Combining degree centrality and betweenness centrality of molecular networks can effectively pinpoint individuals at high risk of HIV transmission within the network

Abstract

Abstract Background. HIV molecular network technology can identify HIV transmission hotspots and individuals at risk of HIV transmission, facilitating precise and targeted interventions. This study explored the molecular network parameters, namely degree centrality (DC) and betweenness centrality (BC), to effectively pinpoint individuals at high risk of HIV transmission within the network. Methods. HIV pol gene and demographic and clinical information were collected from a whole-population sampling cohort comprising all newly diagnosed people living with HIV (PLWH) in Shenyang, from 2016 to 2019. Molecular networks were constructed, and three groups were identified based on DC and BC: high DC group (>the median DC), high BC group (>the mean BC), and high DC+BC group (both >the median DC and >the mean BC). The average risk of HIV transmission was calculated by dividing the total probability of recent HIV infections (identified by HIV-1 LAg-Avidity EIA) by the number of cases in each group. A multivariate logistic regression analysis was conducted to identify the characteristics of men who have sex with men (MSM) in the three groups. Results. Of the 2882 PLWH, 1162 were included in the molecular network. The median DC and the mean BC of all nodes were 3 (range: 1-28) and 0.09 (range: 0-1), respectively. The average risk of HIV transmission was highest in the high DC+BC group (n=92) at 0.49, followed by the high BC group (n=201) at 0.37, and the high DC group (n=169) at 0.29. The characteristics of MSM with high DC+BC overlapped with those of MSM with high DC (low education levels, P=0.031) and MSM with high BC (unemployed, P=0.027 and high baseline viral load[≥105copies/mL], P=0.003). Conclusion. The combined utilization of DC and BC can effectively identify individuals at high risk of HIV transmission, enabling precisely targeted interventions using molecular network technology.