Phylogenetic methods inconsistently predict direction of HIV transmission among heterosexual pairs in the HPTN052 cohort.
Rose R., Hall M., Redd AD., Lamers S., Barbier AE., Porcella SF., Hudelson SE., Piwowar-Manning E., McCauley M., Gamble T., Wilson EA., Kumwenda J., Hosseinipour MC., Hakim JG., Kumarasamy N., Chariyalertsak S., Pilotto JH., Grinsztejn B., Mills LA., Makhema J., Santos BR., Chen YQ., Quinn TC., Fraser C., Cohen MS., Eshleman SH., Laeyendecker O.
Background:We evaluated use of phylogenetic methods to predict the direction of HIV transmission. Methods:For 33 index-partner pairs with genetically-linked infection, samples were collected from partners and indexes close to time of partners' seroconversion (SC); 31 indexes also had an earlier sample. Phylogenies were inferred using env next-generation sequences (one tree per pair/subtype). Direction of transmission (DoT) predicted from each tree was classified as correct or incorrect based on which sequences (index or partner) were closest to the root. DoT was also assessed using maximum-parsimony to infer ancestral node states for 100 bootstrap trees. Results:DoT was predicted correctly for both single pair and subtype-specific trees in 22 pairs (67%) using SC samples and 23 pairs (74%) using early index samples. DoT was predicted incorrectly for four pairs (15%) using SC or early index samples. In the bootstrap analysis, DoT was predicted correctly for 18 pairs (55%) using SC samples and 24 pairs (73%) using early index samples. DoT was predicted incorrectly for seven pairs (21%) using SC samples and four pairs (13%) using early index samples. Conclusions:Phylogenetic methods based solely on tree topology of HIV env sequences, particularly without consideration of phylogenetic uncertainty, may be insufficient for determining DoT.