A Comprehensive Review of Performance of Next-Generation Sequencing Platforms.

Background: Next-generation sequencing methods have been developed and proposed to investigate any query in genomics or clinical activity involving DNA. Technical advancement in these sequencing methods has enhanced sequencing volume to several billion nucleotides within a very short time and low cost. During the last few years, the usage of the latest DNA sequencing platforms in a large number of research projects helped to improve the sequencing methods and technologies, thus enabling a wide variety of research/review publications and applications of sequencing technologies. Objective: The proposed study is aimed at highlighting the most fast and accurate NGS instruments developed by various companies by comparing output per hour, quality of the reads, maximum read length, reads per run, and their applications in various domains. This will help research institutions and biological/clinical laboratories to choose the sequencing instrument best suited to their environment. The end users will have a general overview about the history of the sequencing technologies, latest developments, and improvements made in the sequencing technologies till now.
Results: The proposed study, based on previous studies and manufacturers' descriptions, highlighted that in terms of output per hour, Nanopore PromethION outperformed all sequencers. BGI was on the second position, and Illumina was on the third position.
Conclusion: The proposed study investigated various sequencing instruments and highlighted that, overall, Nanopore PromethION is the fastest sequencing approach. BGI and Nanopore can beat Illumina, which is currently the most popular sequencing company. With respect to quality, Ion Torrent NGS instruments are on the top of the list, Illumina is on the second position, and BGI DNB is on the third position. Secondly, memory- and time-saving algorithms and databases need to be developed to analyze data produced by the 3rd- and 4th-generation sequencing methods. This study will help people to adopt the best suited sequencing platform for their research work, clinical or diagnostic activities.

1. Introduction

DNA sequencing methods have a history of only 60 years back, but these methods evolved very rapidly and can be said an outstanding example of progress resulting in enormous improvement and enhancement in cost reduction, high throughput, capability, and applications [1-3]. History of DNA sequencing started when two fundamental methods, i.e., Sanger sequencing [4] and Maxam and Gilbert's approach [5], were introduced. Developments in polymerize chain reaction [6, 7], availability of good quality enzymes to modify DNA, and fluorescent automated sequencing enabled to sequence first human genome in 2001 [8, 9]. Afterwards, giant revolution in DNA sequencing methods, chemistries, and bioinformatics analysis approaches were observed.

Since 2005, several Next-generation sequencing (NGS) methods have been developed and proposed to investigate any query in genomics or clinical activity involving DNA [10, 11]. NGS proposes a novel way of sequencing constituting various approaches that depend on the amalgamation of preparing template, determining order of the bases, aligning sequences and genome assembly [12]. A major advantage of NGS over traditional mutation detection methods is the ability to sequence multiple genes and highlight millions of variants simultaneously. Other advantages include minimal DNA input, faster turnaround time; NGS has revolutionized the speed of genetic and genomic discovery and advanced our understanding of the molecular mechanisms of disease and potential treatment options. Technical advancements in these sequencing methods (replacing radiolabeling with fluorescent dyes and gel electrophoresis with capillary array electrophoresis) introduced automation in the sequencing approaches and enhanced sequencing volume to several thousand base pairs in a single run [13]. The NGS instruments can generate several billion nucleotides within a very short time and low cost [14-17]. These capabilities enabled NGS methods to use in a number of areas such as whole-genome sequencing (WGS), whole-exome sequencing (WES/ES), variant calling (VC), targeting sequencing (TS), and transcriptome sequencing (TCS) [18]. During the last few years, the usage of the latest DNA sequencing platforms in a large number of research projects helped to improve the sequencing methods and technologies, thus enabling a wide variety of research/review publications and applications of sequencing technologies. Each year, several hundreds of publications are being published, highlighting the importance of sequencing technologies.

Over the last decade, dozens of excellent studies describing advantages, disadvantages, and applications of sequencing methods [2, 12, 19, 20] including Sanger sequencing also termed as first-generation sequencing (1stGS), NGS also called as second-generation sequencing (2ndGS), third-generation sequencing (3rdGS), and fourth-generation sequencing (4thGS) have been published. History of sequencing methods reveals amazing pace of developments and improvements in these technologies that now enabled us to sequence genomes of all species at very low cost and a high speed.

