With the rapid advances in single-cell sequencing technology, it is now feasible to conduct in-depth genetic analysis in individual cells. Study on the dynamics of single cells in response to perturbations is of great significance for understanding the functions and behaviors of living organisms. However, the acquisition of post-perturbation cellular states via biological experiments is frequently cost-prohibitive. Predicting the single-cell perturbation responses poses a critical challenge in the field of computational biology. In this work, we propose a novel deep learning method called coupled variational autoencoders (CoupleVAE), devised to predict the postperturbation single-cell RNA-Seq data. CoupleVAE is composed of two coupled VAEs connected by a coupler, initially extracting latent features for controlled and perturbed cells via two encoders, subsequently engaging in mutual translation within the latent space through two nonlinear mappings via a coupler, and ultimately generating controlled and perturbed data by two separate decoders to process the encoded and translated features. CoupleVAE facilitates a more intricate state transformation of single cells within the latent space. Experiments in three real datasets on infection, stimulation and cross-species prediction show that CoupleVAE surpasses the existing comparative models in effectively predicting single-cell RNA-seq data for perturbed cells, achieving superior accuracy.

Accurate prediction of epidemics is pivotal for making well-informed decisions for the control of infectious diseases, but addressing heterogeneity in the system poses a challenge. In this study, we propose a novel modelling framework integrating the spatio-temporal heterogeneity of susceptible individuals into homogeneous models, by introducing a continuous recruitment process for the susceptibles. A neural network approximates the recruitment rate to develop a Universal Differential Equations (UDE) model. Simultaneously, we pre-set a specific form for the recruitment rate and develop a mechanistic model. Data from a COVID Omicron variant outbreak in Shanghai are used to train the UDE model using deep learning methods and to calibrate the mechanistic model using MCMC methods. Subsequently, we project the attack rate and peak of new infections for the first Omicron wave in China after the adjustment of the dynamic zero-COVID policy. Our projections indicate an attack rate and a peak of new infections of 80.06% and 3.17% of the population, respectively, compared with the homogeneous model's projections of 99.97% and 32.78%, thus providing an 18.6% improvement in the prediction accuracy based on the actual data. Our simulations demonstrate that heterogeneity in the susceptibles decreases herd immunity for ~37.36% of the population and prolongs the outbreak period from ~30 days to ~70 days, also aligning with the real case. We consider that this study lays the groundwork for the development of a new class of models and new insights for modelling heterogeneity.

Non-pharmaceutical interventions (NPIs) are effective in mitigating infections during the early stages of an infectious disease outbreak. However, these measures incur significant economic and livelihood costs. To address this, we developed an optimal control framework aimed at identifying strategies that minimize such costs while ensuring full control of a cross-regional outbreak of emerging infectious diseases. Our approach uses a spatial SEIR model with interventions for the epidemic process, and incorporates population flow in a gravity model dependent on gross domestic product (GDP) and geographical distance. We applied this framework to identify an optimal control strategy for the COVID-19 outbreak caused by the Delta variant in Xi'an City, Shaanxi, China, between December 2021 and January 2022. The model was parameterized by fitting it to daily case data from each district of Xi'an City. Our findings indicate that an increase in the basic reproduction number, the latent period or the infectious period leads to a prolonged outbreak and a larger final size. This indicates that diseases with greater transmissibility are more challenging and costly to control, and so it is important for governments to quickly identify cases and implement control strategies. Indeed, the optimal control strategy we identified suggests that more costly control measures should be implemented as soon as they are deemed necessary. Our results demonstrate that optimal control regimes exhibit spatial, economic, and population heterogeneity. More populated and economically developed regions require a robust regular surveillance mechanism to ensure timely detection and control of imported infections. Regions with higher GDP tend to experience larger-scale epidemics and, consequently, require higher control costs. Notably, our proposed optimal strategy significantly reduced costs compared to the actual expenditures for the Xi'an outbreak.

