Identification of Clinically Relevant HIV Vif Protein Motif Mutations through Machine Learning and Undersampling.

Human Immunodeficiency virus (HIV) and its clinical entity, the Acquired Immunodeficiency Syndrome (AIDS) continue to represent an important health burden worldwide. Although great advances have been made towards determining the way viral genetic diversity affects clinical outcome, genetic association studies have been hindered by the complexity of their interactions with the human host. This study provides an innovative approach for the identification and analysis of epidemiological associations between HIV Viral Infectivity Factor (Vif) protein mutations and four clinical endpoints (Viral load and CD4 T cell numbers at time of both clinical debut and on historical follow-up of patients. Furthermore, this study highlights an alternative approach to the analysis of imbalanced datasets, where patients without specific mutations outnumber those with mutations. Imbalanced datasets are still a challenge hindering the development of classification algorithms through machine learning. This research deals with Decision Trees, Naïve Bayes (NB), Support Vector Machines (SVMs), and Artificial Neural Networks (ANNs). This paper proposes a new methodology considering an undersampling approach to deal with imbalanced datasets and introduces two novel and differing approaches (MAREV-1 and MAREV-2). As theses approaches do not involve human pre-determined and hypothesis-driven combinations of motifs having functional or clinical relevance, they provide a unique opportunity to discover novel complex motif combinations of interest. Moreover, the motif combinations found can be analyzed through traditional statistical approaches avoiding statistical corrections for multiple tests.

Mexican HIV-1 Protease Sequence Diversity.

Protease is one of three enzymes encoded within HIV's pol gene, responsible for the cleavage of viral Gag-Pol polypeptide into mature viral proteins and a target of current anti-retroviral therapy. Protease diversity analysis in Latin America has been lacking in spite of extensive studies of protease-inhibitor resistance mutations. We studied the diversity of 777 Mexican protease sequences and found that all were subtype B except one (CRF02_AG). Phylogenetic analysis suggested the existence of six different clades with geospecific contributions. Thirty-three percent of sites were conserved, 25% had conservative substitutions, and 41% exhibited physicochemical changes. The most conserved regions surrounded the active site, most of the flap domain, and a region between the 60's loop and C-terminal triad. A single sequence exhibited an active site mutation (T26S). Variable sites were mapped to a crystallographic structure, providing further insight into the distribution and functional relevance of variable sites among Mexican isolates.

Molecular Characterization of Mexican HIV-1 Vif Sequences.

The viral infectivity factor (Vif) is an HIV accessory protein that counteracts host antiviral proteins of the APOBEC3 family. Accumulating evidence highlights the pivotal role that accessory HIV proteins have on disease pathogenesis, a fact that has made them targets of interest for novel therapeutic and preventive strategies. Little is known about Vif sequence diversity outside of African or white populations. Mexico is home to Americas' third largest HIV-affected population and Mexican Hispanics represent an ever-increasing U.S. minority. This study provides a detailed analysis of the diversity seen in 77 Mexican Vif protein sequences. Phylogenetic analysis shows that most sequences cluster with HIV-1 subtype B, while less than 10% exhibit greater similarity to subtype D and A subtypes. Although most functional motifs are conserved among the Mexican sequences, substantial diversity was seen in some APOBEC binding sites, the nuclear localization inhibitory signal, and the CBFß interaction sites.