The proposed study presents history, needs, and reasons of evolving the sequencing technologies. For this purpose, 120 relevant articles from PubMed and journals web sites were downloaded. The keywords such as “NGS,” “Sequencing technology,” “Sequencing chemistry,” “Comparison of NGS instruments,” and “Quality of NGS instruments” were provided to Google search engine to search these articles. At the end, 65 articles having detailed information about the history, efficiency, quality, and comparison of sequencing technologies/instruments were selected for writing this review article. It provides a detailed overview of the sequencing approaches starting from first- to fourth-generation sequencing methods. The technical features of the new and most popular sequencing instruments by various companies such as Illumina, Ion Torren, GenapSys, QIAGEN, and BGI were also summarized and compared. The proposed study contributed by highlighting the most efficient and accurate NGS instruments and helped the researchers and clinicians to get DNA sequenced through an instrument best suited to them. This study will provide end users with the knowledge of history, background chemistries, and latest developments in the sequence technologies and help them in selecting the most suitable NGS instrument based on their needs.

2. Evolution of High-Throughput Sequencing Technologies

Initial studies which were performed before 2005 including human genome project used DNA sequencing approaches were generally called as 1stGS (1970). The most famous among them were the sequencing methods discovered by Sager and Maxam and Gilbert [21, 22]. Slow speed and high cost of sequencing DNA by 1stGS methods raised the need of fast and cheap DNA sequencing technologies. 2ndGS methods based on the concept of massively parallel sequencing were made available in 2005. The most popular 2ndGS platforms were developed and commercialized by Roche Life Sciences, Thermo Fisher Scientific, Illumina, and Applied Biosystems. These methods are also termed as NGS platforms and have revolutionized the DNA sequencing. NGS has several advantages over 1stGS. Some more important benefits are (1) massive throughput, generating a number of short DNA sequences called as reads in parallel, (2) high speed, and (3) low cost. The reads generated by NGS methods range from 50 bp to 300 bp in length. NGS technologies are classified into two groups, sequencing by hybridization and sequencing by synthesis (SBS). Sequencing by synthesis (SBS) is actually Illumina sequencing technology and is the most popular approach generating 90% of the world's sequence data [23]. The 3rdGS approaches (2010) include Single-Molecule Sequencing (SMS) and True Single-Molecule Sequencing (tSMS). These technologies need less starting DNA material and work without amplifying the template DNA. The 4thGS (2014) also called as nanopore sequencing include majorly the MinION by Oxford Nanopore Technology (ONT). This approach actually incorporated nanopore technology in 3rdGS. The 4thGS has capability to sequence fixed cells and tissues in real time without requiring amplification and repeated cycles in the synthesis phase [21]. Figure 1 shows evolution of sequencing methods.

Figure 1

Evolution of the sequencing methods in chronological order.

3. Detailed Overview of the Sequencing Methods

3.1. First-Generation Sequencing

The first process of DNA sequencing, called as Sanger sequencing, was published in 1977. This method uses sequencing by synthesis (SBS) approach of radioactively labeled DNA strand complementary to the template strand by employing the dideoxy chain termination technique. The fragments are then investigated using polyacrylamide gel electrophoresis. This technique was then improved, automated, and made available for commercial purpose [24, 25] and is termed as 1stGS. Major improvements were the introduction of capillary electrophoresis with gel electrophoresis [26, 27]; replacement of radioactively labeled DNA with fluorescent labeled DNA further advancement was ensured by using recombinant DNA [28] and the PCR technologies [6]. The major drawback of 1stGS is generation/analysis of only one sequence per electrophoresis lane or capillary tube. This is the reason of dividing DNA into fragments. One thousand long read lengths were sequenced with 99.99% accuracy. Major limitations of this method were low throughput and high cost 29. For example, this process was so costly that the human genome project consumed almost 13 years and US$ 2.7 billion. Later on, improvement in 1stGS enabled to sequence of another human genome for US$ 10 million. However, with the passage of time, 1stGS reached its limit and was taken as costly approach. One thing to note is that this technology is still used for validating DNA sequences and target resequencing [1].

3.2. Next/Second-Generation Sequencing

The NGS, high throughput, and massively parallel sequencing are the terms used for this type of sequencing. It is also called as 2ndGS. This approach works without separating the sequencing reactions into lanes, capillaries, or tubes. NGS allows billions of sequencing processes to be happened simultaneously in parallel on a slid surface (glass or beads), an enormous improvement in throughput and cost compared to 1stGS.

