Multiscale mathematical models of hepatitis C infection have been instrumental in our understanding of direct acting antivirals. These models include the mechanisms driving intracellular viral production and explicitly model the intracellular concentration of viral RNA. Incorporating proliferation of infected hepatocytes in these models can be subtle, as infected daughter cells inherit viral RNA from the proliferating mother cell. In this note, we show how to incorporate this inheritance within a multiscale model of HCV infection. As in typical multiscale models of HCV infection, we show that this model is mathematically equivalent to a system of ordinary differential equations and perform bifurcation analysis of the resulting ODE that demonstrates that proliferation of infected hepatocytes can lead to infection persistence even if the basic repoductive number is less than one.

Large language models (LLMs) are powerful artificial intelligence (AI) tools transforming how research is conducted. However, their use in research has been met with skepticism, due to concerns about hallucinations, biases and potential harms to research. These emphasize the importance of clearly understanding the strengths and weaknesses of LLMs to ensure their effective and responsible use. Here, we present a roadmap for integrating LLMs into cross-disciplinary research, where effective communication, knowledge transfer and collaboration across diverse fields are essential but often challenging. We examine the capabilities and limitations of LLMs and provide a detailed computational biology case study (on modeling HIV rebound dynamics) demonstrating how iterative interactions with an LLM (ChatGPT) can facilitate interdisciplinary collaboration and research. We argue that LLMs are best used as augmentative tools within a human-in-the-loop framework. Looking forward, we envisage that the responsible use of LLMs will enhance innovative cross-disciplinary research and substantially accelerate scientific discoveries.

Abstract Introduction AZD5582 (AZD) is a SMAC mimetic that reverses latency by activating the non-canonical NF-κB pathway. AZD can trigger 3-4 logs transient viremia in ART-suppressed macaques; however, the reductions in the latent reservoir have been modest. Detailed analysis of the mechanisms is difficult to study due to low amplitude and fluctuating viral load signals. Methods We analyzed longitudinal data from 29 ART-suppressed, SIV-infected rhesus macaques treated with AZD5582 in combination with other agents (e.g., SIV antibodies, N-803, and/or MT807R1). We used a nonlinear mixed-effect modeling framework to fit an ensemble of mechanistic viral-dynamics models to the plasma SIV RNA and SIV CA-DNA measurements. We performed model selection using Bayesian information criteria and parameter identifiability checks, ultimately using the best models’ parameter estimates to infer the AZD5582 mode of action and treatment response mechanisms. Results The viral-dynamics model ensemble successfully reproduced the viral dynamics observed, specifically the AZD5582-associated increase in plasma SIV RNA and the changes in SIV DNA levels. Parameter estimates from the top-performing models consistently indicated that cells reactivated by AZD exhibited lower viral production and a significantly reduced death rate, ranging from 32 to 89% less than other infected cells. Conclusion Mechanistic modeling provides an explanation for the limited reduction in the viral reservoir despite the strong latency reversal achieved by AZD5582. The model revealed that reactivated cells behave differently from fully productively infected cells. This finding suggests that combining AZD5582 with other type of latency reversing treatment may improve the overall effectiveness and robustness of the drug. Funding Source R01 OD011095, R01 AI028433, R01 AI152703, U54 HL143541, UM1-AI164561, Los Alamos National Laboratory 20220791PRD2. Topic Categories Computational and Systems Immunology (COMP)

HIV infection accelerates biological aging, but the contribution of the host's age to this process is unknown. We investigated the influence of SIV infection in macaques (SIVmac) on the risk of comorbidities and aging in young and old rhesus macaques (RMs) by assessing pathogenesis markers, DNA methylation-based epigenetic age (EA), and EA acceleration (EAA) in blood and tissues. Initially, upon SIV infection, the young RMs showed greater resilience to CD4+ T cell depletion, better control of T cell activation, hypercoagulation, and excessive inflammation, yet this resilience was progressively lost in the advanced stages of infection. During the late stages of infection, the young RMs, but not the aged ones, showed an increase in EA in PBMCs; also, EAA in the cerebellum and heart of young RMs was higher compared with old RMs. SIV infection was more pathogenic in aged animals in early stages, leading to a more rapid disease progression; however, accelerated aging mostly affected young animals, so that the levels of multiple key pathogenesis markers in the young RMs converged toward those specific to aged ones in the late stages of infection. We conclude that SIV infection-driven age acceleration is tissue specific, and that host age influences the susceptibility of different tissues to enhanced aging.

