A spatiotemporally resolved single cell atlas of the Plasmodium liver stage

Malaria infection involves an obligatory, yet clinically silent liver stage1,2. Hepatocytes operate in repeating units termed lobules, exhibiting heterogeneous gene expression patterns along the lobule axis3, but the effects of hepatocyte zonation on parasite development have not been molecularly explored. Here, we combine single-cell RNA sequencing4 and single-molecule transcript imaging5 to characterize the host’s and parasite’s temporal expression programs in a zonally-controlled manner for the rodent malaria parasite Plasmodium berghei ANKA. We identify differences in parasite gene expression in distinct zones, and a sub-population of periportally-biased hepatocytes that harbor abortive infections associated with parasitophorous vacuole breakdown. These ‘abortive hepatocytes’ up-regulate immune recruitment and key signaling programs. They exhibit reduced levels of Plasmodium transcripts, perturbed parasite mRNA localization, and may give rise to progressively lower abundance of periportal infections. Our study provides a resource for understanding the liver stage of Plasmodium infection at high spatial resolution and highlights heterogeneous behavior of both the parasite and the host hepatocyte.

Health Status Prediction with Local-Global Heterogeneous Behavior Graph

Health management is getting increasing attention all over the world. However, existing health management mainly relies on hospital examination and treatment, which are complicated and untimely. The emergence of mobile devices provides the possibility to manage people’s health status in a convenient and instant way. Estimation of health status can be achieved with various kinds of data streams continuously collected from wearable sensors. However, these data streams are multi-source and heterogeneous, containing complex temporal structures with local contextual and global temporal aspects, which makes the feature learning and data joint utilization challenging. We propose to model the behavior-related multi-source data streams with a local-global graph, which contains multiple local context sub-graphs to learn short-term local context information with heterogeneous graph neural networks and a global temporal sub-graph to learn long-term dependency with self-attention networks. Then health status is predicted based on the structure-aware representation learned from the local-global behavior graph. We take experiments on the StudentLife dataset, and extensive results demonstrate the effectiveness of our proposed model.

Computational Behavioral Models in Public Goods Games with Migration Between Groups

In this study we have simulated numerically two models of linear Public Goods Games where players are equally distributed among a given number of groups. Agents play in their group by using two simple sets of rules that are inspired by the observed behavior of human participants in laboratory experiments. In addition, unsatisfied agents have the option of leaving their group and migrating to a new random one through probabilistic choices. Stochasticity, and the introduction of two types of players in the population, help simulate the heterogeneous behavior that is often observed in experimental work. The numerical simulation results of the corresponding dynamical systems show that being able to leave a group when unsatisfied favors contribution and avoids free-riding to a good extent in a range of the enhancement factor where defection would prevail without migration. Our numerical simulation results are qualitatively in line with known experimental data when human agents are given the same kind of information about themselves and the other players in the group. This is usually not the case with customary mathematical models based on replicator dynamics or stochastic approaches. As a consequence, models like the ones described here may be useful for understanding experimental results and also for designing new experiments by first running cheap simulations instead of doing costly preliminary laboratory work. The downside is that models and their simulation tend to be less general than standard mathematical approaches.

Single-cell profiling defines the prognostic benefit of CD39high tissue resident memory CD8+ T cells in luminal-like breast cancer

Luminal-like breast cancer (BC) constitutes the majority of BC subtypes, but, differently from highly aggressive triple negative BC, is poorly infiltrated by the immune system. The quality of the immune infiltrate in luminal-like BCs has been poorly studied, thereby limiting further investigation of immunotherapeutic strategies. By using high-dimensional single-cell technologies, we identify heterogeneous behavior within the tissue-resident memory CD8+ T (Trm) cells infiltrating luminal-like tumors. A subset of CD127− CD39hi Trm cells, preferentially present in the tumor compared to the adjacent normal breast tissue or peripheral blood, retains enhanced degranulation capacity compared to the CD127+ CD39lo Trm counterpart ex vivo, and is specifically associated with positive prognosis. Nevertheless, such prognostic benefit is lost in the presence of highly-suppressive CCR8hi ICOShi IRF4+ effector Tregs. Thus, combinatorial strategies aiming at boosting Trm function and infiltration while relieving from Treg-mediated immunosuppression should be investigated to achieve proper tumor control in luminal-like BCs.

Clinical - epidemiological aspects and spatial analysis of leprosy relapses in an endemic area of Brazil

Objective: to analyze the trend, spatial distribution and factors associated with leprosy recurrence in Sergipe. Methods: this is a population-based ecological study, using spatial analysis and logistic regression techniques. Data from all leprosy recurrences reported to SINAN (Notifiable Diseases Information System) in the state of Sergipe, Brazil, were used from 2007 to 2017. Results: there was a tendency for an increase in recurrences in the period from 2011 to 2017: APC: 14.69 (p-value = 0.003). The recurrence incidence map showed a heterogeneous behavior, with the Moran index (I = 0.16; p-value: 0.0159). The multivariate analysis showed a higher risk of recurrence in adults (aOR = 2.81) and young adults (aOR = 2.85) Conclusion: the risk factors associated with the appearance of recurrences are: the age group, the zone, the operational classification , the clinical form and the degree of disability.

