| BackgroundInfluenza is an acute respiratory infectious disease caused by influenza virus in humans and other animals.Influenza spreads rapidly and widely.It is generally characterized by fever,chills,cough,sore throat,runny nose or nasal congestion,headache,muscle aches,and fatigue.In severe cases,multiple complications may occur,including viral pneumonia,bacterial pneumonia,acute respiratory distress syndrome,disseminated other extra-pulmonary symptoms like intravascular coagulation and shock.Humans are generally susceptible to influenza viruses,especially adolescents,the elderly,and those with weakened immune systems.Worldwide,there are up to 1 billion cases of seasonal influenza each year,with approximately 3-5 million severe cases and up to 650,000 respiratory infections-related deaths.China is a country with a high burden of influenza infection.A population-based study on deaths due to influenza virus respiratory infection shows that an average of 88,000 influenza deaths occur in China each year,accounting for about 13.6% of the global influenza-related deaths.This research is carried on when the world is facing the most severe and most widespread new respiratory infection epidemic---Corona Virus Disease 2019(COVID-19)since Severe acute respiratory syndrome(SARS).Up to now,the “global pandemic” of COVID-19 has spread to more than 200 countries and regions.According to data from the World Health Organization,as of 24:00 on March 30,2022,the cumulative number of confirmed cases of COVID-19 in the world exceeded 460 million,which has seriously affected the production and life of human society around the world.Influenza virus infection is highly seasonal and harmful.In the context of the COVID-19 epidemic,the trend and risk of seasonal influenza is a new proposition in the field of respiratory infectious diseases.PurposeThis study is through epidemiological investigation and pathogen monitoring to analyze the epidemiological characteristics and changing patterns of influenza virus infection and explore the genetic evolution of influenza virus HA,NA and PA genes in South China during the COVID-19 epidemic;meanwhile to construct a predictive model of influenza epidemic in the context of the COVID-19 epidemic using multi-source parameters such as influenza positive case monitoring data,weather and climate data and influenza virus gene evolution index,combined with artificial intelligence learning algorithm.This provides an effective scientific basis for China to respond to and formulate precise control strategies against influenza in the context of the COVID-19 epidemic.Methods(1)Based on the data and information from the Clinical Respiratory Virus Surveillance Network of the National Clinical Research Center for Respiratory Disease,descriptive epidemiological statistical methods were used to analyze and conclude the etiological and epidemiological characteristics and changing patterns of influenza epidemics in Guangdong region,China during 2018-2021.(2)Select influenza virus nucleic acid-positive respiratory clinical samples from December 2019 to January 2020 in Guangdong region,extract RNA from the samples,use specific amplification primers to amplify the HA,NA and PA genes of three types of influenza strains with reverse transcription-polymerase chain reaction(RT-PCR),extract and purify the amplification products,and then perform full-length sequencing of the HA,NA and PA gene sequences,and compare the HA,NA and PA gene sequences with those of selected vaccine strains and reference strains.As for the software used,“Tracy” is used for the gene sequence splicing,“blast” is used for sequence homology analysis,“Muscle” is used for multiple sequence alignment,“IQ-TREE” is used to construct phylogenetic tree and “Data Monkey” is used for selective pressure analysis.(3)Using multi-source data like the time series of influenza-positive cases in Guangdong Province,influenza-positive case monitoring data from the Clinical Respiratory Virus Surveillance Network of the National Clinical Research Center for Respiratory Disease,weather and climate data,influenza virus genetic evolution index and use of mask,based on the classic SIRS epidemiological model,combined with the time series model,nonlinear fitting model and neural network learning algorithm,a new influenza predictive model is constructed.Train the newly-constructed SIRSNet predictive model with the number of active influenza infections per 1,000 people in Guangdong Province from January 2018 to July 2019,and fit the curve of the influenza infection epidemic from September 2019 to January 2020,and verify the fitting curve with real data,so as to test the effectiveness of the predictive model.Results1.Epidemiological characteristics of influenza in Guangdong Province from 2018 to 2021Before the COVID-19 epidemic,influenza virus infection in Guangdong Province was distinctly seasonal.Influenza epidemics can occur in winter and spring,or in summer.During 2018-2019,influenza-like cases in the south tended to be more active in spring and winter.In the 2020-2021 flu season,during the outbreak of COVID-19 as well,the influenza virus is at a low prevalence.The analysis of susceptible populations shows that the main groups affected by influenza in Guangdong Province are people aged 5-14,15-49 and 50-64,and the number of cases in these three age groups accounts for more than 70% of the total number of cases.The susceptible age groups of influenza A virus are different in these three years,5-14 years old in 2018,50-64 in 2019,and 15-49 in 2020,which may be related to the different epidemic subtypes of influenza A virus in different years.There is no significant change in the susceptible populations of influenza B virus in the three years,all of whom are 15-49 years old.Overall,the proportion of influenza virus positivity in the 15-49 age group is increasing year by year.From 2018 to 2021,the proportion of each subtype of influenza virus in South China keeps changing every year,and the dominant strains are also different.Compared with the previous three years,the detection rate of influenza virus in people of all ages in 2021 is low.In 2018 and 2019 before the COVID-19 epidemic,the positive detection rates of influenza A and B viruses in people aged 15-49 are both higher than 12%.Although the positive detection rate of influenza