BACKGROUND:The prolonged viral shedding from the gastrointestinal tract is well documented for numerous pathogens, including SARS-CoV-2. However, the impact of prolonged viral shedding on epidemiological inferences using wastewater data is not yet fully understood. METHODS:To gain a better understanding of this phenomenon at the population level, we extended a wastewater-based modeling framework that integrates viral shedding dynamics, viral load data in wastewater, case report data, and an epidemic model. RESULTS:Our results indicate that as an outbreak progresses, the viral load from recovered individuals gradually becomes predominant, surpassing that from the infectious population. This phenomenon leads to a dynamic relationship between model-inferred and reported daily incidence over the course of an outbreak. Sensitivity analyses on the duration and rate of viral shedding for recovered individuals reveal that accounting for this phenomenon can considerably advance prediction of transmission peak timing. Furthermore, extensive viral shedding from the recovered population toward the conclusion of an epidemic wave may overshadow viral signals from newly infected cases carrying emerging variants, which can delay the rapid recognition of emerging variants based on viral load. CONCLUSIONS:These findings highlight the necessity of integrating post-recovery viral shedding to enhance the accuracy and utility of wastewater-based epidemiological analysis.

Crimean-Congo hemorrhagic fever (CCHF) is a tick-borne disease endemic to many regions of Africa, the Middle East, Southeast Asia and the Balkans. Caused by the CCHF virus (CCHFV), CCHF has been a recognized cause of illness in Uganda since the 1950s and recently, more intensive surveillance suggests CCHFV is widely endemic within the country. Most surveillance has been focused on the Ugandan cattle corridor due to the risk of CCHFV exposure associated with livestock practices. Here we evaluated the seroprevalence of CCHFV in several Southern Ugandan communities outside the cattle corridor combined with longitudinal sample sets to measure the immune response to CCHFV for up to a decade. Interestingly, across three community types, agrarian, trading and fishing, we detected CCHFV seroprevalence in all three but found the highest seroprevalence in fishing communities. We also measured consistent CCHFV-specific antibody responses for up to a decade. Our findings support the conclusion that CCHFV is widely endemic in Uganda and highlight that additional communities may be at risk for CCHFV exposure.

The cornerstone of functional cure for chronic hepatitis B (CHB) is hepatitis B surface antigen (HBsAg) loss from blood. HBsAg is encoded by covalently closed circular DNA (cccDNA) and HBV DNA integrated into the host genome (iDNA). Nucleos(t)ide analogs (NUCs), the mainstay of CHB treatment, rarely lead to HBsAg loss, which we hypothesized was due to continued iDNA transcription despite decreased cccDNA transcription. To test this, we applied a multiplex droplet digital PCR that identifies the dominant source of HBsAg mRNAs to 3,436 single cells from paired liver biopsies obtained from 10 people with CHB and HIV receiving NUCs. With increased NUC duration, cells producing HBsAg mRNAs shifted their transcription from chiefly cccDNA to chiefly iDNA. This shift was due to both a reduction in the number of cccDNA-containing cells and diminished cccDNA-derived transcription per cell; furthermore, it correlated with reduced detection of proteins deriving from cccDNA but not iDNA. Despite this shift in the primary source of HBsAg, rare cells remained with detectable cccDNA-derived transcription, suggesting a source for maintaining the replication cycle. Functional cure must address both iDNA and residual cccDNA transcription. Further research is required to understand the significance of HBsAg when chiefly derived from iDNA.

Chronic hepatitis B virus (HBV) infection is strongly associated with increased risk of liver cancer and cirrhosis. While existing treatments effectively inhibit the HBV life cycle, viral rebound frequently occurs following treatment interruption. Consequently, functional cure rates of chronic HBV infection remain low and there is increased interest in a novel treatment modality, capsid assembly modulators (CAMs). Here, we develop a multiscale mathematical model of CAM treatment in chronic HBV infection. By fitting the model to participant data from a phase I trial of the first-generation CAM vebicorvir, we estimate the drug's dose-dependent effectiveness and identify the physiological mechanisms that drive the observed biphasic decline in HBV DNA and RNA, and mechanistic differences between HBeAg-positive and negative infection. Finally, we demonstrate analytically and numerically that the relative change of HBV RNA more accurately reflects the antiviral effectiveness of a CAM than the relative change in HBV DNA.

