PNAS Nexus

A novel coronavirus (COVID-19) was identified in Wuhan, Hubei Province, China, in December 2019 and has caused over 40,000 cases worldwide to date. Previous studies have supported an epidemiological hypothesis that cold and dry (low absolute humidity) environments facilitate the survival and spread of droplet-mediated viral diseases, and warm and humid (high absolute humidity) environments see attenuated viral transmission (i.e., influenza). How-ever, the role of absolute humidity in transmission of COVID-19 has not yet been established. Here, we examine province-level variability of the basic reproductive numbers of COVID-19 across China and find that changes in weather alone (i.e., increase of temperature and humidity as spring and summer months arrive in the North Hemisphere) will not necessarily lead to declines in COVID-19 case counts without the implementation of extensive public health interventions.

Are COVID-19 fatalities large when a federal government does not enforce containment policies and instead allow states to implement their own policies? We answer this question by developing a stochastic extension of a SIRD epidemiological model for a country composed of multiple states. Our model allows for interstate mobility. We consider three policies: mask mandates, stay-at-home orders, and interstate travel bans. We fit our model to daily U.S. state-level COVID-19 death counts and exploit our estimates to produce various policy counterfactuals. While the restrictions imposed by some states inhibited a significant number of virus deaths, we find that more than two-thirds of U.S. COVID-19 deaths could have been prevented by late November 2020 had the federal government enforced federal mandates as early as some of the earliest states did. Our results quantify the benefits of early actions by a federal government for the containment of a pandemic.

Precise regulation of ion channel biophysics is an essential life process that governs electrical signaling in excitable tissues. Many ion channels including voltage-gated Na+ channels (NaVs) exist in the membrane as clusters, which show distinct biophysical behavior not predicted by single-channel measurements. In both heterologous and native systems, we report that single-channel-based predictions significantly overestimated Na+ current (INa) amplitudes from multi-channel clusters. Computational modeling suggested that these observations could reflect interactions between adjacent channels, such as recently reported between NaVs, and identified specific biophysical consequences thereof. This updated model not only accurately predicted behaviors observed from NaV clusters and consequent cellular physiology, but also suggested the possibility that clustered NaVs may respond differently to use-dependent pharmacological agents. Experiments validated the latter prediction and further identified modulation of clustering as a novel approach to correcting macroscopic electrophysiological dysfunction resulting from NaV defects linked to life-threatening arrhythmias and seizures. Thus, our study not only motivates a fundamental revision of how ion channels behave when clustered but also highlights resulting biophysical effects as important considerations for pharmacology and a potential therapeutic target to address human disease.

Contagion happens through heterogeneous interpersonal relations (homophily) which induce contamination clusters. Group testing is increasingly recognized as necessary to fight the asymptomatic transmission of the COVID-19. Still, it is plagued by false negatives. Homophily can be taken into account to design test pools that encompass potential contamination clusters. I show that this makes it possible to overcome the usual information-theoretic limits of group testing, which are based on an implicit homogeneity assumption. Even more interestingly, a multiple-step testing strategy combining this approach with advanced complementary exams for all individuals in pools identified as positive identifies asymptomatic carriers who would be missed even by costly exhaustive individual tests. Recent advances in group testing have brought large gains in efficiency, but within the bounds of the above cited information-theoretic limits, and without tackling the false negatives issue which is crucial for COVID-19. Homophily has been considered in the contagion literature already, but not in order to improve group testing.

Over a life-course, human adaptive immunity to antigenically mutable pathogens exhibits competitive and facilitative interactions. We hypothesize that such interactions may lead to cyclic dynamics in immune responses over a lifetime. Here, we demonstrated a long-term periodicity (about 24 years) in individual antibody responses, by analyzing hemagglutination inhibition titers against 21 historical influenza A(H3N2) strains spanning 47 years from a cohort in Guangzhou, China. The reported cycles were robust to analytic and sampling approaches. Simulations suggested that individual-level cross-reaction between antigenically similar strains likely explain the reported cycle. We showed that the reported cycles are predictable at both individual and birth-cohort level and that cohorts show a diversity of phases of these cycles. Phase of cycle was associated with the risk of response to circulating strains, after accounting for age and pre-existing titers of the circulating strains. Our findings reveal the existence of long-term periodicities in individual antibody responses to A(H3N2). We hypothesize that these cycles are driven by pre-existing antibody responses blunting responses to antigenically similar pathogens (by preventing infection and/or robust antibody responses upon infection), leading to reductions in antigen specific responses over time until individuals increasing risk leads to an infection with an antigenically distant enough virus to generate a robust immune response. These findings could help disentangle cohort-effects from individual-level exposure histories, improve our understanding of observed heterogeneous antibody responses to immunizations, and inform targeted vaccine strategy.