3.2.1. Illumina

Illumina, Inc. [29] is a leading manufacturer of various sequencing instruments. It was established in 1998 in San Diego, CA. Currently, it provides a number of sequencing platforms categorized in two groups: Benchtop Sequencers (BTS) and Production Scale Sequencers (PSS). All BTSs provide support for (1) WGS for small organisms such as microbes and viruses, (2) target gene sequencing (TGS), (3) target gene expression profiling (TGEP), (4) miRNA and sRNA analysis profiling, and (5) 16S metagenomic sequencing (MS) (except iSeq100). NextSeq 550 Series and NextSeq 1000 & 2000 have extra applications such as exome sequencing (ES), s-cell profiling, chip-seq analysis, methylation sequencing, MS, and cell-free sequencing (CFS). Comparison of BTS is given in Table 1. Among PSSs, only NovaSeq 6000 supports WGS of humans, plants, and animals. Functionalities provided by other PSSs are similar to those of benchtop sequencers. Table 2 provides a summary of applications, features, and performance of the PSSs. HiSeq 2500, HiSeq 3000, HiSeq 4000, HiSeq X Ten, and HiSeq X five have been declared to discontinue (Illumina, 2021). However, their support will be provided up to March 31, 2024. So, these sequencers are not discussed here. Illumina sequencing method is based on SBS. Reaction system is a mixture of DNA polymerase, primers, and 4 dNTP with base specific fluorescent markers. The 3′-OH of dNTPs ensures addition of one base at time. On completion of the sequencing reaction, DNA polymerase and the unused dNTP are eluted. For fluorescence excitation, buffer solution is then added. The fluorescence signal is recorded by optical equipment. Optical signals, generated by optical equipment, are used for base calling. To perform next round of sequencing reaction, a chemical reagent is used for quenching fluorescence signal and remove the dNTP 3′-OH protective group. Sequencing data generated during the same experiment have the same length. The latest sequencing platforms can generate DNA sequence in paired-end fashion (22 × 300 bp), i.e., can read both ends of a fragment [30]. Signal decay and dephasing occurred due to incorrect cleavage of fluorescent label or terminating moieties. Average error rates of the sequencing platforms are 1-1.5% [31].

Table 1

A comparison of Illumina benchtop sequencers [29].

Methods/applications	iSeq 100	MiniSeq	MiSeq series	NextSeq 550 series	Next Seq 1000 & 2000
Ideal for	Every size lab	TG sequencing	Long read applications	Exome and transcriptome sequencing	miRNA and sRNA analysis
Major applications	sWGS (microbes) and TGS	iSeq 100+TG EP and 16S MS	iSeq 100+16S MGS	iSeq 100+TCS	sWGS (microbes), ES, SC profiling, TS, miRNA, and sRNA analysis
Max. data quality	>85% > Q30	>85% > Q30	>90% > Q30	>80% > Q30	>90% > Q30
Run time	9.5–19 h	4–24 hours	4–55 hours	12–30 hours	11-48 hours
Maximum output	1.2 Gb	7.5 Gb	15 Gb	120 Gb	330 Gb∗
Maximum reads per run	4 million	25 million	25 million	400 million	1.1 billion
Maximum read length	2 × 150 bp	2 × 150 bp	2 × 300 bp	2 × 150 bp	2 × 150 bp

Table 2

A comparison of Illumina production scale sequencer sequencers [29].

Methods/applications	NextSeq 550	NextSeq 550Dx	NextSeq 1000 & 2000	NovaSeq 6000
Ideal for	Research	Research+in vitro diagnostic	Targeted sequencing	Long read applications
Major applications	sWGS (microbes), TGS, and TCS	NextSeq 550+clinical NGS applications	NextSeq 550 series+SCP	NextSeq 550 series+NextSeq 1000 & 2000+lWGS
Max. data quality	>80% > Q30	>75% > Q30	>90% > Q30	>90% > Q30
Run time	12-30 hours	35 hours	11-48 hours	13-44 hours
Maximum output	120 Gb	90 Gb	360 Gb	6000 Gb
Maximum reads per run	400 million	300 million	1.2 billion	20 billion
Maximum read length	2 × 150 bp	2 × 150 bp	2 × 150 bp	2 × 250 bp

