The experiment I perform is just subsequent additions of DNA over a solution of the complex. then, I don´t really know if I have to consider the absorption bands of pi-pi* transitions (320 nm)
Record the statring material alone; titrate the complex solution with DNA solution; record UV spectra at different DNA concentrations; chose the band that changes most.
Using the most intense band of the UV-Vis spectrum will lead you, as it has been suggested, to more accurate reading.
In any case the metal you are using may absorb in the UV (i.e. Copper) causing the band in this region to be due to both DNA and metal. In this case you can either just choose the visible band for your experiment or carry out (if you have a double beam instrument) a differential UV-vis titration.
Thank you very much for your answers! I was a bit confused as I thought that in order to check whether there was interaction with DNA I had to choose a specific absorption band of my metal complex. So far I had been using the most intense one as it was the one that changes the most and I could clearly see the changes. However , I read lately that some authors only consider the one where the metal is involved, which normally it is one of the less intense bands and it is quite difficult to appreciate the changes. So, I´ll keep choosing the one that changes the most regardless of its origin. Thanks!