Long-read sequencing
Circular consensus sequencing in nanowells: inserts sheared to ~15–20 kb are made into circles, read in many passes by one polymerase, and combined into reads usually above Q30.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
PacBio HiFi
Circular consensus sequencing in nanowells: inserts sheared to ~15–20 kb are made into circles, read in many passes by one polymerase, and combined into reads usually above Q30.
How it differs from the standard method
- Insert size is chosen at shearing and size selection; longer means fewer passes.
- Loading targets about one molecule per well; single-loaded wells top out near 37%.
- DNA damage repair matters: damage stalls the polymerase and costs passes.
- Kinetics of incorporation also report base modifications such as 5mC.
1One molecule, read end to end
Short-read platforms read millions of clonal clusters a few hundred bases at a time, then rebuild genomes computationally. Long-read platforms read single molecules thousands to hundreds of thousands of bases long, with no amplification.
Circular consensus
A polymerase in a nanowell copies a circular template over and over; the passes are combined into one highly accurate read of 10–25 kb.
Strand through a pore
A motor protein feeds DNA through a protein pore and the ionic current is decoded into bases; reads are as long as the molecules, sometimes over 100 kb.
Read length
Half of all bases sit in reads at least this long; the standard summary of a long-read run.
2What sets length, accuracy and yield
- DNA integrity first: you can shear long DNA shorter but never make short DNA longer. Gentle extraction and handling decide the ceiling.
- HiFi trades insert length against passes. A polymerase reads a roughly fixed total length per well, so longer inserts get fewer laps and lower consensus quality.
- Loading follows Poisson statistics in nanowells. At most about 37% of wells hold exactly one molecule, at an average of one per well. Overloading creates multi-loaded wells that are discarded.
- Pores capture molecules, not mass, so nanopore libraries are loaded in fmol calculated from the real fragment length. Too much long DNA clogs pores and shortens the run.
- Nanopore accuracy depends on the basecalling model: faster models are noisier, and the raw signal can be re-basecalled later.
- Short fragments are captured or loaded preferentially, so size selection protects read length.
3Choosing between them
HiFi reads are both long and accurate (typically Q30 or better), which suits small-variant calling, phasing and assembly. Their length is limited by the trade against passes.
Key terms
Practise these as flashcards →A nanowell small enough that only the polymerase at its floor is illuminated, so one molecule is watched at a time.
A circular template made by ligating hairpin adapters to both ends of a double-stranded insert.
One traversal of the insert by the polymerase; several are combined into the consensus read.
The total sequence a polymerase produces before it stops; it bounds the number of passes.
The share of active nanopores reading a strand at a given moment rather than waiting for one.
The length such that half of all sequenced bases are in reads at least that long.
A read that covers a feature and anchors in unique sequence on both sides; needed to size repeats and resolve structural variants.
The neural network that turns raw nanopore signal into bases; fast, high-accuracy and super-accuracy variants trade compute for accuracy.
When to use it
- Sizing repeat expansions and resolving structural variants, segmental duplications and complex rearrangements.
- De novo genome assembly, phasing of haplotypes, and closing gaps in reference genomes.
- Full-length transcript isoforms (long-read RNA or cDNA sequencing).
- Native methylation and other base modifications without bisulfite conversion.
- Whole-plasmid or whole-amplicon verification in a single read.
Compared with alternatives
Cheaper per base and very accurate; prefer it for counting applications and small variants in unique sequence.
One accurate ~800 bp read per reaction; still the quick check for one junction or variant.
The application layer: germline, somatic and assembly analyses can be run on long or short reads.
Short-read library prep; long-read libraries keep fragments long and are usually not amplified.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
Circular consensus sequencing in nanowells: inserts sheared to ~15–20 kb are made into circles, read in many passes by one polymerase, and combined into reads usually above Q30.
- Insert size is chosen at shearing and size selection; longer means fewer passes.
- Loading targets about one molecule per well; single-loaded wells top out near 37%.
- DNA damage repair matters: damage stalls the polymerase and costs passes.
- Kinetics of incorporation also report base modifications such as 5mC.