3.2.2. Ion Torrent

Ion Torrent was launched in 2011 [32]. Ion Torrent is an SBS-based approach and uses pH measurements for generating nucleotide sequences. Length of sequencing reads generated by Ion Torrent varies. Ion Torrent sequencing machines cannot generate sequencing from either ends of a fragment [30]. There are four Ion Torrent instruments; GeneXus system has ability to produce data analysis report in a single day using an automated workflow with only two touch points. This is economical for the lowest sample input and can be placed in lab or a house regardless of the level of NGS expertise. This is also termed as in-house NGS system. Ion GeneStudio S5 systems support efficient, scalable, and low-cost targeted sequencing. Based upon the Ion chips, there are five variants of this instrument with ability of generating 2M to 130M reads and 0.3 to 50 Gb data in a single run by consuming 3-21.5 hours. Table 3 describes applications, performance, and features of Ion GeneStudio S5 systems. The PGM Dx system is suitable for regulated lab environments and in vitro diagnostic. It is an integrated system of NGS instrument, reagents, consumables and software tools for sequencing and data analysis. The Ion Chef System is an improved version of Ion GeneStudio S5 systems. It is an automated approach to prepare library for Ion AmpliSeq, reproducible template and to load chip [33].

Table 3

Comparison of various Ion Torrent NGS instruments [33].

Methods/applications	Ion GeneStudio S5 System	Ion GeneStudio S5 Plus System	Ion GeneStudio S5 Prime System	Ion PGM 314	Ion PGM 316	Ion PGM 318	Ion Torrent GeneXus
Major applications	ES, TCS, EP, TGS, and sWGS (microbes)			TS (DNA/RNA), CNV, sRNA sequencing, de novo MICS, ChIP-seq, MA, SNP verification, and GbS			WGS, WES, TGS
Run time (hours)	4.5–21.5	3-20	3-10	2.3-3.7	3.0-4.9	4.4-7.3	2.5
Maximum output	15 Gb	30 Gb	50 Gb	300–100 Mb	300 Mb–1 Gb	600 Mb–2 Gb	15 Gb
Maximum reads per run	2-80 million	2-130 million	2-130 million	400–550 thousand	2–3 million	4–5.5 million	12-15 million
Maximum read length	200-600 bp	200-600 bp	200-600 bp	200-400 bp	200-400 bp	200-400 bp	200
Compatible chips	Ion 510, 520, 530, 540 chips	Ion 510, 520, 530		Ion 314 Chip V2 or Ion 314 Chip v2 BC	Ion 316 Chip V2 or Ion 316 Chip v2 BC	Ion 318 Chip V2 or Ion 318 Chip v2 BC	Ion Torrent GX5 Chip
Max. data quality	>99% > Q30	>80% > Q30	>99% > Q30	>99% > Q30	>99% > Q30	>99% > Q30	>99% > Q30

3.2.3. GenapSys

The GenapSys (founded in 2010) is a company from the Stanford Genome Technology Center. The GenapSys Sequencer enhanced SBS technique by embedding thermal detection of nucleotide incorporations [34]. It is a small (less than ten pounds), low-priced, and easy to use, even good for beginners in the genomic filed. The electrical chip has several million sensors each having a single bead coated in thousands of clonal copies of a nucleotide sequence. The DNA bases are poured across the chip in a sequence, and successful incorporation is noticed by changes in impedance as the complementary DNA strand grows. Three versions of the chip, based on varying number of sensors, are available: 1 million sensors, 16 million sensors, and 144 million sensors. This technology has enabled the sequencer to produce a massive range of data quantity. For example, the GenapSys with sixteen million sensor chips can generate thirteen million reads per day providing read length of 150 bp and accuracy level of >80% > Q30 (raw accuracy 99.9%). However, its performance can be enhanced to ES, TS, and SCP by using a cluster of chips. The GenapSys can be used for identifying pathogen, sRNA, sWGS, targeted mRNA, SCP, and gene editing [35].

