Sequencing technologies: past, present and future

The sequencing field is fast changing, with new technologies allowing dramatic drop in the cost of sequencing while improving data quality and accessibility.

1970-2000

  • Over four decades ago, Frederick Sanger and his colleagues developed a method to decode the genetic information stored in DNA. 
  • The capillary sequencing method allowed the sequencing of the 1st human genome and, although it is labour intensive, it still remains today’s gold standard in terms of accuracy.

Capillary sequencing machines used in the Human Genome Project.
Image credit: Genome Research Limited

2000-2010

  • Since the 1st human genome, the cost and turnaround time to sequence one human genome, or even several genomes in parallel, has dropped extraordinarily, far more than anyone predicted. 
  • This rapid development was made possible in the early 2000s by two main factors: a new generation of high throughput sequencing technologies referred as second generation sequencing, and the rapid development of computing power.
  • Second generation sequencing had a straightforward sample preparation process and a high parallelisation of the sequencing process – in other words, it sequenced many short DNA strands at the same time. 
  • Machines produced by the biotechnology company Illumina became the best known on the sequencing market with its “$1000 dollars genome”.

Illumina sequencing machines in the sequencing centre at the Sanger Institute in 2009.
Image credit: Genome Research Limited

2010-today

  • From 2010 onward, a third generation of sequencers emerged led by PacBio and Oxford Nanopore Technology (ONT). Both of these technologies use single-molecule sequencing and allow us to decode much longer stretches of DNA fragments, also called long reads.  
  • Although these new machines are more expensive and were initially less accurate than previous sequencing technologies, the benefits of long reads lie in their name. Just as a jigsaw puzzle with a few large pieces is much easier to solve than one with many small ones, assembling a genome is much easier with long read data.
  • During the last decade, 3rd generation sequencers have improved greatly and they now compete with the previous generation sequencers in terms of accuracy (data quality). 
  • Their cost remains high but the (extra-)long reads they produce are extremely useful when we want to sequence species never sequenced before. 
  • Some of the 3rd generation sequencers are no bigger than a smart phone, and were even used on board of the International Space Station.
  • Thanks to this rapid boost in the sequencing world, sequencing technologies can now be introduced into hospitals, or used for very large-scale projects such as the surveillance of the Covid-19 pandemic.
 

MinION sequencing machine (attached to a computer) being used by astronaut and biologist Kate Rubins on the International Space Station.
Image credit: NASA

 

  1970s 1990s 2000s 2010s 2020s
Generation 1st generation sequencing Next generation sequencing (NGS) Future Generations
2nd generation 3rd generation
Technologies Sanger sequencing (manual) Sanger sequencing (automated)
  • Illumina
  • Roche 454
  • Ion Torrent
  • PacBio
  • ONT
  • Genapsys
  • MGI
Breakthrough Gel-based Capillary High-throughput Long reads  
Pros
  • Accuracy
  • High scale
  • Low cost
  • Accuracy
  • Long reads
  • Ultra-long reads, simplifying data analysis
  • Fast turnaround, rapid diagnostics & portable equipment
  • Aiming for cheaper and highly accurate methods
Cons
  • Labour intensive
  • High cost
  • Short reads making the data analysis more difficult
  • High cost
  • Some have similar accuracy to 2nd gen
 

Article written by Louise Aigrain, Senior Staff Scientist in the DNA Pipelines Research and Development team at the Wellcome Sanger Institute.

This page was last updated on 2021-12-14

Í