ABSTRACT
For humans, the COVID-19 pandemic and Coronavirus have undeniably been a nightmare. Although there are effective vaccines, specific drugs are still urgent. Normally, to identify potential drugs, one needs to design and then test interactions between the drug and the virus in an in silico manner for determining candidates. This Drug-Target Interaction (DTI) process, can be done by molecular docking, which is too complicated and time-consuming for manual works. Therefore, it opens room for applying Artificial Intelligence (AI) techniques. In particular, Graph Neural Network (GNN) attracts recent attention since its high suitability for the nature of drug compounds and virus proteins. However, to introduce such a representation well-reflecting biological structures of biological compounds is not a trivial task. Moreover, since available datasets of Coronavirus are still not highly popular, the recently developed GNNs have been suffering from overfitting on them. We then address those issues by proposing a novel model known as Atom-enhanced Graph Neural Network with Multi-hop Gating Mechanism. On one hand, our model can learn more precise features of compounds and proteins. On the other hand, we introduce a new gating mechanism to create better atom representation from non-neighbor information. Once applying transfer learning from very large databanks, our model enjoys promising performance, especially when experimenting with Coronavirus. © 2022 IEEE.
ABSTRACT
Background: Haiphong is a Vietnamese city of 2 million people and a historic hotspot for HIV and drug use. The DRIVE community research program recently demonstrated the end of the HIV epidemic among PWID in the city, with an incidence of 0.085/100PY, and a substantial decrease in HCV incidence in the past 5 years. After the emergence of COVID-19, a one-month strict lockdown was imposed in April 2020 in Vietnam, followed by lighter social distancing restrictions over the year. We investigated whether those measures affected PWID in terms of risk behaviors and access to prevention and care. Methods: Participants were PWID that had been enrolled in a respondent-driven sampling (RDS) survey as part of DRIVE in the last quarter of 2019. They were recalled and interviewed in the last quarter of 2020 by peer educators on their socioeconomic situation, drug use and sexual behaviors, relations to methadone maintenance treatment (MMT) and ART services. They were tested for drugs and methadone in the urine, and for HIV, HCV, and HIV plasma viral load when HIV(+). Changes following the restrictions were assessed by comparing these "after" data to the "before" data collected one year earlier during the RDS survey. In-depth interviews were conducted with 30 participants including 5 female sex workers (FSW). Results: 780 PWID were enrolled. Their mean age was 44 years and 94% were male. 56% were still actively injecting (100% heroin) at the time of the interview;their monthly consumption had decreased from 24 to 17 days on average. The main source of syringes remained pharmacies for 83% before, during and after the lockdown. The proportion of PWID still engaging in sharing decreased from 6.0 to 1.5%. No change in the frequency of condom use was reported. The proportion of PWID on MMT increased from 68.7 to 75.3%. There was no alteration in the HIV cascade of care that was still above 90/90/90. No HIV seroconversion was observed, and HCV incidence remained stable (2.6/100PY, 95%CI: 0.7-6.7). 53% reported a monthly income of less than 130 USD "after" compared to only 9% "before". One FSW reported accepting unsafe sex during the lockdown due to financial pressure. Conclusion: Six months after the beginning of COVID-19-related restrictions, access to harm reduction materials and care services for PWID was maintained and no increase in the number of new HIV or HCV infections was observed. However, this period was a major financial challenge, especially for FSW that were more likely to engage in risky sexual behaviors.