4

My question is about nanopore sequencing and specifically about the current that is measured by the device in each measurement. The question is: In each measurement in nanopore sequencing, the change in the current between two sides of the membrane at the sensing region is recorded by the sequencer. Is it a single nucleotide or a k-mer or the difference between the first and last base in the kmer that contributes to one single measurement? I am citing two papers here that I looked at:

  1. Nanopore sequencing technology, bioinformatics and applications (Nature 2021)
  2. Three decades of nanopore sequencing (Nature 2016)

Based on my finding in these papers, I realized that all the nucleotide molecules in the k-mer in the sensing region contribute to the measurement, and not only one base, yet I am not sure if it is a good assumption.

I know some models like HMM and NN sequential models can computationally infer the individual bases (for example some basecallers). But my question is more about what is physically measured? In the second cited paper, it is stated that, in more recent nanopore designs, DNA strand can be fixed such that one individual nucleotide is positioned in the sensing region. Still, the diameter of the sensing region is not exactly one base. Moreover, since the current change is recorded not the current itself, the contributor could be the difference between the base which just passed from the sensing region and the one base that just entered?

So I restate my question; what is the contributing factor in single measurements of nanopore sequencing? In case it is the k-mer that is measured, how could we find k? Is it based on the properties of the device and the kit information?

Marjan
  • 309
  • 1
  • 6

1 Answers1

3

The sequencing channels measure electric current, i.e. the flow of charged particles through the nanopore. The circuits are not measuring the bases directly; they're measuring the way the flow of ions is changed when different things are introduced into the pores.

Older R9 pores had a length that could fit about five bases in the pore at any one time, the newer R10 pores have a double-size pore that fits about ten bases. Any of these bases can potentially influence the flow of current, and that degree of influence is increased at the narrower pinch points of the Nanopore. Inconsistent unwinding of double-stranded DNA (or hybrid DNA/RNA) can also affect the current profile over time, so there may also be some effects on the current that transfer from where the unwinding occurs, 20 or so bases prior to the pore introduction.

The 5/6 kmer models are (or were, with R9) a reasonable sequencing model, because that's the number of bases that would fit in the pore, so those are the bases that have the greatest influence on the current.

I'll explain using a bigger physical model of hydroelectricity:

The mental model that I have in my head of a flow cell is a dam with a lake / reservoir, and holes in the dam that slowly drain the lake, spinning turbines that produce electricity (i.e. current). The interesting bits in the lake are the things going through the holes that aren’t water, but the water flow is the only thing that can be measured: there are no cameras attached to the holes, only current meters attached to the turbines.

It just so happens that those non-water things have a strange enough shape that they interrupt the water flow in large and predictable ways, which means that observing the water flow can be used as a good proxy for observing the things themselves.

If the holes get blocked, nothing can move through the holes, so there is no current flow: sequencing cannot happen.

Eventually that water reservoir will get depleted. If the reservoir is fully drained [ionic balance of the carrier ions in the flow cell is equalised], then it doesn’t matter how clean the holes are; the dam will produce no more electricity [sequencing cannot happen].

As an aside, this model helps me to understand why keeping the pores occupied with long DNA prolongs the sequencing life of the flow cell. The flow of carrier ions is reduced when a pore is actively sequencing, so the carrier ion reservoir drains slower, and can keep sequencing for longer.

gringer
  • 14,012
  • 5
  • 23
  • 79