Globally, five variants of concern (VOC) and two variants of interest (VOI) of SARS-CoV-2 have been identified by 1 January 2022, ¹ with varying ability to escape from pre-existing immunity following prior infection or vaccination. ² The ‘serial interval’ (SI) and ‘incubation period’ (IP) are two key epidemiological metrics, defined as the duration between symptom onset of successive cases in a transmission chain and the duration between exposure and symptom onset of a case, respectively, which can be used to characterize the COVID-19 transmission. Reliable estimates of these parameters are essential for determining the duration of the quarantine needed to minimize the transmission risk. ³

South Africa (SA) first detected the B.1.1.529 SARS-CoV-2 variant from a specimen collected on 9 November 2021 and reported the variant to the World Health Organization (WHO) on 24 November 2021. ¹ Two days later, the WHO named the variant Omicron and classified it as a variant of concern (VOC) because of increasing detections in SA and large number of mutations in the immunogenic regions of the spike protein. ¹ In addition to increasing cases in SA, the new variant may have been prevalent in other locations in southern Africa. By 30 November 2021, over 11 countries had reported detections of the Omicron variant, with the first reports of international importations coming from Hong Kong, Israel, Japan and France. ² Three days later, it was reported in 29 additional countries. ² As of 2 December 2021, >50 countries had enacted border controls to slow the global spread of Omicron. ³ For example, Japan and Israel closed their borders to all foreign travellers. ⁴ The Omicron variant might have been spreading cryptically in these and other countries before the WHO declaration, undetected because of the limited viral sequencing capacity.

Mobile data are a feasible way for us to understand and reveal the feature of human mobility. However, it is extremely hard to have a fine-grained picture of large-scale mobility data, in particular at an urban scale. Here, we present a large-scale dataset of 2-million mobile phone users with time-varying locations, denoted as the temporal network of individuals, conducted by an open-data program in Changchun Municipality. To reveal human mobility across locations, we further construct the aggregated mobility network for each day by taking cellular base stations as nodes coupled by edges weighted by the total number of users’ movements between pairs of nodes. The resulting temporal network of mobile phone users and the dynamic, weighted and directed mobility network are released in simple formats for easy access to motivating research using this new and extensive data of human mobility.

Phages are the functional viruses that infect bacteria and they play important roles in microbial communities and ecosystems. Phage research has attracted great attention due to the wide applications of phage therapy in treating bacterial infection in recent years. Metagenomics sequencing technique can sequence microbial communities directly from an environmental sample. Identifying phage sequences from metagenomic data is a vital step in the downstream of phage analysis. However, the existing methods for phage identification suffer from some limitations in the utilization of the phage feature for prediction, and therefore their prediction performance still need to be improved further. In this article, we propose a novel deep neural network (called MetaPhaPred) for identifying phages from metagenomic data. In MetaPhaPred, we first use a word embedding technique to encode the metagenomic sequences into word vectors, extracting the latent feature vectors of DNA words. Then, we design a deep neural network with a convolutional neural network (CNN) to capture the feature maps in sequences, and with a bi-directional long short-term memory network (Bi-LSTM) to capture the long-term dependencies between features from both forward and backward directions. The feature map consists of a set of feature patterns, each of which is the weighted feature extracted by a convolution filter with convolution kernels in the CNN slide along the input feature vectors. Next, an attention mechanism is used to enhance contributions of important features. Experimental results on both simulated and real metagenomic data with different lengths demonstrate the superiority of the proposed MetaPhaPred over the state-of-the-art methods in identifying phage sequences. The source code and data are available at: https://github.com/dww01/MetaPhaPred .

Background: With the emergence of SARS-CoV-2 variants that have eluded immunity from vaccines and prior infections, vaccine shortages and vaccine effectiveness pose unprecedented challenges for governments in expanding booster vaccination programs. The fractionation of vaccine doses might be an effective strategy for helping society to face these challenges, as fractional doses may have efficacies comparable with those of the standard doses. Objective: This study aims to investigate the relationship between vaccine immunogenicity and protection and to project efficacies of fractional doses of vaccines for COVID-19 by using neutralizing antibody levels. Methods: In this study, we analyzed the relationship between in vitro neutralization levels and the observed efficacies against both asymptomatic infection and symptomatic infection, using data from 13 studies of 10 COVID-19 vaccines and from convalescent cohorts. We further projected efficacies for fractional doses, using neutralization as an intermediate variable, based on immunogenicity data from 51 studies included in our systematic review. Results: In comparisons with the convalescent level, vaccine efficacy against asymptomatic infection and symptomatic infection increased from 8.8% (95% CI 1.4%-16.1%) to 71.8% (95% CI 63%-80.7%) and from 33.6% (95% CI 23.6%-43.6%) to 98.6% (95% CI 97.6%-99.7%), respectively, as the mean neutralization level increased from 0.1 to 10 folds of the convalescent level. Additionally, mRNA vaccines provided the strongest protection, which decreased slowly for fractional dosing with dosages between 50% and 100% of the standard dose. We also observed that although vaccine efficacy increased with the mean neutralization level, the rate of this increase was slower for vaccine efficacy against asymptomatic infection than for vaccine efficacy against symptomatic infection. Conclusions: Our results are consistent with studies on immune protection from SARS-CoV-2 infection. Based on our study, we expect that fractional-dose vaccination could provide partial immunity against SARS-CoV-2 and its variants. Our findings provide a theoretical basis for the efficacy of fractional-dose vaccines, serving as reference evidence for implementing fractional dosing vaccine policies in areas facing vaccine shortages and thereby mitigating disease burden. Fractional-dose vaccination could be a viable vaccination strategy comparable to full-dose vaccination and deserves further exploration.

Phage virion proteins (PVPs), which form the structural components of phages, are crucial for maintaining phage structures and infecting host bacteria. Identifying PVP can lead to the development of novel therapeutic agents to combat bacterial infections, attracting great research attention in recent years. However, most of the existing methods for PVP identification heavily depend on the effectiveness of feature extraction and have no ability to precisely classify specific classes. In this article, we propose a bidirectional encoder representations-based Transformer model called BERTPVP for the identification and classification of PVP. BERTPVP uses a stack of transformer encoders to effectively capture contextual information from the entire protein sequence through the multi-head self-attention mechanism. We initially pre-train the model using the masked language modeling task to learn contextual information from phage protein sequences, which aids in understanding the significance and relevance of individual components in relation to the entire sequence. Subsequently, the pre-trained model is fine-tuned for PVP identification and classification tasks. Our experimental results show the superiority of the proposed BERTPVP over the state-of-the-art methods in identifying and classifying PVP. Moreover, the ablation study demonstrates the necessary of the pre-training and fine-tuning components of BERTPVP in accelerating convergence and improving prediction performance.