Table 1.

Comparison of the molecular based approaches for determination of HIV recency.

Assay or Test Type or Recent Infection Algorithm	Brief Summary of the Assay or Test	Strengths	Limitations	References
High Resolution Melting Assay	Measures diversity of generated amplicons by using melting temperature of DNA duplexes.	Relatively inexpensive Does not require viral genotyping	PCR may be poor scalable in rural and resource constrained settings. Sensitive to indels that are common in HIV. Unable to distinguish between infections caused by single or multiple viral strains.	[69,70,71]
Counting Sequence Ambiguities	Ambiguous bases in the viral sequence indicate heterogeneous virus population. The number of ambiguous bases is small in recently infected individuals and increased overtime.	Can be cost effective if routine drug resistance testing is a part of clinical care	Not cost effective in resource limited settings. Can underestimate ambiguous positions. The impact of ART exposure is unknown and could be a serious concern. The number of ambiguous positions in the late stage of HIV infection could be reduced, which may lead to misclassifications.	[18,19,72,73]
Naïve Bayes Classifier (NBC)	Utilized the frequency of ambiguous sites together with CD4+ cell counts and any concurrent AIDS defining illness. The Bayesian probability framework estimates the probability of a patient to be in one of four stages of HIV infection.	High positive predictive value Can be retrospectively fitted to available genotypic and clinical data	Requires substantial validation. The method has been applied only once in HIV-1 subtype B settings.	[74]
Hamming Distance (HD)	The HD is a number that denotes the difference between two sequences of equal length. It is the simplest measure of HIV diversity. HD can measure the number of nucleotide differences between a pair of virus sequences. If applied to viral quasispecies from the host, HD can estimate the stage of HIV infection.	High sensitivity and specificity Simplicity	The HD approach has not been validated in long-term non-progressors, rapid progressors, and among ART-experienced individuals. It is unclear how indels and viral recombination can affect the HD estimates. Requires viral quasispecies. May have limited use in the resource-constrained settings.	[21,75]
Sequence Clustering Based Diversity Measure (SCBD)	Intra-cluster genetic diversity is used as the measure of time since infection. Inter-cluster diversity is used to determine whether there were multiple founder strains and the dot matrix incorporates information on indels and recombination.	Good accuracy High sensitivity and specificity	It is unclear how indels and viral recombination can affect the estimates. Is time consuming and expensive.	[76]
Multi-Assay Algorithms (MMA)	Results of serology-based test of recent infection combined with: Clinical data (e.g., CD4+ cell counts, HIV-1 RNA load, ART status) Measure of HIV diversity Combination of Assays can be optimized to increase accuracy.	Provides more accurate estimate of the HIV recency Reduces false recency	Not validated across HIV-1 subtypes and different populations. Requires clinical data (e.g., CD4+ cell counts, viral load count). Might be problematic logistically in resource-limited settings.	[69,77,78]