Table 1.

Advantages and limitations of illumina and nanopore‐based metagenomics in microbiome research

	Library preparation and sequencing	Reads‐based community and functional analysis	Assembly and binning
Illumina‐based metagenomics
Advantages	Readily available commercial sequencing service with a relatively low price Low requirement on the input DNA for library construction in terms of both DNA quality and quantity, for example, 1 ng DNA is enough for library construction	Massive SRs with high community coverage, easy to capture signal for populations with very low abundance Various mature bioinformatic frameworks to carry out community, functionality as well as metagenomic binning analysis	High‐accuracy SRs could ensure the accuracy of assembled MAGs
Limitations	High instrumental cost, which results in relatively longer turn‐around time to obtain sequencing data at centralized labs or sequencing companies Unavoidable biases against high‐GC populations by the bridge‐PCR	Generally difficult to assign SRs to a specific phylogenetic lineage, for example, species level	Hard to assemble exogenous elements, resulting in highly fragmented MAGs. For example, even high‐quality MAGs still have >50 contigs
Nanopore‐based metagenomics
Advantages	Relatively low instrument price, which enables short turn‐around time to obtain sequencing results within 48 h at every lab Higher feasibility to customize sequencing protocols for specific sequencing purpose, for example, ReadUntil sequencing No systematic bias during sequencing, but has an evident base‐calling constrain	Long read length enables easy assignment of LRs to a specific phylogenetic lineage (e.g., species level), but correction must be applied to ensure reliable functional annotation	Outstanding capability to obtain highly continuous MAGs from metagenomic assembly
Limitations	High overall sequencing price by commercial sequencing service at present Strict requirement on DNA purity and quantity (>400 ng DNA) to ensure a successful sequencing run with expected data output	High error rates of raw LRs generated by mainstream chemistry, namely, 5%–10% for R9.4 chemistry and 3%–5% for R10.4 chemistry Regular bioinformatic pipelines, like, Prokka, MetaWRAP, unapplicable for raw nanopore‐LRs analysis	Difficulty to assemble low coverage populations due to the sequencing throughput limit which is most often associated with the high sequencing cost Persistence of indel and chimera errors on the assembled the MAGs which limited its application as reference genome

Abbreviations: GC, gas chromatography; LR, long read; MAG, metagenome‐assembled genome; PCR, polymerase chain reaction; SR, short read.