3.2.4. QIAGEN

QIAGEN provides GeneReader for NGS data generation. The nucleotides are detected by matching fluorescent signals templates clonally amplified by GeneRead QIAcube. The GeneReader can be used only by the qualified persons trained in MB approaches and GeneReader itself. It is claimed to a complete workflow that eliminates challenges faced during sample preparation and provides very good understanding of the results. The GeneReader system helps in all sample processing and sequencing phases such as DNA extraction, library preparation, sequencing, bioinformatics data analysis, clinical implications, and evidence. It employs “QCI Analyze” and “QCI Interpret” for analyzing biological data, variant calling and their annotation, read mapping, and visualization of the alignment. Quality (>85% > Q30) is assured at run level to validate each variant for minimizing false-positive and false-negative indications [36, 37].

3.2.5. Complete Genomics Technology/BGI

Complete genomics, founded in 2006, is specialized in whole human genome sequencing. In 2013, it was purchased by BGI-Shenzhen, China, that is one of the world's leading institutions providing genomic services. The BGI provides a number of sequencing (Table 4) and data analysis tools and technologies for research, agriculture, medical, and environment applications [38]. The complete genomics developed a technology by emerging sequencing by hybridization and ligation [39], called as DNA nanoball (DNB) sequencing. Rolling circle replication is used to amplify DNA fragments consisting of 440-500 bp into DNBs. This needs generation of entire circular templates before the generation of nanoballs. DNBs are poured into a flow cell, one nanoball in each well. The template bases ranging from 1 to 10 are processed in paired-end fashion comparable to Exact Call Chemistry in SOLiD sequencing [40-42]. After eliminating ligated sequences, new probes are added, according to various interrogated positions. The process of annealing, washing, ligation, and image reading is iterated for all positions nearby to one end of one adapter. This procedure is performed for all remaining termini of the adapter. The main disadvantage of DNB sequencing is run time and short read lengths. The key advantage of this technique is the high quantity of DNBs (almost 350 million). Later on, the Retrovolocity approach was incorporated for generating high quality WG and WE sequence having 50x coverage in <8 days [43]. As per their claim, more than 20,000 whole genomes of humans have been sequenced using the propriety instrument and procedures [38].

Table 4

Comparison of various BGI NGS instruments [38].

Methods/applications	DNBSEQ-T7	DNBSEQ-G400 FAST	DNBSEQ-G400	DNBSEQ-G50
Major applications	WGS, DES, EGS, TS	WGS, WES, TS, MGS, RNA-seq	WGS, WES	Targeted sequencing (DNA & RNA), pathogen identification, and SPS
Max. run time (hours)	30	13	37	40
Maximum output	6 Tb	330 Gb	1440 Gb	150 Gb
Maximum reads per run	5000 million	550 million	1800 million	770 million
Maximum read length	150 PE	150 PE	200 PE/400 SE	150 PE
Data quality	>85% > Q30	>85% > Q30	>85% > Q30	>85% > Q30

3.2.6. Roche 45

The Roche GS-FLX 454 Genome Sequencer was the first commercial system launched as the 454 Sequencer in 2004 [42, 44]. Using this platform, the second complete genome of an individual (James D. Watson) was sequenced. The upgraded 454 GS FLX Titanium system introduced by Roche in 2008 enhanced the average read length and accuracy to 700 bp and 99.997%, respectively. This platform improved an output of 0.7 Gb of data per run within 24 hours. The GS Junior bench-top sequencer system produced the average read length of 700 bp, throughput of 70 Mb, and runtime of 10 to 18 hours. However, Roche decided to reduce its focus on gene sequencing and shut down 454 Life Sciences sequencing services by the end of 2013, so Roche NGS instruments will not be discussed more in this study [45-47].

3.3. Third-Generation Sequencing

Second-generation sequencing approaches require PCR amplification of the template DNA which causes sequencing errors. This limitation can be overcome if sequencing is performed based on a single molecule without amplification. Secondly, time needed to produce results is also long because several scanning and washing cycles have to be run. Due to the addition of each nucleotide, synchronicity is also lost which may result in noisy sequencing data and short length of the reads.

The Single-Molecule Sequencing (SMS) which is 3rdGS approach is also termed as single template approach. The most famous SMS approach is Single-Molecule Real Time Sequencing (SMRT) by Pacific Biosciences (PacBio). This method uses sequencing by synthesis chemistry similar to some 2nd-generation sequencing methods but needs less starting material and PCR amplification of the template DNA which results in low error rate and produce long reads with less run time [48]. SMRT can generate tens of kilobases long reads; for examples, the latest PacBio sequencer (Sequel IIe System released on Oct. 05, 2020) can produce 4 million reads with more 99% accuracy in just 30 hours. This system was shown to have more contiguity (N50), correctness (quality score), and completeness (genome size) compared to Nanopore and Illumina (Table 5) whereas cost of PacBio HiFi Sequencing was also reported very low (Table 6) compared to its competitors [49].