The human immunodeficiency virus (HIV) can persist in a latent form as integrated DNA (provirus) in resting CD4+ T cells unaffected by antiretroviral therapy. Despite being a major obstacle for eradication efforts, it remains unclear which infected cells survive, persist, and ultimately enter the long-lived reservoir. Here, we determine the genetic divergence and integration times of simian immunodeficiency virus (SIV) envelope sequences collected from infected macaques. We show that the proviral divergence and the phylogenetically estimated integration times display a biphasic decline over time. Investigating the dynamics of the mutational distributions, we show that SIV genomes in short-lived cells are, on average, more diverged, while long-lived cells contain less diverged virus. The change in the mutational distributions over time explains the observed biphasic decline in the divergence of the proviruses. This suggests that long-lived cells harbor viruses deposited earlier in infection, while short-lived cells predominantly harbor more recent viruses.

Human immunodeficiency virus (HIV) persistence during antiretroviral therapy (ART) is associated with heightened plasma interleukin-10 (IL-10) levels and PD-1 expression. We hypothesized that IL-10 and PD-1 blockade would lead to control of viral rebound following analytical treatment interruption (ATI). Twenty-eight ART-treated, simian immunodeficiency virus (SIV)mac239-infected rhesus macaques (RMs) were treated with anti-IL-10, anti-IL-10 plus anti-PD-1 (combo) or vehicle. ART was interrupted 12 weeks after introduction of immunotherapy. Durable control of viral rebound was observed in nine out of ten combo-treated RMs for >24 weeks post-ATI. Induction of inflammatory cytokines, proliferation of effector CD8(+) T cells in lymph nodes and reduced expression of BCL-2 in CD4(+) T cells pre-ATI predicted control of viral rebound. Twenty-four weeks post-ATI, lower viral load was associated with higher frequencies of memory T cells expressing TCF-1 and of SIV-specific CD4(+) and CD8(+) T cells in blood and lymph nodes of combo-treated RMs. These results map a path to achieve long-lasting control of HIV and/or SIV following discontinuation of ART.

BACKGROUND:Hepatitis C virus (HCV) infects nearly one-fourth of people with human immunodeficiency virus (HIV). The role of direct-acting antivirals (DAAs) on immune activation in people with HIV (PWH) and HCV is poorly understood. METHODS:We quantified plasma HCV RNA and CXCL10 in persons with HCV monoinfection versus HIV/HCV coinfection receiving sofosbuvir-velpatasvir. Single-cell laser capture microdissection was applied to liver biopsies obtained before and within 4-7 days of DAA initiation to estimate HCV clearance and changes in interferon-stimulated genes (ISGs). RESULTS:We enrolled 10 people with chronic genotype 1a HCV: 5 were PWH with ART-suppressed viremia and CD4+ T cell counts >200 cells/µL. First- and second-phase plasma HCV RNA kinetics were not different between groups. Median (min-max) proportions of infected hepatocytes at biopsy 1 were 0.06 (0.01-0.59) in HCV monoinfection and 0.21 (0.04-0.87) in HIV/HCV coinfection and did not differ. Participants had lower intracellular HCV RNA levels at biopsy 2. CXCL10 levels declined in both groups but were higher in coinfection than in monoinfection even at the end of treatment. The proportion of cells expressing ISGs diminished in monoinfection but increased in coinfection. CONCLUSIONS:Whereas DAAs rapidly cleared intrahepatic HCV in both groups, immune activation was slower to diminish in PWH. Residual immune activation in PWH warrants further exploration. Clinical Trials Registration. NCT02938013.

Introduction:Vaccine platforms used in successful, licensed vaccines have varied among pathogens. However, antibody level is still the main clinical correlate of protection in most approved vaccines. Decisions as to the best vaccine platform to pursue for a given pathogen may be informed through improved understanding of the process of antibody generation and its temporal dynamics, as well as the relationship between these processes and the type of vaccine. Methods:We have analyzed the dynamics of antibody generation for different vaccine platforms against diverse pathogens, and developed a consensus mathematical model that captures antibody dynamics across these diverse systems. Initially, the model was fitted to a rich dataset of antibody and immune cell concentrations in a SARS-CoV-2 vaccine experiment. We then used concepts from machine learning, such as transfer learning, to apply the same model to a variety of systems, involving different pathogens, vaccine platforms, and booster dose use/timing, fixing most parameter values relating to the dynamics of the immune system. Results:The model includes B cell proliferation and differentiation, as well as the generation of plasma cells, which secrete large amounts of antibody, and memory B cells. Overall, the model describes antibody generation in all systems tested well and shows that the main differences across platforms are related to the dynamics of antigen presentation. Discussion:This model can be used to predict antibody generation in pairs of vaccine platform/pathogen, allowing for the use of in silico results to narrow down experimental burden in vaccine development.

