Metagenomics: Retrospect and Prospects in High Throughput Age

Table 2

Comparison of the unique features of NGS platforms widely applied in metagenomic research.

Sequencer	Roche/454 GS FLX Titanium	HiSeq 2000	SOLiDv4
NGS chemistry	Pyrosequencing	Sequencing by synthesis	Sequencing by ligation and exact call chemistry
Library/template preparation	Emulsion PCR (emPCR)	Solid phase amplification	Emulsion PCR for fragment/mate-pair end sequencing
Average read length	250–310 bp (highest among the NGS platforms); Now approaching 400–500 (titanium) pyroreads	Initially it was 36, now approaching 150	35
Run time (days)	24 hours (fastest of all)	4 days (fragment run); 9 days (mate-pair run)	7 days (fragment run); 14 days (mate-pair run)
Output data/run	0.7 Gb	600 Gb; (over 1 Tb with Illumina’s HiSeq X Ten)	120 Gb
Advantage	Longer reads; Least time for one run; Amenable to multiplexing allowing many samples in single run	High throughput; Most widely used platform	Highest accuracy due to ECC (exact call chemistry)
Limitations	High error rate in homopolymer region; High cost of reagents; Low in throughput; Artificial replicate sequences during ePCR [88]	Short read length; Low multiplexing capability of samples; Single base error with GGC motifs; High error rate at tail end reads [89]	Long run time; Short read length