][15]17,18,22,[26][27][28][29][30][31]37 . However, there is not only substantial methodological variation between published approaches and the performed here, but also biological variation between each dataset which probably precludes the development and application of a universal or standardized approach for applying hybrid metagenomics to microbial communities.…”

“…A primary goal for many microbial community metagenomic sequencing approaches is to reconstruct high-quality, ideally essentially complete genomes of as many of the members as possible [1][2][3][4][5][6][7][8][9][10]12,13 . Hybrid assembly of microbial metagenomes has already been employed to recover microbial genomes from a variety of ecosystems from waste sludge to human guts, and more [1][2][3][4][5][6][7][8][9][10]12,13 .…”

“…Though the increasing variety of high-throughput short-and long-read (meta)genomic sequencing technologies are only within their first decades of existence, both the sequencing technologies and software development have flourished and has already been implemented to study microbial ecosystems [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15] . Integrating both short-and long-read platforms for single microorganisms or microbial communities is gaining popularity because they compensate for the…”

“…other's weaknessesshorter reads achieve higher accuracy while longer reads offer better contiguity 7,11,[16][17][18][19][20] . Despite the technical costs decreasing and technologies improving, hybrid strategies currently are more cost-effective for recovering high-quality Metagenome-Assembled Genomes (MAGs) from microbial communities than either platform alone 12 .…”

A metagenomics pipeline reveals insertion sequence-driven evolution of the microbiota

“…Identified IS elements were filtered using the following criteria: 1) multiple (minimum of 2 copies) complete copies of the element must be present in at least one genome and 2) the elements must contain identified inverted repeat sequences. To supplement these genomes, we also identified IS elements in hybrid-assembled MAGs from a large human intestinal microbiota study (26). These metagenomes were annotated with Prokka v1.14.6 (90), processed as described above, and then combined with the IS elements identified from the genomes from NCBI Assembly.…”