We examine the flavor dependence of the nuclear modification factor R AA in the pQCD calculations at LHC energies. The computations are performed accounting for radiative and collisional parton energy loss with running coupling constant. Our results show that the recent LHC data on the R AA for charged hadrons, D-mesons and non-photonic electrons agree reasonably with the pQCD picture of the parton energy loss with the dominating contribution from the radiative mechanism.1. The parton energy loss in the quark-gluon plasma (QGP) is widely believed to be a source for strong suppression of high-p T hadrons in AA-collisions (usually called the jet quenching) observed at RHIC and LHC. Understanding the underlying mechanisms of the parton energy loss is of great importance for application of the jet quenching to probing the hot QCD matter produced in AA-collisions. In the pQCD picture fast partons lose energy mostly due to induced gluon radiation [1,2,3,4,5,6]. The effect of collisional energy loss [7] for the RHIC and LHC conditions is likely to be relatively small [8,9]. Unfortunately, uncertainties in the pQCD-based models of the jet quenching remain large (mostly due to difficulties in modeling multiple gluon emission). For the nuclear modification factor R AA they are perhaps about a factor two. Despite this, it seems relatively safe to assume that predictions for variation of the R AA should be more robust, if the parameters are already adjusted to fit some set of experimental data.From the point of view of the underlying physics of the jet quenching it is very interesting to compare R AA for light and heavy flavors. It was suggested [10] that for the heavy quarks the dead cone effect should suppress induced gluon emission and give rise to an increase of the R AA . However, the observed at RHIC strong suppression of the nonphotonic electrons from the B/D−meson decays [11,12,13] seemed to be in contradiction with this picture. It may indicate that for RHIC conditions the dead cone suppression is not very strong or that the radiative mechanism is not the dominating one at all. It stimulated the renewed interest in the collisional energy loss [14]. Although, by adjusting the coupling constant one can obtain a sufficiently strong heavy quark suppression due to the collisional mechanism alone, this scenario does not seem to be realistic (at least for p T ∼ > 5 − 10 GeV). Calculations of the radiative and collisional energy losses with the 1 same α s and the Debye screening mass performed in [8] clearly demonstrate that the collisional loss is relatively small for relativistic partons and unlikely to change significantly the heavy quark energy loss (see also [9]). In [10] the dead cone suppression was estimated from a qualitative analysis neglecting the quantum finite-size effects. Calculations of the induced gluon emission from heavy quarks in a brick of QGP [15] within the light-cone path integral (LCPI) approach [2], which treats accurately the mass effects, demonstrate that at energy ∼ 10 − 20 GeV for c-qu...