Genes, trees and infections: Molecular evidence in epidemiology

Deep Learning and Likelihood Approaches for Viral Phylogeography Converge on the Same Answers Whether the Inference Model Is Right or Wrong

Analysis of phylogenetic trees has become an essential tool in epidemiology. Likelihood-based methods fit models to phylogenies to draw inferences about the phylodynamics and history of viral transmission. However, these methods are often computationally expensive, which limits the complexity and realism of phylodynamic models and makes them ill-suited for informing policy decisions in real-time during rapidly developing outbreaks. Likelihood-free methods using deep learning are pushing the boundaries of inference beyond these constraints. In this paper, we extend, compare, and contrast a recently developed deep learning method for likelihood-free inference from trees. We trained multiple deep neural networks using phylogenies from simulated outbreaks that spread among 5 locations and found they achieve close to the same levels of accuracy as Bayesian inference under the true simulation model. We compared robustness to model misspecification of a trained neural network to that of a Bayesian method. We found that both models had comparable performance, converging on similar biases. We also implemented a method of uncertainty quantification called conformalized quantile regression that we demonstrate has similar patterns of sensitivity to model misspecification as Bayesian highest posterior density (HPD) and greatly overlap with HPDs, but have lower precision (more conservative). Finally, we trained and tested a neural network against phylogeographic data from a recent study of the SARS-Cov-2 pandemic in Europe and obtained similar estimates of region-specific epidemiological parameters and the location of the common ancestor in Europe. Along with being as accurate and robust as likelihood-based methods, our trained neural networks are on average over 3 orders of magnitude faster after training. Our results support the notion that neural networks can be trained with simulated data to accurately mimic the good and bad statistical properties of the likelihood functions of generative phylogenetic models.

HIV-1 subtype diversity and transmission strain source among men who have sex with men in Guangxi, China

With the rapid increase in HIV prevalence of men who have sex with men (MSM) in recent years and common human migration and travelling across different provinces in China, MSM are now finding it easier to meet each other, which might contribute to local HIV epidemics as well as fueling cross-province transmission. We performed a cross-sectional survey in 2018-2019 to investigate the current HIV subtype diversity and inferred HIV strain transmission origin among MSM in Guangxi province, China based on a phylogenetic analysis. Based on 238 samples, we found that the HIV-1 subtype diversity was more complicated than before, except for three major HIV subtypes/circulating recombinant forms (CRFs): CRF07_BC, CRF01_AE, CRF55_01B, five other subtypes/CRFs (CRF59_01B, B, CRF08_BC, CRF67_01B, CRF68_01B) and five unique recombinant forms (URFs) were detected. In total, 76.8% (169/220) of samples were infected with HIV from local circulating strains, while others originated from other provinces, predominantly Guangdong and Shanghai. The high diversity of HIV recombinants and complicated HIV transmission sources in Guangxi MSM indicates that there has been an active sexual network between HIV positive MSM both within and outside Guangxi without any effective prevention. Inter-province collaboration must be enforced to provide tailored HIV prevention and control services to MSM in China.

Patterns of Bird-Bacteria Associations

Birds, with their broad geographic ranges and close association with humans, have historically played an important role as carriers of human disease and as reservoirs for drug-resistant bacteria. Here, we examine scientific literature over a 15-year timespan to identify reported avian-bacterial associations and factors that may impact zoonotic disease emergence by classifying traits of bird species and their bacteria. We find that the majority of wild birds studied were migratory, in temperate habitats, and in the order Passeriformes. The highest diversity of bacteria was found on birds in natural habitats. The most frequently reported bacteria were Escherichia coli, Salmonella enterica, and Campylobacter jejuni. Of the bacteria species reported, 54% have shown pathogenicity toward humans. Percentage-wise, more pathogens were found in tropical (vs. temperate) habitats and natural (vs. suburban, urban, or agricultural) habitats. Yet, only 22% were tested for antibiotic resistance, and of those tested, 75% of bacteria species were resistant to at least one antibiotic. There were no significant patterns of antibiotic resistance in migratory versus non-migratory birds, temperate versus tropical areas, or different habitats. We discuss biases in detection and representation, and suggest a need for increased sampling in non-temperate zones and in a wider range of avian species.

Emerging infectious disease: what are the relative roles of ecology and evolution?

The increasing threat of infectious diseases in humans has renewed interest in factors leading to the emergence of new diseases and the re-emergence of familiar diseases. Examples of seemingly novel diseases currently spreading in human populations include HIV, dengue hemorrhagic fever and Lyme disease; drug-resistant forms of well-known diseases such as tuberculosis are also increasing. The problem of disease emergence also extends to other animal and plant populations. In most current epidemics, ecological factors (e.g. migration, climate, agricultural practices) play a more significant role in disease emergence than evolutionary changes in pathogens or hosts. Evolutionary biologists and ecologists have much to offer to the development of strategies for the control of emerging diseases.