The human immunodeficiency virus type 1 (HIV-1) has continued to be a global concern. With the new HIV incidence, the emergence of multi-drug resistance and the untoward side effects of currently used HIV protease inhibitors (PIs), there is an urgent need to discover more efficient HIV PIs. Modern computational tools have played vital roles in facilitating drug discovery process. This research focuses on pharmacophore-based similarity search to screen 111,566,735 unique compounds in the PubChem database to discover novel HIV-1 PIs. We used in-silico approach involving 3D-similarity search, physicochemical and ADMET evaluations, HIV protease-inhibitor prediction (IC₅₀/percent inhibition), rigid receptor-molecular docking studies and free-binding energy calculations. The 10 FDA approved HIV PIs (saquinavir, lopinavir, ritonavir, amprenavir, fosamprenavir, atazanavir, nelfinavir, darunavir, tipranavir and indinavir) were used as reference. The in-silico analysis revealed that fourteen out of the twenty-eight selected optimized hit molecules were within the acceptable range of all the parameters investigated. The hit molecules demonstrated significant binding affinity to the HIV protease (PR) when compared to the reference drugs with the residues ASP25, GLY27, ASP29, ASP30, ILE50 involved in essential hydrogen bonding and п-п stacked interactions, which stabilize the optimized hit molecules in the active binding site of the HIV-1 PR (PDB:2Q5K). HPS/002 and HPS004 are most promising in terms of IC₅₀/percent inhibition (90.15%) of HIV-1 PR, in addition to their drug metabolism and safety profile. These hit candidates should be investigated further as possible HIV-1 PIs with improved efficacy and low toxicity through in-vitro experiments and clinical trial investigation.