Expression from a Gateway vector requires two cloning steps:
First, a PCR amplification product is cloned into an entry vector (pENTR or pDONR) using BP clonase.
Secondly, the entry construct is incubated with a destination vector, and recombination is facilitated by LR clonase.
The first step usually involves PCR, the error rate of which is about 1 in 10^3 or so. As such, it is customary to sequence the entirety of whatever is being cloned into the entry vector.
On the other hand, the second step (the LR reaction) does not involve in vitro DNA synthesis. Usually, I just check that the sequence of the vector-insert junction is correct. I typically do not sequence the entirety of the insert a second time, especially if it is a big insert.
What do other people think?