Optimization of combined anti-retroviral therapy (cART) for human immunodeficiency virus (HIV) infection -guided by genotype/phenotype testing for potential virus drug resistance- has been established and recommended in the past decade by international guidelines, to assist clinicians in selecting first-line and subsequent regimens. Several drug resistance prediction algorithms for the nucleotide/nucleoside/non-nucleoside reverse transcriptase, and protease inhibitors are currently available, routinely used, and have been validated thoroughly in independent studies. Computer tools that combine single-drug genotypic/phenotypic resistance interpretation and optimize cART have also been developed and implemented as web applications. In this work, we aim at assessing the state-of-the-art in computational prediction models that incorporate new HIV drugs (e.g. including etravirine, rilpivirine, darunavir), new HIV drug targets (e.g. CCR5-antagonists, fusion and integrase inhibitors), extension to hepatitis drugs, and at evaluating how they have been integrated in computer tools that optimize cART and predict virological response.