Studying the early events that occur after viral infection in humans is difficult unless one intentionally infects volunteers in a human challenge study. Here, we use data about severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in such a study in combination with mathematical modeling to gain insights into the relationship between the amount of virus in the upper respiratory tract and the immune response it generates. We propose a set of dynamic models of increasing complexity to dissect the roles of target cell limitation, innate immunity, and adaptive immunity in determining the observed viral kinetics. We introduce an approach for modeling the effect of humoral immunity that describes a decline in infectious virus after immune activation. We fit our models to viral load and infectious titer data from all the untreated infected participants in the study simultaneously. We found that a power-law with a power h < 1 describes the relationship between infectious virus and viral load. Viral replication at the early stage of infection is rapid, with a doubling time of ~2 h for viral RNA and ~3 h for infectious virus. We estimate that adaptive immunity is initiated ~7 to 10 d postinfection and appears to contribute to a multiphasic viral decline experienced by some participants; the viral rebound experienced by other participants is consistent with a decline in the interferon response. Altogether, we quantified the kinetics of SARS-CoV-2 infection, shedding light on the early dynamics of the virus and the potential role of innate and adaptive immunity in promoting viral decline during infection.

Most people living with HIV-1 experience rapid viral rebound once antiretroviral therapy is interrupted; however, a small fraction remain in viral remission for an extended duration. Understanding the factors that determine whether viral rebound is likely after treatment interruption can enable the development of optimal treatment regimens and therapeutic interventions to potentially achieve a functional cure for HIV-1. We built upon the theoretical framework proposed by Conway and Perelson to construct dynamic models of virus-immune interactions to study factors that influence viral rebound dynamics. We evaluated these models using viral load data from 24 individuals following antiretroviral therapy interruption. The best-performing model accurately captures the heterogeneity of viral dynamics and highlights the importance of the effector cell expansion rate. Our results show that post-treatment controllers and non-controllers can be distinguished based on the effector cell expansion rate in our models. Furthermore, these results demonstrate the potential of using dynamic models incorporating an effector cell response to understand early viral rebound dynamics post-antiretroviral therapy interruption.

Abstract We present an ensemble transfer learning method to predict suicide from Veterans Affairs (VA) electronic medical records (EMR). A diverse set of base models was trained to predict a binary outcome constructed from reported suicide, suicide attempt, and overdose diagnoses with varying choices of study design and prediction methodology. Each model used twenty cross-sectional and 190 longitudinal variables observed in eight time intervals covering 7.5 years prior to the time of prediction. Ensembles of seven base models were created and fine-tuned with ten variables expected to change with study design and outcome definition in order to predict suicide and combined outcome in a prospective cohort. The ensemble models achieved c-statistics of 0.73 on 2-year suicide risk and 0.83 on the combined outcome when predicting on a prospective cohort of $$\sim$$ ∼ 4.2 M veterans. The ensembles rely on nonlinear base models trained using a matched retrospective nested case-control (Rcc) study cohort and show good calibration across a diversity of subgroups, including risk strata, age, sex, race, and level of healthcare utilization. In addition, a linear Rcc base model provided a rich set of biological predictors, including indicators of suicide, substance use disorder, mental health diagnoses and treatments, hypoxia and vascular damage, and demographics.

Vaccines have historically played a pivotal role in controlling epidemics. Effective vaccines for viruses causing significant human disease, e.g., Ebola, Lassa fever, or Crimean Congo hemorrhagic fever virus, would be invaluable to public health strategies and counter-measure development missions. Here, we propose coverage metrics to quantify vaccine-induced CD8+ T cell-mediated immune protection, as well as metrics to characterize immuno-dominant epitopes, in light of human genetic heterogeneity and viral evolution. Proof-of-principle of our approach and methods are demonstrated for Ebola virus, SARS-CoV-2, and Burkholderia pseudomallei (vaccine) proteins.

