Analysis of drug resistance in HIV protease

Abstract

BackgroundDrug resistance in HIV is the major problem limiting effective antiviral therapy. Computational techniques for predicting drug resistance profiles from genomic data can accelerate the appropriate choice of therapy. These techniques can also be used to select protease mutants for experimental studies of resistance and thereby assist in the development of next-generation therapies.ResultsThe machine learning produced highly accurate and robust classification of HIV protease resistance. Genotype data were mapped to the enzyme structure and encoded using Delaunay triangulation. Generative machine learning models trained on one inhibitor could classify resistance from other inhibitors with varying levels of accuracy. Generally, the accuracy was best when the inhibitors were chemically similar.ConclusionsRestricted Boltzmann Machines are an effective machine learning tool for classification of genomic and structural data. They can also be used to compare resistance profiles of different protease inhibitors.

Evolution of drug resistance in HIV protease

“…Vectors were generated for each sequence by obtaining the neighbors of each position of this sequence from the Delaunay triangulation as was done in [ 7 – 9 , 11 , 15 , 17 ]. The coordinates of the carbon atoms were used, and all arcs of the triangulation were used.…”

“…Our previous work [ 7 – 9 , 11 , 15 , 17 ] has concentrated on developing models for predicting the resistance to single inhibitors. Shen et al [ 11 ] and Pawar et al [ 17 ] demonstrated classification accuracies higher than 99%. However, many of the resistant strains have lost susceptibility to all clinical inhibitors.…”

“…It is important, therefore, to apply machine learning to the prediction of resistance to multiple inhibitors. Our previous work [ 17 ] showed that there was significant cross-prediction accuracy where models trained on one inhibitor predict the response to other inhibitors. This suggests that there are commonalities in resistance mechanisms and the first step to studying these commonalities is to build a machine learning model that describes them.…”

Epidemiological data analysis of viral quasispecies in the next-generation sequencing era

“…As a result, these drug-resistant haplotypes will begin to dominate over time [102,113]. The second problem is that drug resistance is connected with haplotypes rather than with the mutations themselves, but haplotypes are harder to detect and so the drug resistance analysis can be significantly improved with more sensitive haplotyping tools [114].…”

<strong>Comprehensive Interactome Development Technique</strong>

“…The hypothetical examination of cooperations relies upon heterogeneous wellsprings of organic data, for example, succession and basic information bases, the writing, and exploratory information. The principle data sets containing trial data about proteinprotein collaborations are: the Database of Interacting Proteins (DIP) [17], the Biomolecular Interaction Network Database (BIND) [18], the Molecular Interaction Database (MINT) [19], INTACT [20,21], and Human Protein Reference Database (HPRD) [22]. The writing information on chose protein successions is accessible from the iHOP [23] and STRING [24] data sets.…”