Sharp increases in COVID-19 cases occurred after reopening in the United States. We show that the post-intervention effective reproduction number is a strong predictor of the surge in late June. Lax interventions in the early stages coupled with elevated virus spread are primarily responsible for surges in most affected states.

East, South, and Southeast Asia (together referred to as Southeastern Asia hereafter) have been recognized as critical areas fuelling the global circulation of seasonal influenza. However, the internal migration network of seasonal influenza within Southeastern Asia remains unclear, including how pandemic-related disruptions altered the network structure and circulation dynamics in this region. Here, we leveraged genetic, epidemiological, and airline travel data between 2007-2023 to characterise the multiyear dispersal patterns of influenza A/H3N2 and B/Victoria viruses both out of and within Southeastern Asia, including during seasons marked by perturbations such as the 2009 A/H1N1 and COVID-19 pandemics. We show consistent Autumn-Winter movement waves of A/H3N2 and B/Victoria from Southeastern Asia to temperate regions during interpandemic seasons. During the COVID-19 pandemic this trend was interrupted for both subtypes, however the A/H1N1 pandemic only disrupted A/H3N2 spread. For influenza strains circulating in Southeastern Asia, we find a higher persistence of A/H3N2 than B/Victoria. We find pandemic-related disruptions in A/H3N2 antigenic evolution, with a greater time-advanced antigenic evolution during the 2009 A/H1N1 pandemic, and a greater time-lagged pattern during the COVID-19 pandemic, compared to inter-pandemic levels. Internally, in comparison to the interpandemic seasons, the inferred dispersal rates within Southeastern Asia decreased by 54.7% and 79.2% during the 2009 A/H1N1 and COVID-19 pandemic seasons, respectively; further, the internal movement structure within Southeastern Asia markedly diverged during the COVID-19 pandemic season, and to a lesser extent, during the 2009 A/H1N1 pandemic season. Analyses of the trunk location and phylogenetic similarity further reveal a temporally varying pattern within Southeastern Asia, suggesting a complex source-sink network, with a notable decrease in the mixing of lineages around the COVID-19 pandemic season. Our findings provide insights into the heterogeneous interplay between influenza circulation in Southeastern Asia and two distinct pandemic-related disruptions (strong decline in human movements during the COVID-19 pandemic, pronounced pathogen interference during the A/H1N1 pandemic), which can help anticipate the effects of potential mitigation strategies and the emergence of future influenza pandemic strains on influenza dynamics.

Little attention has been paid to governances impacts on the evolution of SARS-CoV-2, the virus that causes COVID-19. To evaluate such impacts on the evolution of vaccine resistance, we analyzed a stochastic compartmental model to quantify the risk a mutant strain capable of evading immunity emerges post-vaccine rollout. We calibrated the model with publicly available data for four territories in the Western Hemisphere qualitatively differing in pandemic interventions. The model shows an immune-evading strain to be readily selected over all infectivities in Texas. In Panama, only a high level of transmission permits immune evasion to evolve. No invasion appears likely in Costa Rica and Uruguay. Programs combining pharmaceutical and nonpharmaceutical interventions are best positioned to remove the epidemiological space SARS-CoV-2 needs to evolve vaccine resistance. One Sentence SummaryModes of governance and production help set the evolutionary trajectories of vaccine resistance in SARS-CoV-2 before vaccine campaigns begin.

