A simple structure-based model for the prediction of HIV-1 co-receptor tropism

Abstract

BackgroundHuman Immunodeficiency Virus 1 enters host cells through interaction of its V3 loop (which is part of the gp120 protein) with the host cell receptor CD4 and one of two co-receptors, namely CCR5 or CXCR4. Entry inhibitors binding the CCR5 co-receptor can prevent viral entry. As these drugs are only available for CCR5-using viruses, accurate prediction of this so-called co-receptor tropism is important in order to ensure an effective personalized therapy. With the development of next-generation sequencing technologies, it is now possible to sequence representative subpopulations of the viral quasispecies.ResultsHere we present T-CUP 2.0, a model for predicting co-receptor tropism. Based on our recently published T-CUP model, we developed a more accurate and even faster solution. Similarly to its predecessor, T-CUP 2.0 models co-receptor tropism using information of the electrostatic potential and hydrophobicity of V3-loops. However, extracting this information from a simplified structural vacuum-model leads to more accurate and faster predictions. The area-under-the-ROC-curve (AUC) achieved with T-CUP 2.0 on the training set is 0.968±0.005 in a leave-one-patient-out cross-validation. When applied to an independent dataset, T-CUP 2.0 has an improved prediction accuracy of around 3% when compared to the original T-CUP.ConclusionsWe found that it is possible to model co-receptor tropism in HIV-1 based on a simplified structure-based model of the V3 loop. In this way, genotypic prediction of co-receptor tropism is very accurate, fast and can be applied to large datasets derived from next-generation sequencing technologies. The reduced complexity of the electrostatic modeling makes T-CUP 2.0 independent from third-party software, making it easy to install and use.

Comparative analyses of error handling strategies for next-generation sequencing in precision medicine

“…A recent study suggests that lowering the threshold of resistance testing might come with a reduction of specificity in HIV-1 resistance testing, but increases the identification of people at risk of virological failure, i.e., sensitivity 20 . So far there are cases where patients experienced therapy failure during a co-receptor antagonist, while the X4 ratio was beneath the threshold of phenotypic testing 11 . Depending on the applied error-strategy the threshold for a decision might be exceeded, e.g., by using the worst-case-scenario for low X4-ratio, which could then, in turn, lead to the wrong treatment decision.…”

“…We used the R package seqinr 21 to clean the raw data concerning ambiguities, leading to a dataset of 3,580 R5-and 214 X4-strains (see Table 1). The sequences were translated into protein sequences using seqinr and subsequently predicted with T-CUP 11 . Only those sequences that were correctly predicted (R5: 3, 395, i.e., specificity 94 83% = .…”

“…Several prediction tools are available, e.g., geno2pheno[coreceptor] 10 , T-CUP 11 , or PhenoSeq 12 . In order to avoid biases with respect to specific unknown sequences that have been used for training of existing tools, we used our tool T-CUP as baseline prediction, since it shows the best performance in comparison to other tools, and delivers a pseudo-probability 11 , where the training data is publicly available 13 and which has been demonstrated to be reliable also for NGS data 14 . In contrast to existing tools such as geno2pheno[coreceptor] 10 or PhenoSeq 12 , T-CUP offers high-throughput analyses in an automated manner, and is thus, also applicable to the huge amounts of sequences used in the current study.…”

Learning the Relationship between the Primary Structure of HIV Envelope Glycoproteins and Neutralization Activity of Particular Antibodies by Using Artificial Neural Networks

“…A simplified structure-based model of the V3 loop is used in [6] to model co-receptor tropism in HIV-1. Moreover, the incorporation of ENV determinants outside the V3 loop is demonstrated to be able to improve the reliability of co-receptor usage [7].…”

gCUP: rapid GPU-based HIV-1 co-receptor usage prediction for next-generation sequencing

“…This model is based on our recently developed method T-CUP (Dybowski et al, 2010a;Heider et al, 2014), but was redeveloped, parallelized and optimized for the use on graphics processing units (GPUs). In Heider et al (2014), we demonstrate the accuracy of our T-CUP model for the prediction of HIV-1 tropism compared with other available models. On an independent dataset, T-CUP outperformed Geno2Pheno as well as the model of Bozek et al (2013).…”