A comparison of two open-source methods for molecular phylogenetic analysis of Human Immunodeficiency Virus (HIV): Hypothesis testing using Phylogenetics (HyPhy) and Molecular Evolutionary Genetics Analysis (MEGA)

We compared two widely used methods for Human Immunodeficiency Virus (HIV) molecular phylogenetic analysis (MPA) - Hypothesis testing using Phylogenetics (HyPhy) and Molecular Evolutionary Genetics Analysis (MEGA) – to consider which could strengthen surveillance and better target prevention interventions in Queensland, Australia.

HIV pol sequences generated for drug resistance testing were linked to de-identified case reports in the state-wide register of notified HIV cases. The study used MEGA, 6.0 patristic distance ≤1.5% and HyPhy, patristic distance ≤2% to identify molecular transmission clusters. We compared time taken to perform each analysis; the effectiveness of identifying sequences that clustered; the efficiency of identifying clusters and their sizes; and the ability simply to illustrate cluster networks and their evolution over time.

Of 1776 unique sequences identified 1,563 (88.5%) were linked to a notification record. Analysis with HyPhy (30 minutes) was 600 times faster than MEGA (324 hours). With HyPhy, 1084 (61.4%) sequences clustered compared to MEGA where 595 (33.7%) clustered making HyPhy 54% (595/1084) more effective. Overall, HyPhy identified 82 more transmission clusters than MEGA (266 versus 184) performing 45% (82/184) more efficiently. In size terms, HyPhy found 565 sequences clustered in 50 moderate or large clusters with MEGA finding 261 sequences in 21 moderate and 2 large clusters: of these clusters, 43 were identified by both methods but with MEGA nearly half (20; 47%) were small. HyPhy also more efficiently established cluster size. The HyPhy network cluster maps can more simply illustrate molecular transmission clusters, include patient characteristics, show their timelines and are easier to update than the cluttered MEGA circular phylogenetic trees.

We are confident that HyPhy is better suited than MEGA to use in the Queensland context to generate comprehensive network transmission cluster maps for monitoring HIV transmission and to translate near real-time phylogeny data into action for targeted prevention interventions.