Hi,

I do not dephosphorylate the cut vector.  Instead, I cut both the vector and the insert with the appropriate restriction enzymes and gel-purify both the cut vector and the insert.  The cut plasmid will migrate more slowly on the gel than any uncut plasmid.  Then I run both on an agarose gel alongside a DNA mass Ladder and estimate [DNA]  by comparing the intensity of the bands produced by vector and insert with the intensity of the bands on the DNA mass ladder.  Then to set up a DNA ligation use a 3:1 molar excess of insert to vector.

Terry

It is strange that you get positive result in colony PCR and negative in sequencing.

1. Did you get a single amplicon band corresponding to the size of your insert in colony PCR?

2.To further check you can also do a minipreparation of DNA from the positive colonies and digest it with the restriction enzyme to check the release of desired sized insert. If both way you get the same positive result then send them for sequencing.  

3. I didn't understand If you are using directional cloning, where you use two different restriction enzyme site in the multiple cloning sites of the vector, then they would result in non-compatible ends after digestion so vector re-ligation could be reduced. But I do have seen where 2 such vectors re-ligate and give false results. Dephosphorylation should further avoid such chances of getting false positive clones. Also as suggested by Terry, using 1:3 molar ratio of vector:insert, all these enhances your chance of getting positive clones. 

Hope it helps!