Non-structural carbon compounds (NSCs) serve to buffer short-term imbalances between carbon supply and demand in trees; however, their seasonal dynamics throughout the entire tree remain inadequately understood. We quantified year-round non-structural carbohydrate storage and fluxes in a temperate pine forest by integrating monthly measurements of soluble sugars, starch, and lipids across five tissues with biometric scaling to ecosystem stocks. Soluble sugars were consistently highest in canopy tissues and maintained a relatively stable concentration, even as sugar fluxes exhibited pronounced seasonal variations and reversals. In contrast, starch showed clear seasonality, increasing during the mid-growing season and decreasing later, whereas lipid pools remained relatively stable and contributed minimally to short-term fluctuations. Ecosystem-scale analyses indicated that sugars predominantly contributed to NSC turnover, accounting for approximately 80% of the total annual flux, while stored pools exhibited slower changes. The net annual NSC flux, approximately 65 g C m-2 yr-1, was relatively modest in comparison to biomass production, which totaled around 522g C m-2 yr -1. These findings indicate that seasonal changes in carbon balance are primarily driven by rapid redistribution of soluble carbon rather than by significant changes in overall NSC storage.

We study the impact of college reopening in Fall 2020 on county-level COVID-19 cases and deaths using the information of 1,076 randomly chosen four- and two-year undergraduate degree-granting colleges from the National Center for Education Statistics. These institutions include public, private nonprofit, and for profit schools from 50 US states and the District of Columbia. We match college and county characteristics using several methods and calculate the average treatment effects of three teaching modalities: in-person, online, and hybrid on COVID-19 outcomes up to two months after college reopening. In pairwise comparison, colleges reopened with in-person teaching mode were found to have about 35 percentage points more cases within 15 days of reopening, compared to those that reopened online, and the gap widens over time at a decreasing rate. Death rates follow the pattern with a time lag. However, colleges with hybrid mode reach up to the rates of in-person mode after some time. We also find that greater endowment and student population, bigger class size, and fewer Republican voters in the county are major predictors of choosing remote teaching modes over in-person.

Stable isotopes of mammoths and mastodons have the potential to illuminate ecological changes in late Pleistocene landscapes and megafaunal populations as these species approached extinction. The ecological factors at play in this extinction remain unresolved, but isotopes of bone collagen ({delta}13C, {delta}15N) and tooth enamel ({delta}13C, {delta}18O, 87Sr/86Sr) from the Midwest, USA are leveraged to examine ecological and behavioral changes that occurred during the last interglacial-glacial cycle. Both species had significant C3 contributions to their diets and experienced increasing levels of niche overlap as they approached extinction. A subset of mastodons after the last glacial maximum (LGM) exhibit low {delta}15N values that may represent expansion into a novel ecological niche, perhaps densely occupied by other herbivores. Stable isotopes from serial and micro-sampled enamel show increasing seasonality and decreasing temperatures as mammoths transitioned from Marine Isotope Stage (MIS) 5e to glacial conditions (MIS 4, MIS 3, MIS 2). Isotopic variability in enamel suggests mobility patterns and life histories have potentially large impacts on the interpretation of their stable isotope ecology. This study further refines the ecology of midwestern mammoths and mastodons demonstrating increasing seasonality and niche overlap as they responded to landscape changes in the final millennia before extinction.

The subtype of the first influenza infection shapes infection risk, illness and mortality associated with subsequent infection subtypes, as well as vaccine efficacy. This phenomenon, termed immune imprinting, may result from immune recognition rooted in structural similarity of the hemagglutinin protein within subtypes of influenza viruses. However, it is difficult to isolate from cohort effects resulting from the annual dominance of specific influenza subtypes. We used a long-term cohort study in southern China to examine whether hemagglutination inhibition responses show evidence of imprinting, and explore whether such patterns could emerge in the absence of imprinting using simulations. Our statistical analysis reveals patterns consistent with immune imprinting. However, simulations showed that similar patterns can emerge without an imprinting mechanism, meaning these may partly be due to age-related variations in immune responses driven by other factors. This work lays the foundations for further research into immune imprinting while accounting for cohort effects.

