the bidirected bruijn graph for genome assembly

[Yele]

Hi guys,

I was reading again the 5.2 epigraph of the "Computability of Models
for Sequence Assembly" paper, and they put an example in the figure 5.
(page 11). They gave a k-molecule spectrum and they construct the
bidirected de Bruijn graph, I have a doubt with the construction of
this graph, specially when inserting the nodes and establishing the
edges directions, so that the graph will be balanced.

Can someone who has fully understood this example explain it to me?
(I don't really want to keep this doubt until next class)

Regards,

Yele.

[Michael Brudno]

On the graph on the right, the nodes are the possible 2-long sequences
of DNA, the edges the 3-long ones that join the 2-long ones together.
The directionality is as explained in class, you have to "balance"
arrows when you walk through every node. Perhaps you can ask a more
precise question, and I'll be glad to explain more...