B virus in people of all ages is lower than 3% in 2020,it has shown a gradual rise since 2021.In general,influenza A virus is the main pathogen causing influenza epidemics in South China.During the COVID-19 epidemic,influenza A virus(H3N2),influenza A virus(H1N1),and influenza B virus Victoria are the main epidemic strains2.Molecular epidemiological study of influenza virus during the COVID-19epidemicDuring the COVID-19 epidemic,the HA gene of influenza A virus(H3N2)has undergone mutations in multiple antigenic epitopes and presented various antigenic variation patterns.Among them,the mutations of HA protein at sites 202(amino acid G changed to D)and 206(amino acid D changed to N)are new mutations.The mutation of NA protein at site 344(amino acid E changed to K)accounts for 85.71%of the mutation ratio of all samples;the mutations of PA protein at sites 626(amino acid changed from K to R)and 321(amino acid Y changed to C)account for 67.86%.Compared with the influenza virus vaccine strains recommended by WHO,the influenza A virus(H3N2)have antigenic mutations at sites 144(amino acid K changed to S),159(amino acid S changed to Y),121(amino acid N changed to K)and 62(amino acid E changed to G).The common gene mutations in the HA gene of influenza A virus(H1N1)are G204 A,Q206E,V267 A and so on.Compared with the influenza virus vaccine strains recommended by WHO,the mutation sites of antigenic epitopes of influenza A virus(H1N1)during the COVID-19 epidemic in South China are mainly sites 91,180,181,202 and 200.The mutations of G148 R,R151E and NDKN178 N commonly happen in the HA gene sequence of influenza B virus.Compared with the influenza virus vaccine strains recommended by WHO,the mutations in antigenic epitopes of influenza B virus during the COVID-19 epidemic are mainly G144 D at site 144 in 120-loop region of the HA gene,G148 R at site 148,and K151 E at site 151.The analysis of the selective pressure of genetic evolution shows that the HA,NA and PA genes of influenza virus in Guangdong Province are subjected to purifying selective pressure during the COVID-19 epidemic,and the evolutionary driving force is self-adaptive change.The selective pressure of HA and NA genes is greater than that of PA genes.There are also partly positive selection sites in the HA gene of influenza A virus(H1N1)and influenza B virus.Drug resistance mutation analysis shows that no influenza virus strains are found to produce drug resistance mutations at the main resistance sites of neuraminidase inhibitors(119E,136 Q,151D,198 D,222I,224 R,274H,276 E,292R,294 N 371R),which indicates that the influenza viruses in Guangdong Province don’t develop universal resistance to neuraminidase inhibitors during the COVID-19 epidemic.3.Prediction of influenza epidemic trend in the context of the COVID-19epidemicThrough the SIRSNet influenza predictive model we constructed,the influenza prevalence in Guangdong is simulated: Assuming that there is no outbreak of COVID-19 from December 2019 to February 2020,the outbreak peak of influenza A is in February and March 2020 during which there will be 8 active cases of influenza infection per 1,000 people.However,there are actually only 2.45 active cases of infection per 1,000 people,and no influenza A epidemic has occurred under the implementation of interventions against COVID-19.Assuming that without masks,the number of active influenza A infection cases in Guangdong Province will increase since December 16,2020,and peak in January 2021,at a peak of 15.86 cases of active influenza infection per 1,000 people.If only 10% of the population in Guangdong Province(about 13 million people)wear masks,the number of active influenza infection cases is second only to the number of active cases at the peak,with 15 active influenza infection cases per 1,000 people and all together 6,501,300 cases based on the total population.If 50% of the population(about 65 million people)wear masks,the number of active influenza infections per 1,000 people will be significantly reduced to 5.64,with all together 2,602,600 influenza cases based on the total population,which is lower than half of the infections at the peak.Assuming that 90%of the total population(about 117 million people)wear masks,the simulated influenza infection cases will be greatly reduced to 785,200.The “mask wearing” index from the predictive model shows that the higher the mask usage rate at the first outbreak,the faster the Rt value of the effective reproduction number in subsequent outbreaks will decrease;the higher the mask usage rate,the lower the peak of Rt value,which is crucial for reducing the infected population base during influenza outbreaks.Conclusions1.From 2018 to 2021,different subtypes of influenza viruses in Guangdong Province have different epidemic patterns,and the dominant strains are constantly changing.During the COVID-19 epidemic,influenza A virus(H3N2),influenza A virus(H1N1),and influenza virus B Victoria are the main epidemic strains.2.During the COVID-19 epidemic,the genes of influenza viruses in Guangdong Province are evolving rapidly.The HA gene of the viruses has undergone mutations in multiple antigenic epitopes and presented various antigenic variation patterns.Compared with the influenza virus vaccine strains recommended by WHO,gene mutations at antigenic cluster sites may weaken the effectiveness of vaccine,resulting in influenza epidemics.3.During the COVID-19 epidemic,the HA,NA and PA genes of influenza virus in Guangdong Province are all subject to purifying selective pressure,and the evolutionary driving force is self-adaptive mutation.No influenza virus strains are found to produce drug resistance mutations at the main resistance sites of neuraminidase inhibitors,which indicates that the influenza viruses don’t develop universal resistance to neuraminidase inhibitors.4.A SIRSNet influenza predictive model based on multi-source data(such as weather factors and evolution indexes)and a variety of artificial intelligence algorithms is successfully constructed.Through the “mask wearing” index from the predictive model,wearing a mask can effectively reduce the Rt value of the effective reproduction number of the epidemic development.Wearing a mask has an extremely important influence on the development trend of potential influenza epidemics in the future. |
PDF Full Text Request