What are the differences and similarities of the mapping and shotgun approaches?

Question

Mapping approach was carried out by the human genome project

Shotgun approach was carried out by the clera genomics

Answer 1

This is really difficult to answer without drawing figures, but here goes:

To start you must consider the context in which the two methods developed. The first, Ordered Clone Sequencing, was developed in the late 80's/ early 90's. The cost to sequence DNA was ~$3 to $5 per base (expensive!) and the best computers on the market were a fraction of the speed of today's machines.

Ordered Clone Sequencing, used by HuGo to map the human genome, is designed to minimize the amount of DNA sequenced. You start by building a library: take the DNA you want to sequence, randomly cut it into pieces, and ligate these pieces into vectors that can be stored in bacteria or yeast.  (see: http://homepages.strath.ac.uk/~dfs99109/... Now, you have millions of pieces of your DNA in your library. Because you “cut it into pieces” randomly, these pieces will overlap. You now need to try to find a series of overlapping pieces that together contain all of the DNA from your genome. I won’t detail how to do this, but trust me it sucks, and is very time consuming. In the end you have a bunch of fragments of DNA, that overlap, and represent your entire genome. Look here for a cartoon diagram: http://www.sanger.ac.uk/Projects/C_elega...

Now you sequence each piece of DNA. Because these pieces are fairly large, this takes awhile, years in fact, but in the end you will get the entire genome except for areas that are hard to sequence (highly repetitive areas like centromeres, etc.)

The Shotgun method was developed by this guy: http://en.wikipedia.org/wiki/Craig_Vente...

It was developed at a time when sequencing was cheap (less than a penny per base) and computers were a lot more powerful. You still start by building a library (just like above), but this time you skip the part where you line the pieces of the library up: you just start sequencing at random. Because you didn’t line up your clones, you will end up sequencing some areas two, three, four times or more times, but that’s ok, because sequencing is cheap. In the end you will have lot’s of DNA sequences, which you line up on computers.

This method is faster, and cheaper, but it has one major flaw: You don’t get the entire genome. Because you are sequencing your library at random, after awhile the unsequenced DNA in you library will get rarer and rarer (To envision this take a bag of black marbles. Pull one out and put a white one back in its place. Keep doing this. Eventually you the black marbles will be very hard to find.). In the end you will have most of the genome with a few small gaps. You can now go back and specifically find the sequences in your library that contain these gaps, and sequence them specifically (http://www.genome.ou.edu/gifs/GapClosure...