Efficiently Predicting HIV-1 Protease Cleavage Sites by Using Deep CNN-Assisted Hybridised Approach

Abstract

Acquired Immuno Deficiency Syndrome (AIDS) is considered a vital
danger over the feasible development and on account of its pestilence
effect and nonattendance of reparable medicines. HIV-1 AIDS can be
constrained by utilising protease inhibitors. Different procedures for
predicting sites are utilised to comprehend the highlighted sites of those
proteases. Arrangement-based sites like physiochemical elements and
construction-based sites are separated from HIV-1 proteases. In this
article, a procedure for choosing those sites using a deep CNN-assisted
hybridised approach will be used for effectively predicting cleavage sites.
The proposed methodology was evaluated based on various Type-1 and
Type-2 parameters. The proposed approach gives superior results on
Type-1 and Type-2 parameters. Data746_setset provides an accuracy of
0.924, data 1625_set provides an accuracy of 0.946, Data_schilling_set
provides an accuracy of 0.9389, Data_impens_set provides an accuracy

of 0.911 and average dataset provides an accuracy of 0.921 for Type-
1 parameters. Data746_setset provides an accuracy of 0.84664, data

1625_set provides an accuracy of 0.34179, data_schilling_set provides
an accuracy of 0.69529 and data_impens_set provides an accuracy of
0.59511 for Type-2 parameters.

Saini S, Kaur G. Efficiently Predicting HIV-1

Protease Cleavage Sites by Using Deep CNN-
Assisted Hybridised Approach. J Commun Dis.

2025;57(1):91-95.

DOI: https://doi.org/10.24321/0019.5138.202511