Macrolide resistance rates of Mycoplasma pneumoniae in the Beijing population were as high as 68.9%, 90.0%, 98.4%, 95.4%, and 97.0% in the years 2008 to 2012, respectively. Common macrolide-resistant mobile genetic elements were not detected with any isolate. These macrolide-resistant isolates came from multiple clones rather than the same clone. No massive aggregation of a particular clone was found in a specific period. Mycoplasma pneumoniae is one of the important pathogens causing human respiratory tract infection, especially in community-acquired pneumonia (1, 2). The major clinical treatment for M. pneumoniae infection is the use of macrolide antibiotics (ML). With the widespread use of the drug, ML-resistant isolates have been reported worldwide (3-5). The resistance mechanism has been identified as a point mutation in the 23S rRNA gene. Other mechanisms of macrolide resistance cannot be excluded and have not been identified. In recent years, ML-resistant M. pneumoniae has become very serious in Asia (6, 7) and has attracted the attention of scientists. Studies on ML-resistant M. pneumoniae in China have only recently been conducted, and the limited reports have been mainly ML resistance analyses of a small number of strains isolated during a few months and from specific populations, such as children or adults (8-11). These reports are lacking continuous full-population surveillance data of M. pneumoniae drug resistance. In view of the above-mentioned information, we have studied drug resistance of 309 M. pneumoniae isolates from a whole population of strains isolated from people with respiratory infections in Beijing, China, from 2008 to 2012, a study which will help us to understand the status of drug-resistant M. pneumoniae in Beijing in recent years.M. pneumoniae strains. A total of 309 M. pneumoniae strains were isolated from 1,183 respiratory infection specimens from Beijing Chao-Yang Hospital, Beijing Children's Hospital, and Beijing Centers for Diseases Control and Prevention. One hundred fifty-six isolates were from 388 pediatric specimens of patients Ͻ14 years of age, and the remaining 153 isolates were collected from 795 adolescent and adult specimens. All 309 isolates were purified, cultured, and identified with a real-time PCR method (12).Detection of macrolide resistance at the gene level. Genomic DNA of 309 M. pneumoniae isolates was extracted using the QIAamp DNA minikit (Qiagen). The extracts were distributed into aliquots and saved at Ϫ20°C. The domain V region of the 23S rRNA gene was amplified by PCR methods described previously (6). The amplification products were sequenced by the Beijing Genomics Institute (BGI). The results showed that there were existing point mutations in domain V of the 23S rRNA gene region of 280 strains in the 309 M. pneumoniae isolates. In 272 of the 280 isolates (97.1%), the mutation was identified as A2063G. Seven of the 280 isolates (2.5%) had the A2064G mutation, one of the 280 isolates (0.4%) had an A2063T mutation, and the remaining 29 isolates did not h...

Dengue fever (DF) is one of the most rapidly spreading diseases in the world, and accurate forecasts of dengue in a timely manner might help local government implement effective control measures. To obtain the accurate forecasting of DF cases, it is crucial to model the long-term dependency in time series data, which is difficult for a typical machine learning method. This study aimed to develop a timely accurate forecasting model of dengue based on long short-term memory (LSTM) recurrent neural networks while only considering monthly dengue cases and climate factors. The performance of LSTM models was compared with the other previously published models when predicting DF cases one month into the future. Our results showed that the LSTM model reduced the average the root mean squared error (RMSE) of the predictions by 12.99% to 24.91% and reduced the average RMSE of the predictions in the outbreak period by 15.09% to 26.82% as compared with other candidate models. The LSTM model achieved superior performance in predicting dengue cases as compared with other previously published forecasting models. Moreover, transfer learning (TL) can improve the generalization ability of the model in areas with fewer dengue incidences. The findings provide a more precise forecasting dengue model and could be used for other dengue-like infectious diseases.

No abstract

No abstract

From June to July 2007, 36 cases of Guillain-Barre syndrome (GBS) occurred in a township in north China. Serological study and bacteria culture were performed to investigate the association between preceding Campylobacter jejuni infection and this GBS outbreak. Anti-C. jejuni antibodies were found in significantly higher numbers of GBS patients (IgM 84%, IgG 87.5%) than in healthy inspection cases (IgM 33%, IgG 27%). IgG anti-GM1 was the dominant anti-ganglioside antibody among the GBS patients. Seven C. jejuni isolates (four from human stool and three from poultry specimens taken from the patients' houses) were obtained. Serotyping and molecular analysis were used to investigate the genetic relatedness among these C. jejuni isolates. The four human isolates, collected from residents of the same district, were indistinguishable by both pulsed-field gel electrophoresis and multilocus sequence typing, suggesting these patients had a common source of infection. A new sequence type, sequence type-2993, was assigned to the human C. jejuni isolates, three of which belonged to Penner serotype heat-stable (HS):41. Both serotype and molecular subtype of the human C. jejuni isolates were different from those of isolates obtained from poultry specimens. Our results suggest that the antecedent C. jejuni infection triggered this GBS outbreak in China.