RNA-seq, including bulk RNA-seq and single-cell RNA-seq, is a next-generation sequencing-based RNA profiling method capable of measuring gene expression patterns with high resolution, and has gradually become an essential tool for the analysis of differential gene expression at the whole transcriptome level. Differential gene identification is a key problem in many biological studies such as disease genetics. Two-sample location test methods are widely used in case-control studies to identify the significant differential genes. However, due to the high cost of labelled data collection, many studies face the small sample problem since there is only small labelled data available, for which the traditional methods often lose power. To address this issue, we propose a novel rank-based nonparametric test method called WMW-A test based onWilcoxon-Mann-Whitiney test by introducing a three-sample statistic through anotherauxiliary sample, which is either given or generated in form of unlabelled data. By combining the case, control and auxiliary samples together, we construct a three-sample WMW-A statistic based on the gap between the average ranks of the case and control samples in the combined samples. The extensive simulation experiments and real applications on different gene expression datasets, including one bulk RNA-seq dataset and two single cell RNA-seq datasets, show that the WMW-A test could significantly improve the test power for two-sample problem with small sample sizes, by either available or generated auxiliary data. The applications on two real small SARS-CoV-2 datasets further show the improvement of WMW-A test for differentially expressed gene identification with small labelled samples.

During the COVID-19 pandemic, control measures, especially massive contact tracing following prompt quarantine and isolation, play an important role in mitigating the disease spread, and quantifying the dynamic contact rate and quarantine rate and estimate their impacts remain challenging. To precisely quantify the intensity of interventions, we develop the mechanism of physics-informed neural network (PINN) to propose the extended transmission-dynamics-informed neural network (TDINN) algorithm by combining scattered observational data with deep learning and epidemic models. The TDINN algorithm can not only avoid assuming the specific rate functions in advance but also make neural networks follow the rules of epidemic systems in the process of learning. We show that the proposed algorithm can fit the multi-source epidemic data in Xi'an, Guangzhou and Yangzhou cities well, and moreover reconstruct the epidemic development trend in Hainan and Xinjiang with incomplete reported data. We inferred the temporal evolution patterns of contact/quarantine rates, selected the best combination from the family of functions to accurately simulate the contact/quarantine time series learned by TDINN algorithm, and consequently reconstructed the epidemic process. The selected rate functions based on the time series inferred by deep learning have epidemiologically reasonable meanings. In addition, the proposed TDINN algorithm has also been verified by COVID-19 epidemic data with multiple waves in Liaoning province and shows good performance. We find the significant fluctuations in estimated contact/quarantine rates, and a feedback loop between the strengthening/relaxation of intervention strategies and the recurrence of the outbreaks. Moreover, the findings show that there is diversity in the shape of the temporal evolution curves of the inferred contact/quarantine rates in the considered regions, which indicates variation in the intensity of control strategies adopted in various regions.

Infectious diseases attack humans from time to time and threaten the lives and survival of people all around the world. An important strategy to prevent the spatial spread of infectious diseases is to restrict population travel. With the reduction of the epidemic situation, when and where travel restrictions can be lifted, and how to organize orderly movement patterns become critical and fall within the scope of this study. We define a novel diffusion distance derived from the estimated mobility network, based on which we provide a general model to describe the spatiotemporal spread of infectious diseases with a random diffusion process and a deterministic drift process of the population. We consequently develop a multi-source data fusion method to determine the population flow in epidemic areas. In this method, we first select available subregions in epidemic areas, and then provide solutions to initiate new travel flux among these subregions. To verify our model and method, we analyze the multi-source data from mainland China and obtain a new travel flux triggering scheme in the selected 29 cities with the most active population movements in mainland China. The testable predictions in these selected cities show that reopening the borders in accordance with our proposed travel flux will not cause a second outbreak of COVID-19 in these cities. The finding provides a methodology of re-triggering travel flux during the weakening spread stage of the epidemic.