This paper introduces an innovative method for fine-tuning a larger multi-label model for abnormality detection, utilizing a smaller trainer and advanced knowledge distillation techniques. It delves into the effects of fine-tuning on various abnormalities, noting varied improvements based on the Original Models performance in specific tasks. The experimental setup, optimized for on-device inference and fine-tuning with limited computational resources, demonstrates moderate yet promising enhancements in model performance post-fine-tuning. Key insights from the study include the importance of aligning the {micro}-Trainers behavior with the Original Model and the influence of hyper-parameters like the batch size on fine-tuning outcomes. The research acknowledges limitations such as the limited exploration of loss functions in multi-label models and constraints in architectural design, suggesting potential avenues for future investigation. While the proposed Naive Continual Fine-tuning Process is in its early stages, it highlights the potential for long-term model personalization. Moreover, using weight transfer exclusively for fine-tuning amplifies user privacy protection through on-device fine-tuning, devoid of transferring data or gradients to the server. Despite modest performance improvements after fine-tuning, these layers represent a small fraction (0.7%) of the total weights in the Original Model and 1.6% in the {micro}-Trainer. This study establishes a foundational framework for advancing personalized model adaptation, on-device inference, and fine-tuning while emphasizing the importance of safeguarding data privacy in model development.

Understanding the cellular and physiological mechanisms underlying muscle remodeling requires model systems that allow rapid, reliable, and quantitative assessment of muscle state. The cricket Gryllus lineaticeps naturally undergoes non-pathological striated muscle breakdown (histolysis), making it a promising system for studying this process. However, current assessments of muscle state are largely qualitative, subjective, and poorly standardized across experiments. Here, we developed and validated a continuous, quantitative muscle color metric to objectively capture histolysis progression and functional changes in muscle. We show that this metric robustly tracks variation in muscle color across remodeling stages, including the challenging fully transparent stage, and strongly predicts protein content, mitochondrial abundance, and iron content in a muscle- and trait-specific manner. The reproducibility of these relationships across independent datasets demonstrates the generality and robustness of this approach. By providing a rapid, objective, and biologically informative proxy of muscle state, this framework not only advances the utility of G. lineaticeps as a model for muscle remodeling but also offers a strategy for exploring the cellular dynamics underlying age-related muscle diseases and disorders, addressing an increasing public health concern in aging populations.

The pump-leak mechanism (PLM) first, described by Tosteson and Hoffman (1960), demonstrates how the activity of the Na+ - K+ ATPase (NKA) can counteract the osmotic influx of water stimulated by the presence of impermeant intracellular molecules. We derive analytical solutions for the steady state ion concentrations, voltage, and volume of a cell, by including impermeant extracellular molecules, variable impermeant charge, and Cation-Chloride Co-transporters (CCC). We demonstrate that impermeant extracellular molecules could stabilize a cell without NKA activity but argue that it is unlikely to play a significant role in vivo. Significantly we have shown that the precise form of the NKA is unimportant for determining the steady state in PLMs. We have derived an analytical expression for the steady state of the PLM with one of the Cation-Chloride Co-transporters, either KCC, NCC, or NKCC, active. Notably, we have demonstrated that NCC at high pump rates can destabilize cells, which could account for the rarity of this co-transporter. In addition, we show that the reversal of any of the CCCs is unlikely. Importantly, we link the thermodynamics of the NKA to the PLM to show that there is a natural limit to the energy utilized by the PLM that prevents futile cycles. We show that the average charge on the intracellular impermeant molecules influences ion distributions but has no impact on energy utilization. Our study shows that analytical mathematical solutions from physically well-grounded models provide insight into ion transport systems that could only be obtained from numerical simulations with great difficulty. Significance StatementThe regulation of cell volume is fundamental to the stability of all tissue. Animal cells regulate their volume by actively pumping sodium and potassium ions, preventing the waters osmotic influx from blowing up the cell. Based on the physical laws that determine ion and water fluxes, we derive equations that allow one to predict how pump rates and ion conductances combine to stabilize cell volume. The action of the sodium pump consumes about 30% of a cells energy budget, and we demonstrate the rate of ion pumping is constrained so that cells do not consume excessive energy. Our work also demonstrates the power of closed-form mathematical equations in characterizing such pump-leak systems.