Table 5

An overview of human genome assembly quality metrics between PacBio system, Nanopore, and Illumina [49].

	Nanopore+Illumina	PacBio HiFi sequencing
Contiguity (N50)	32.3 Mb	98.7 Mb
Correctness (quality score)	Q34	Q51
Completeness (genome size)	2.8 Gb	3.1 Gb

Table 6

Overall costs for sequencing a human genome [49].

	Nanopore+Illumina	PacBio HiFi sequencing
	(US $)
Consumables	4800	3800
Compute	5050	3850
Data storage	5200	3900

The 3rd-generation sequencing has several advantages over 2nd-generation sequencing; for example, higher throughput, detecting haplotype directly, longer read lengths, better consensus accuracy to identify rare variants, whole chromosome phasing, and small amount of sample are the salient features of the 3rd-generation sequencing which had it useful in clinical diagnostic [50].

3.4. Fourth-Generation Sequencing

The fourth-generation sequencing integrated nanopore technology into SMS. This technology performs real-time sequencing without amplification and repeated cycles by eliminating synthesis and therefore is called as 4G sequencing. The 4thGS, also called in situ sequencing technology, has opened new horizons in DNA sequencing by making it possible to identify order of nucleotides in the fixed cells and tissues [21]. It differs from other sequencing generation approaches in two ways. Firstly, spatial distribution of the DNA reads over the sample can be observed which provide very useful information for highlighting tissue heterogeneity based upon the known markers. The second difference is that large number of cells can be analyzed simultaneously. For example, robust single cell RNA sequencing approaches were developed, which are cheap and are capable to sequence a number of cells with very few pictograms of the starting material [51]. Drawback of this technique is that tissue material is composed of several thousands of cells and sequencing single cells is not technically and computationally an easy job. However, it is predicted that in situ sequencing will be used to extract clinically important information from data produced by conventional NGS approaches. Targeted in situ sequencing method may be applied for filtering validated biomarkers directly on the samples whereas nontargeted technique may be useful for developing molecular profiles of the samples for classifying a disease on the molecular level or to satisfy the patients. Integrating in situ sequencing in the conventional NGS methods would expedite the development of these methods and these will eventually become essential tools for personalized medicine. Nanopore sequencing, the most popular 4thGS platform, has ability to identify molecules (proteins, DNA, RNA, etc.) while they are passed through nanoscale holes entrenched in a thin membrane [52]. In this approach, an electric field forces individual molecules to pass through a nanopore having 2 nm diameter. Due to very thin pore, single-stranded molecules are passed through the pore in a firm linear order. Distinguished electric signals are generated as DNA molecule passes through the pore. The most famous nanopore technology is the Oxford nanopore Technology. It is one of the most robust sequence technologies and can sequence whole genome with 1 million base pairs long reads and diagnose diseases very efficiently and with very low cost [53]. The MinION, which was released in 2014, is the first application of nanopore technology. Other higher throughput nanopore devices from Oxford Nanopore Technologies are GridION Mk1 and PromethION 24/48. GridION Mk1has 1-5 flow cells with the ability of generating 250 GB data. PromethION 24/48 has 1-48 flow cells and can produce data up to 15 TB [54]. Nanopore sequencing is classified into three categories. In case of 1D, single-stranded DNA is sequenced. In 2D, two strands of the DNA were bounded by a hairpin-like structure. The first sequence of one strand of DNA is obtained, and then, the second strand DNA is sequenced. In this way, sequencing is repeated twice to raise base calling quality. 1D2 is very close to 2D, but hairpin structure is not needed for keeping connected two strands of DNA.

4. Comparison of Sequencing Platforms

All sequencing instrument manufacturing companies offer a variety of sequencing platforms. Some produce small data and others produce huge amount of data in a single run. Reads' length and time consumed to generate data also vary among these sequencers. Table 7 provides comparison of various high-performing sequencers, and Table 8 shows analysis in terms of advantages and disadvantages of the sequencing generations. Per hour output analysis of high-performing sequencers showed that Nanopore PromethION outperformed all sequencers. BGI was on the second position and Illumina was on the third position (Figure 2).

