I'm studying bioinformatics and I'm confused by shotgun sequencing. In Sanger sequencing we break up the DNA and use ddNTPs in order to determine the exact position of each neucleotide. How exactly does shotgun sequencing help us determine the sequence structure?

If we break up the sequence randomly and then stitch it back together by looking at the overlapping parts, how does that tell us what order the nucleotides are in? How exactly are the nucleotides read so that shotgun sequencing helps us?

  • 1
    $\begingroup$ @AliceD Basically, I'm under the impression that sanger sequencing isn't used in cases for very long DNA strands. Shotgun sequencing works better here but I'm not sure how shotgun sequencing accomplishes the same thing that sanger sequencing does? Sanger sequencing tells me the position of the nucleotides and therefore the structure of the DNA strand. How does shotgun sequencing do this? $\endgroup$ – tryingtolearn Apr 6 '17 at 20:54
  • 1
    $\begingroup$ Fair enough I guess. I'll leave this one up to our molecular biologists to assess. I held my last pipette back in 2001 O.o Auch, I'm getting old $\endgroup$ – AliceD Apr 6 '17 at 20:57
  • 1
    $\begingroup$ Does this answer your question? $\endgroup$ – AliceD Apr 6 '17 at 21:07
  • 1
    $\begingroup$ @AliceD Not exactly. I appreciate you taking the time to search for things though. The only thing that makes sense to me is that the shotgun sequencing involves breaking up the dna strands that are too long, performing sanger sequencing on them, and then looking at overlap. But it doesn't seem to say that on the Wiki, and it says a LOT of other stuff that I don't relaly understand what it's for. $\endgroup$ – tryingtolearn Apr 6 '17 at 21:25
  • 2
    $\begingroup$ Would it answer your question if I said that shotgun sequencing isn't actually a sequencing technique, per se, but a method of assembling a large number of short sequence reads generated by Sanger sequencing? In other words, to sequence long DNA molecules, you would break it down into short fragments that Sanger sequencing can handle and then reassemble the reads based on overlapping nucleotide sequence. $\endgroup$ – canadianer Apr 6 '17 at 22:54

The question appears to confuse the method of identifying and determining the order of bases in a fragment of DNA — Sanger* sequencing, Maxam–Gilbert or more modern ‘next generation’ methods — with the strategy for selecting fragments to analyse and reassemble into the larger contiguous sequence from which they were derived, of which shotgun sequencing is one.

Shotgun sequencing is a strategy in which the selection of fragments to sequence is random, the method of sequencing is a matter of choice (it was often the Sanger method when it was first developed), and the assembly is done by computer programs that identify matching stretches. (It fails — or is not completely successful — if by chance the fragments do not cover the whole DNA with suitable overlaps or there are certain types of repeated regions.)

Shotgun sequencing should be contrasted with what might be called ordered sequencing strategies in which one generally starts with a restriction map of the whole DNA and initially selects specific fragments to clone and sequence (by whatever method seems best). On the basis of the sequence of the fragment one can choose to sequence further from a position within the already sequenced fragment, and this procedure repeated. Here there is no apparent assembly step as this is inherent in the choice of fragments to sequence. (This is much slower than shotgun sequencing, and cannot be automated so that it is much more expensive. However, it is more precise.)


Part of the confusion may arise from the fact that modern automated methods of sequencing genomes are designed to work on a mixture of many uncloned fragments of DNA, so inherently involve the shotgun sequencing strategy. They tend to employ technologies other than the classic Sanger method. If a ‘manual’ clone-based ordered sequencing strategy is required to ‘finish’ missing or ambiguous regions of a genome, this would most employ the Sanger method of determining the sequence. Hence, the contemporary writer might employ what is, in fact, a misleading contrast.

*Fred Sanger was an eminent Nobel laureate, so his surname is capitalized.

  • $\begingroup$ I hope this is clear. I can add a diagram if necessary. $\endgroup$ – David Apr 7 '17 at 8:02
  • $\begingroup$ This is clear! Thank you for explaining. Would it still be possible for you to provide a diagram as they usually drastically improve my understanding. $\endgroup$ – tryingtolearn Apr 16 '17 at 15:09

In Sanger sequencing we break up the DNA and use ddNTPs in order to determine the exact position of each neucleotide.

No. Sanger sequencing doesn't do this. The results of one Sanger sequencing reaction is a string of nucleotides, about 700-1200 bp long. There is nothing in that which inherently gives that string a position.

"Shotgun" just means that you aren't take advantage of any kind of known mapping information to target your sequencing reactions; you sequence everything randomly, and let a computer sort out what overlaps with what.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.