DE-MHAIPs: Identification of SARS-CoV-2 phosphorylation sites based on differential evolution multi-feature learning and multi-head attention mechanism

The rapid spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) around the world affects the normal lives of people all over the world. The computational methods can be used to accurately identify SARS-CoV-2 phosphorylation sites. In this paper, a new prediction model of SARS-CoV-2 phosphorylation sites, called DE-MHAIPs, is proposed. First, we use six feature extraction methods to extract protein sequence information from different perspectives. For the first time, we use a differential evolution (DE) algorithm to learn individual feature weights and fuse multi-information in a weighted combination. Next, Group LASSO is used to select a subset of good features. Then, the important protein information is given higher weight through multi-head attention. After that, the processed data is fed into long short-term memory network (LSTM) to further enhance model's ability to learn features. Finally, the data from LSTM are input into fully connected neural network (FCN) to predict SARS-CoV-2 phosphorylation sites. The AUC values of the S/T and Y datasets under 5-fold cross-validation reach 91.98% and 98.32%, respectively. The AUC values of the two datasets on the independent test set reach 91.72% and 97.78%, respectively. The experimental results show that the DE-MHAIPs method exhibits excellent predictive ability compared with other methods.

Modeling the effects of vaccination, nucleic acid testing, and face mask wearing interventions against COVID-19 in large sports events.

The transmission of SARS-CoV-2 leads to devastating COVID-19 infections around the world, which has affected both human health and the development of industries dependent on social gatherings. Sports events are one of the subgroups facing great challenges. The uncertainty of COVID-19 transmission in large-scale sports events is a great barrier to decision-making with regard to reopening auditoriums. Policymakers and health experts are trying to figure out better policies to balance audience experiences and COVID-19 infection control. In this study, we employed the generalized SEIR model in conjunction with the Wells-Riley model to estimate the effects of vaccination, nucleic acid testing, and face mask wearing on audience infection control during the 2021 Chinese Football Association Super League from 20 April to 5 August. The generalized SEIR modeling showed that if the general population were vaccinated by inactive vaccines at an efficiency of 0.78, the total number of infectious people during this time period would decrease from 43,455 to 6,417. We assumed that the general population had the same odds ratio of entering the sports stadiums and becoming the audience. Their infection probabilities in the stadium were further estimated by the Wells-Riley model. The results showed that if all of the 30,000 seats in the stadium were filled by the audience, 371 audience members would have become infected during the 116 football games in the 2021 season. The independent use of vaccination and nucleic acid testing would have decreased this number to 79 and 118, respectively. The combined use of nucleic acid testing and vaccination or face mask wearing would have decreased this number to 14 and 34, respectively. The combined use of all three strategies could have further decreased this number to 0. According to the modeling results, policymakers can consider the combined use of vaccination, nucleic acid testing, and face mask wearing to protect audiences from infection when holding sports events, which could create a balance between audience experiences and COVID-19 infection control.