Table 7

Comparison of various high-performing sequencing instruments∗.

Manufacturer	Read length	Data output	Max. run time (hours)	Chemistry	Key applications∗∗
Illumina (NovaSeq 6000)	300 PE	6 Tb (6000 Gb)	44	Sequencing by synthesis	SS-WGS and TGS, TGEP, 16sMGS, WES, SCP, LS-WGS, CA, MS, MGP, CFS, LBA
Thermo Fisher Scientific Ion Torrent (Ion GeneStudio S5 Prime)	600 SE	50 Gb	12	Sequencing by synthesis	WGS, WES, TGS
GenapSys (16 chips)	150 SE	2 Gb	24	Sequencing by synthesis	TS, SS-WGS, GEV, 16S rRNA sequencing, sRNA sequencing, TSCAS
QIAGEN (GeneReader)	100 SE	Not available	Not available	Sequencing by synthesis	Cancer research and identifying mutations
BGI/Complete Genomics	400 SE	6 Tb (6000 Gb)	40	DNA nanoball	Small and large WGS, WES and TGS
PacBio (HiFi Reads)	25 Kb	66.5 Gb	30	Real-time sequencing	DN sequencing, FT, identifying ASI, mutations, and EPM
Nanopore (PromethION)	4 Mb	14 Tb (14000 Gb)	72	Real-time sequencing	SV, GS, phasing, DNA and RNA base modifications, FT, and isoform detection

∗Performance comparison is given as per manufacturer's description. ∗∗Applications by all sequencers of the respective manufacturer are listed. ∗∗Full names are given in Abbreviations.

Table 8

Advantages and disadvantages of sequencing generations.

Sequencing generation	Advantages	Disadvantages
First generation	High accuracy	High cost
	Helps in validating findings of NGS	Low throughput
Second generation	High throughput	Short read length
	Low cost	Difficult sample preparation
	Have clinical applications	PCR amplification
	Short run time	Long run time
Third generation	No PCR amplification	High sequencing error rate10–15% in the PacBio and 5–20% in the ONTFresh DNA requires for ensuring quality of ultralong readsDatabase systems and algorithms/tools are rare for analyzing 3rd and 4th GS data
	Require less starting material
	Longer read lengths
	Very low cost
	Low error rate during library preparation
Fourth generation	Advantages of 3rdGS+
	Ultrafast: scan of whole genome in 15 minutes
	Spatial distribution of the sequencing reads over the sample can be seen

Figure 2

Per hour output analysis of high-performing sequencers.

5. Discussion

Rapid evolving approaches for genome sequencing have resulted in significant reduction in cost and time for NGS data generation and amazing increase in accuracy and throughput by using very less amount of starting material of DNA. Every day brings innovation in these technologies, and the field of genomics is progressing steadily by opening new horizons in various domains of life sciences [55-57]. Two features of NGS systems, i.e., extensive reduction in time and substantial increase in accuracy, have enabled NGS methods to be used in diagnostics, prognostics, and predicting variations [58-61] in the human genomes—leading towards the personalized medicine [62-64]. On the other hand, NGS methods have made it possible to conduct large-scale “omics” studies such as genomics, exomics, epigenomics, metagenomics, and transcriptomics [65, 66] which provided insight into the basic as well as applied research areas.