We consider stochastic models of individual infected cells. The reproduction number, R, is understood as a random variable representing the number of new cells infected by one initial infected cell in an otherwise susceptible (target cell) population. Variability in R results partly from heterogeneity in the viral burst size (the number of viral progeny generated from an infected cell during its lifetime), which depends on the distribution of cellular lifetimes and on the mechanism of virion release. We analyse viral dynamics models with an eclipse phase: the period of time after a cell is infected but before it is capable of releasing virions. The duration of the eclipse, or the subsequent infectious, phase is non-exponential, but composed of stages. We derive the probability distribution of the reproduction number for these viral dynamics models, and show it is a negative binomial distribution in the case of constant viral release from infectious cells, and under the assumption of an excess of target cells. In a deterministic model, the ultimate in-host establishment or extinction of the viral infection depends entirely on whether the mean reproduction number is greater than, or less than, one, respectively. Here, the probability of extinction is determined by the probability distribution of R, not simply its mean value. In particular, we show that in some cases the probability of infection is not an increasing function of the mean reproduction number.

In a subset of SARS-CoV-2-infected individuals treated with the antiviral nirmatrelvir-ritonavir, the virus rebounds following treatment. The mechanisms driving this rebound are not well understood. We used a mathematical model to describe the longitudinal viral load dynamics of 51 individuals treated with nirmatrelvir-ritonavir, 20 of whom rebounded. Target cell preservation, either by a robust innate immune response or initiation of N-R near the time of symptom onset, coupled with incomplete viral clearance, appears to be the main factor leading to viral rebound. Moreover, the occurrence of viral rebound is likely influenced by the time of treatment initiation relative to the progression of the infection, with earlier treatments leading to a higher chance of rebound. A comparison with an untreated cohort suggests that early treatments with nirmatrelvir-ritonavir may be associated with a delay in the onset of an adaptive immune response. Nevertheless, our model demonstrates that extending the course of nirmatrelvir-ritonavir treatment to a 10-day regimen may greatly diminish the chance of rebound in people with mild-to-moderate COVID-19 and who are at high risk of progression to severe disease. Altogether, our results suggest that in some individuals, a standard 5-day course of nirmatrelvir-ritonavir starting around the time of symptom onset may not completely eliminate the virus. Thus, after treatment ends, the virus can rebound if an effective adaptive immune response has not fully developed. These findings on the role of target cell preservation and incomplete viral clearance also offer a possible explanation for viral rebounds following other antiviral treatments for SARS-CoV-2. IMPORTANCE:Nirmatrelvir-ritonavir is an effective treatment for SARS-CoV-2. In a subset of individuals treated with nirmatrelvir-ritonavir, the initial reduction in viral load is followed by viral rebound once treatment is stopped. We show that the timing of treatment initiation with nirmatrelvir-ritonavir may influence the risk of viral rebound. Nirmatrelvir-ritonavir stops viral growth and preserves target cells but may not lead to full clearance of the virus. Thus, once treatment ends, if an effective adaptive immune response has not adequately developed, the remaining virus can lead to rebound. Our results provide insights into the mechanisms of rebound and can help develop better treatment strategies to minimize this possibility.

Mathematical models of viral infection have been developed, fitted to data, and provide insight into disease pathogenesis for multiple agents that cause chronic infection, including HIV, hepatitis C, and B virus. However, for agents that cause acute infections or during the acute stage of agents that cause chronic infections, viral load data are often collected after symptoms develop, usually around or after the peak viral load. Consequently, we frequently lack data in the initial phase of viral growth, i.e., when pre-symptomatic transmission events occur. Missing data may make estimating the time of infection, the infectious period, and parameters in viral dynamic models, such as the cell infection rate, difficult. However, having extra information, such as the average time to peak viral load, may improve the robustness of the estimation. Here, we evaluated the robustness of estimates of key model parameters when viral load data prior to the viral load peak is missing, when we know the values of some parameters and/or the time from infection to peak viral load. Although estimates of the time of infection are sensitive to the quality and amount of available data, particularly pre-peak, other parameters important in understanding disease pathogenesis, such as the loss rate of infected cells, are less sensitive. Viral infectivity and the viral production rate are key parameters affecting the robustness of data fits. Fixing their values to literature values can help estimate the remaining model parameters when pre-peak data is missing or limited. We find a lack of data in the pre-peak growth phase underestimates the time to peak viral load by several days, leading to a shorter predicted growth phase. On the other hand, knowing the time of infection (e.g., from epidemiological data) and fixing it results in good estimates of dynamical parameters even in the absence of early data. While we provide ways to approximate model parameters in the absence of early viral load data, our results also suggest that these data, when available, are needed to estimate model parameters more precisely.