Genotypic predictors of human immunodeficiency virus type 1 drug resistance

Abstract

Understanding the genetic basis of HIV-1 drug resistance is essential to developing new antiretroviral drugs and optimizing the use of existing drugs. This understanding, however, is hampered by the large numbers of mutation patterns associated with crossresistance within each antiretroviral drug class. We used five statistical learning methods (decision trees, neural networks, support vector regression, least-squares regression, and least angle regression) to relate HIV-1 protease and reverse transcriptase mutations to in vitro susceptibility to 16 antiretroviral drugs. Learning methods were trained and tested on a public data set of genotype-phenotype correlations by 5-fold cross-validation. For each learning method, four mutation sets were used as input features: a complete set of all mutations in >2 sequences in the data set, the 30 most common data set mutations, an expert panel mutation set, and a set of nonpolymorphic treatment-selected mutations from a public database linking protease and reverse transcriptase sequences to antiretroviral drug exposure. The nonpolymorphic treatment-selected mutations led to the best predictions: 80.1% accuracy at classifying sequences as susceptible, low͞intermediate resistant, or highly resistant. Least angle regression predicted susceptibility significantly better than other methods when using the complete set of mutations. The three regression methods provided consistent estimates of the quantitative effect of mutations on drug susceptibility, identifying nearly all previously reported genotype-phenotype associations and providing strong statistical support for many new associations. Mutation regression coefficients showed that, within a drug class, crossresistance patterns differ for different mutation subsets and that cross-resistance has been underestimated. antiviral therapy ͉ HIV ͉ linear regression ͉ machine learning T wenty antiretroviral drugs are approved for treating HIV-1 infection: eight protease inhibitors (PIs), seven nucleoside and one nucleotide reverse transcriptase (RT) inhibitors (NRTIs), three nonnucleoside RT inhibitors (NNRTIs), and one fusion inhibitor. Resistance to these drugs is caused by mutations in their molecular targets. Understanding the genetic basis of cross-resistance is essential for designing new antiviral drugs and for using genotypic drug resistance testing to select optimal therapy. Despite the large number of PIs and RT inhibitors, therapy is challenging because drug resistance arises from complex patterns of mutations and because of the high degree of cross-resistance within each drug class.Approaches for using HIV-1 drug resistance mutations to predict changes in drug susceptibility have included decision trees (1), linear regression (2), linear discriminant analysis (3), neural networks (4), and support vector regression (SVR) (5). Here, we compare five statistical learning methods each using four different sets of input mutations to develop quantitative models associating HIV-1 protease and RT mutations with changes in susce...

Detecting and understanding combinatorial mutation patterns responsible for HIV drug resistance

“…Among all the published methods (7,8,23,24) that are related to our study, the one by Haq et al (24) is most closely related. They attempted to achieve a similar goal by using similar datasets.…”

“…Most published works (7,8,23) attempted to predict phenotype (e.g., fold change) from genotype by using genotype-phenotype data from Stanford HIVdb. The phenotype data, unfortunately, were measured in vitro.…”

Point Mutations Associated with HIV-1 Drug Resistance, Evasion of the Immune Response and AIDS Pathogenesis

“…There is a body of evidence suggesting that K65R mutation occurs more commonly in low-income countries when patients with nonsubtype B HIV-1 strains are treated with d4T/ddI and d4T/3TC Hawkins et al, 2009) or TDF/3TC (Rey et al, 2009). In comparison to K65R, L74V point mutation causes intermediate resistance to ddI and ABC, and a slight increase in susceptibility to AZT and TDF (Rhee et al, 2006b). L74V and K65R mutations rarely occur in the same viruses (Shafer & Schapiro, 2008).…”

“…Each of the primary NNRTI resistance mutations, namely, K103N/S; V106A/M; Y181C/I/V, Y188L/C/H and G190A/S/E ( Figure 3 and Table 3) cause high level resistance to nevirapine (NVP) and variable resistance to efavirenz (EFV), ranging from about 2-fold for V106A and Y181C; 6-fold for G190A; 20-fold for K103N; and more than 50-fold for Y188L and G190S (Bacheler et al, 2001;Rhee et al, 2006b). Transient virologic responses to EFV-based salvage therapy regimen occur in some NNRTIexperienced patients but a sustained response is not common (Antinori et al, 2002;Delaugerre et al, 2001;Shulman et al, 2000;Walmsley et al, 2001).…”

Learning from Past Treatments and Their Outcome Improves Prediction of In Vivo Response to Anti-HIV Therapy

“…This topic is similar in a sense that it uses genotypic information as features and the target is the prediction of binary or real-valued response values. Various statistical learning methods have been applied in this area, including linear regression (Rabinowitz et al, 2006;Rhee et al, 2006;Saigo et al, 2007), decision trees (Beerenwinkel et al, 2002), support vector machines (SVMs) (Beerenwinkel et al, 2003;Sing et al, 2005;Sing and Beerenwinkel, 2007), artificial neural networks (Wang and Larder, 2003;Larder, 2007), bayesian networks (Deforche et al, 2008) and Markov models DeGruttola, 2002, 2003).…”