In this talk I will discuss how to exploit observations of transmitted flux close by high-redshift quasars in order to observe the physical state of the intergalactic medium (IGM) at redhsift higher than six. The physical interpretation of such spectral features has been strongly debated during the last few years. Here I will present a semi-analytic criterion which -for the first time- allows to robustly establish whether the window of transmitted flux close by a given quasar is the counterpart of its HII region in frequency space or it is analogous to the proximity effect observed at lower redshift.
Both these two scenarios offer the possibility to explore the IGM state at such high redshifts: the first case allows one to infer reliable estimates of the mean IGM ionization, while the second permits to detect the presence of biased galaxy-clustering in the vicinity of luminous quasars.
I will then present results from the application of the above method to real spectra.