Among the SGS technologies, Illumina has been reported to offer a big variety of benchtop and production scale NGS instruments and they are the most popular [2] among the clients. The instruments are more economical [1] and are among the platforms that have the highest throughput [67, 68]. The Ion Torrent instruments are more automatic in the sense that in addition to automation in NGS data generation and analysis they provide automation in library preparation as well. Some studies have shown that Ion Torrent methods are more suitable for forensic SNP investigation [69] and have better throughput than Illumina HiSeq 2000 [70, 71]. Although Roche 454 was one of the most popular instruments, now they have been discontinued [45-47]. Some studies have reported that Roche instruments are more error prone and costly and have low throughput as compared to other NGS instruments [67, 71]. The GenapSys is lightweight, low-priced, and easy to use, even good for beginners in the genomic filed. This instrument has the electrical chip with different number of sensors: 1 million sensors, 16 million sensors, and 144 million sensors. This technology has enabled the sequencer to produce a massive range of data quantity. The GenapSys with sixteen million sensor chips can generate thirteen million reads per day. The GenapSys can be used for identifying pathogen, sRNA, sWGS, targeted mRNA, SCP, and gene editing [35]. The GeneReader by QIAGEN can be used only by the qualified persons trained in MB approaches and GeneReader itself. It presents a complete workflow starting from sample preparation to NGS data generation and provides very good understanding of the results. It employs “QCI Analyze” and “QCI Interpret” for analyzing biological data and variant calling and their annotation. The GeneReader ensures quality at run level to validate each variant for minimizing false-positive and false-negative indications [36, 37]. Complete genomics, founded in 2006 and purchased by BGI-Shenzhen, China, in 2013, is one of the world's leading institutions providing genomics services. The BGI provides a number of services for research, agriculture, medical, and environment applications [38]. The BGI instruments generate high-quality WG and WE sequence with 50x coverage in <8 days [49]. As per their claim, more than 20,000 whole genomes of humans have been sequenced using the propriety instrument and procedures [38].

The third-generation sequencing technology has some advantages over SGS such as this requires less starting DNA material and does not require PCR amplification of the template DNA. This has enabled SMS to produce more accurate long reads within less time [48]. The latest PacBio sequencer (Sequel IIe System) has the ability to produce 4 million reads with more 99% accuracy in just 30 hours. This system is more accurate as compared to Nanopore and Illumina whereas the cost of PacBio HiFi Sequencing was also reported as very low [49]. The tSMS can sequence millions of individual molecules even from a picogram sample. The tSMS has an important improvement over the SGS in the sense that it can perform RNA sequencing directly [50].

Nanopore sequencing, i.e., integration of nanopore technology into the third-generation sequencing technology, falls in the category of fourth-generation sequencing. It can sequence fixed cells and tissues in real time without requiring amplification and repeated cycles in the synthesis phase [21]. The most famous nanopore technology is the ONT. It can sequence whole genome with 1 million base pair long reads and diagnose diseases very efficiently and with very low cost [53]. The MinION is the first application of nanopore technology. Others are GridION Mk1 and PromethION 24/48. GridION Mk1 can generate 250 GB data and PromethION 24/48 can produce data up to 15 TB [54].

To summarize the discussion, this may be claimed that NGS technologies are being developed with an amazing pace. In the near future, NGS technologies and instruments will be seen in action in clinical and diagnostic labs all around the world, helping us to fulfill the dream of personalized medicine. In addition, there will be very good portable and fully automatic devices for generating NGS data. So, to cater needs of the future, algorithms and databases should be developed for storing, processing, analyzing, and visualizing data of each patient, which may be useful for clinicians to make therapeutic decisions. Major challenges of NGS approaches include the lack of standardized procedures for managing quality, sequencing workflows, sequencing data handling, and analyzing [72, 73].

6. Conclusion

Sequencing platforms have reshaped the genomic era and are helping us in understanding and characterizing genomes of humans, animals, and plants. Every day brings innovation in sequencing chemistry, throughput, and nucleotide detection which enables sequencing process very easy, fast, and low-priced. The proposed study investigated various sequencing instruments and highlighted advantages, disadvantages, and applications based on the previous studies and the material provided by the manufacturers on their websites. Each instrument has different application, run time, and output per hour; however, overall, Nanopore PromethION is the fastest sequencing approach. It can produce 194 Gb data in an hour. BGI with an output of 150 Gb data per hour was on the second position, and Illumina with an output of 136 Gb data per hour was on the third position. The results of the proposed study showed that BGI and Nanopore can beat Illumina, which is currently the most popular sequencing company, and overcome the genomic market very soon. With respect to quality, Ion Torrent NGS instruments are on the top of the list, Illumina is on the second position, and BGI DNB is on the third position. Secondly, memory- and time-saving algorithms and databases need to be developed to analyze data produced by the 3rd- and 4th-generation sequencing methods.

7. Outcome of the Review and Recommendations

The Nanopore PromethION should be used in large-scale projects for getting maximum data in minimum time. The Ion Torrent NGS and Illumina instruments may be used for small projects where quality is an essential element. Tools and databases for storing, analyzing, and visualizing big data biology should be developed so that life science researchers may contribute in improving humans' health effectively.