New variants of SARS-CoV-2 are being reported worldwide. More specifically, the variants reported in South Africa (501Y.V2) and United Kingdom (B.1.1.7) were found to be more contagious than the wild type. There are also speculations that the variants might evade the host immune responses induced by currently available vaccines and develop resistance to drugs under consideration. The first step of viral infection in COVID-19, occurs through the interaction of receptor binding domain (RBD) of the spike protein with peptidase domain of the human ACE-2 (hACE-2) receptor. So, possibly the mutations in the RBD domain of spike protein in the new variants could modulate the protein-protein interaction with hACE-2 receptor leading to the increased virulence. In this study, we aim to get molecular level understanding into the mechanism behind the increased infection rate due to such mutations in these variants. We have computationally studied the interaction of the spike protein in both wild-type and B.1.1.7 variant with hACE-2 receptor using combined molecular dynamics and binding free energy calculations using molecular mechanics-Generalized Born surface area (MM-GBSA) approach. The binding free energies computed using configurations from minimization run and low temperature simulation show that mutant variant of spike protein has increased binding affinity for hACE-2 receptor (i.e. ΔΔG(N501Y,A570D) is in the range −20.4 to −21.4 kcal/mol)The residue-wise decomposition analysis and intermolecular hydrogen bond analysis evidenced that the N501Y mutation has increased interaction between RBD of spike protein with ACE-2 receptor. We have also carried out calculations using density functional theory and the results evidenced the increased interaction between three pairs of residues (TYR449 (spike)-ASP38 (ACE-2), TYR453-HIE34 and TYR501-LYS353) in the variant that could be attributed to its increased virulence. The free energies of wild-type and mutant variants of the spike protein computed from MM-GBSA approach suggests that latter variant is stable by about −10.4 kcal/mol when compared to wild type suggesting that it will be retained in the evolution due to increased stability. We demonstrate that with the use of the state-of-the art of computational approaches, we can in advance predict the more virulent nature of variants of SARS-CoV-2 and alert the world health-care system.