Touchdown PCR is commonly used to increase the sensitivity of primers that do not work optimally: you run several cycles at a lower Tm, and then increase the Tm to the theoretically optimal. Following this strategy, you will get more unspecific amplicons, but you will luckily be able to detect your specific band.

I can't figure out why you would need that when amplifying a recombinant vector.

Maybe you can specify more about your experiment and aims.

As an example, if your primers have a tm of 60ºC, you could add 10-20 cycles of tm 55ºC, and then another 25-30 cycles at 60ºC.

Dear Arif

take a look at this paper

• https://www.researchgate.net/messages/attachment/922916_Touch%20down%20PCR.pdf

In essence you

• commence PCR with 1 cycle @96C for 3-5 min
• Carry out the TD phase by using the standard 3 step PCR but in each cycle reduce the annealing temp by 0.5C per cycle starting at Tm +3C for highest primer and progressing down to Tm-3C for lowest primer in 0.5C intervals 
• This would equate to 12 cycles if primers shared an identical Tm
• Now perform a standard 3 step PCR for a further 25 cycles at Tm-3C; that is the lowest annealing temp for the TD phase
• However when extending perform at 68C instead of 72C to minimise damage to DNA because of protracted cycling conditions 
• In addition during this last phase denature at 94C thereagain to minimise damage to DNA 
• Finally perform 1 cycle of 72C for 30 min then take down to 4C

 Thank you for your explanation