Comparison of Pandemic Intervention Policies in Several Building Types Using Heterogeneous Population Model

In a world where pandemics are a matter of time and increasing urbanization of the world's population, governments should be prepared with pandemic intervention policies (IPs) to minimize the crisis direct and indirect adverse effects while keeping normal life as much as possible. Successful pandemic IPs have to take into consideration the heterogeneous behavior of individuals in different types of buildings and social contexts. In this study, we propose a spatio-temporal, heterogeneous population model and in silico simulation to evaluate pandemic IPs in four types of buildings - home, office, school, and mall. We show that indeed each building type has a unique pandemic spread and therefore a different optimal IP. Moreover, we show that temporal-based IPs (such as mask wearing) have a similar influence on the pandemic spread in all four building types while spatial-based IPs (such as social distance) highly differ.

Jointly Modeling Heterogeneous Student Behaviors and Interactions among Multiple Prediction Tasks

Prediction tasks about students have practical significance for both student and college. Making multiple predictions about students is an important part of a smart campus. For instance, predicting whether a student will fail to graduate can alert the student affairs office to take predictive measures to help the student improve his/her academic performance. With the development of information technology in colleges, we can collect digital footprints that encode heterogeneous behaviors continuously. In this article, we focus on modeling heterogeneous behaviors and making multiple predictions together, since some prediction tasks are related and learning the model for a specific task may have the data sparsity problem. To this end, we propose a variant of Long-Short Term Memory (LSTM) and a soft-attention mechanism. The proposed LSTM is able to learn the student profile-aware representation from heterogeneous behavior sequences. The proposed soft-attention mechanism can dynamically learn different importance degrees of different days for every student. In this way, heterogeneous behaviors can be well modeled. In order to model interactions among multiple prediction tasks, we propose a co-attention mechanism based unit. With the help of the stacked units, we can explicitly control the knowledge transfer among multiple tasks. We design three motivating behavior prediction tasks based on a real-world dataset collected from a college. Qualitative and quantitative experiments on the three prediction tasks have demonstrated the effectiveness of our model.

Phase transition of the 2-Choices dynamics on core–periphery networks

The 2-Choices dynamics is a process that models voting behavior on networks and works as follows: Each agent initially holds either opinion blue or red; then, in each round, each agent looks at two random neighbors and, if the two have the same opinion, the agent adopts it. We study its behavior on a class of networks with core–periphery structure. Assume that a densely-connected subset of agents, the core, holds a different opinion from the rest of the network, the periphery. We prove that, depending on the strength of the cut between core and periphery, a phase-transition phenomenon occurs: Either the core’s opinion rapidly spreads across the network, or a metastability phase takes place in which both opinions coexist for superpolynomial time. The interest of our result, which we also validate with extensive experiments on real networks, is twofold. First, it sheds light on the influence of the core on the rest of the network as a function of its connectivity toward the latter. Second, it is one of the first analytical results which shows a heterogeneous behavior of a simple dynamics as a function of structural parameters of the network.

Heterogeneity of Response and Immune System Activity during Treatment with Nivolumab in Hepatocellular Carcinoma: Results from a Single-Institution Retrospective Analysis

Treatment of hepatocellular carcinoma (HCC) is rapidly evolving, with many new therapeutic options; in particular, immunotherapy (IT) is acquiring a major role, even in combination regimens. Despite these promising results, an important limitation is the lack of prognostic and predictive factors that prevent provision of a tool for patient stratification in order to select the most appropriate strategy. Furthermore, response assessment can be challenging with IT due to peculiar patterns such as mixed responses or pseudoprogression. We analyzed biological and clinical features from the first 10 HCC patients treated with nivolumab in our institution. Analysis of patterns of response in CT assessment revealed complete response in pulmonary lesions, along with heterogeneous behavior in the liver and other organ lesions. Peripheral blood mononuclear cells (PBMC) analysis in the first four patients showed unique alterations in a patient with poor prognosis, both at baseline (lower percentage of effector T cells, higher percentage of natural killer T [NK/T] cells) and during treatment with nivolumab (decrease in nonclassical monocytes, increase in monocytic myeloid-derived suppressor cells [MO-MDSC]), suggesting a possible prognostic role for these features. Although obtained in a small cohort of patients, our results open a new perspective for understanding mechanisms underlying IT outcomes in HCC patients.