The emergence of successive SARS-CoV-2 variants of concern (VOC) during 2020-22, each exhibiting increased epidemic growth relative to earlier circulating variants, has created a need to understand the drivers of such growth. However, both pathogen biology and changing host characteristics - such as varying levels of immunity - can combine to influence replication and transmission of SARS-CoV-2 within and between hosts. Disentangling the role of variant and host in individual-level viral shedding of VOCs is essential to inform COVID-19 planning and response, and interpret past epidemic trends. Using data from a prospective observational cohort study of healthy adult volunteers undergoing weekly occupational health PCR screening, we developed a Bayesian hierarchical model to reconstruct individual-level viral kinetics and estimate how different factors shaped viral dynamics, measured by PCR cycle threshold (Ct) values over time. Jointly accounting for both inter-individual variation in Ct values and complex host characteristics - such as vaccination status, exposure history and age - we found that age and number of prior exposures had a strong influence on peak viral replication. Older individuals and those who had at least five prior antigen exposures to vaccination and/or infection typically had much lower levels of shedding. Moreover, we found evidence of a correlation between the speed of early shedding and duration of incubation period when comparing different VOCs and age groups. Our findings illustrate the value of linking information on participant characteristics, symptom profile and infecting variant with prospective PCR sampling, and the importance of accounting for increasingly complex population exposure landscapes when analysing the viral kinetics of VOCs.

The vast majority of colleges and universities across the United States are bringing students back for in-person instruction in the midst of the COVID-19 pandemic, despite the absence of an effective vaccine or other anti-viral therapeutic treatment. Using data from the New York Times and the American Community Survey, we assess the effect of this return to campus on viral case growth in counties with a significant college student population ("college counties") relative to non-college counties. We find a significant surge in new cases in a 21-day time frame in college counties, a finding consistent across U.S. Census divisions. These results suggest the need for institutions of higher education and the communities where these institutions reside work together quickly and effectively to mitigate viral transmission and to prevent overwhelming local healthcare infrastructure in college counties.

The COVID-19 outbreak is under control in China. Mobility interventions, including both the Wuhan lockdown and travel restrictions in other cities, have been undertaken in China to mitigate the epidemic. However, the impact of mobility restrictions in cites outside Wuhan has not been systematically analyzed. Here we ascertain the relationships between all mobility patterns and the epidemic trajectory in Chinese cities outside Hubei Province, and we estimate the impact of local travel restrictions. We estimate local inter-city travel bans averted 22.4% (95% PI: 16.8-27.9%) more infections in the two weeks after the Wuhan lockdown, while local intra-city travel prevented 32.5% (95% PI: 18.9-46.1%) more infections in the third and fourth weeks. More synchronized implementation of mobility interventions would further decrease the number of confirmed cases in the first two weeks by 15.7% (95% PI:15.4-16.0%). This study shows synchronized travel restrictions across cities can be effective in COVID-19 control.

While photosynthates are partitioned by the relative strength of young developing leaves and roots as sinks for carbon-based resources, many plants also show a close relationship between partitioning, phyllotaxy and vascular connectivity, yielding sectorial patterns of allocation. We examined whether smoke influences phloem vascular sectoriality in a model plant, sunflower (Helianthis annuus L.). Using radioactive 11CO2 administered to single photosynthetically active source leaves, we examined the transport behavior and allocation patterns of 11C-photosynthates using gamma counting and autoradiography. Soil treated with liquid smoke caused significant reductions in phloem sectoriality involving young sink leaves and roots. The resulting increase in vascular connectivity could benefit young plant performance by allowing a more uniform allocation of nutrients and/or stress signal molecules at a critical time of their growth. One-Sentence SummarySmoke-exposed roots exhibit a significant reduction in phloem sectoriality involving carbon transport to young sink leaves and roots.

Large language models (LLMs) increasingly operate in high-stakes settings including healthcare and medicine, where demographic attributes such as race and ethnicity may be explicitly stated or implicitly inferred from text. However, existing studies primarily document outcome-level disparities, offering limited insight into internal mechanisms underlying these effects. We present a mechanistic study of how race and ethnicity are represented and operationalized within LLMs. Using two publicly available datasets spanning toxicity-related generation and clinical narrative understanding tasks, we analyze three open-source models with a re-producible interpretability pipeline combining probing, neuron-level attribution, and targeted intervention. We find that demographic information is distributed across internal units with substantial cross-model variation. Although some units encode sensitive or stereotype-related associations from pretraining, identical demographic cues can induce qualitatively different behaviors. Interventions suppressing such neurons reduce bias but leave substantial residual effects, suggesting behavioral rather than representational change and motivating more systematic mitigation.