Lu Sun

Machine learning techniques and computer-aided methods are now widely used in the pre-discovery tasks of drug discovery, effectively improving the efficiency of drug development and reducing the workload and cost. In this study, we used multi-source heterogeneous network information to build a network model, learn the network topology through multiple network diffusion algorithms, and obtain compressed low-dimensional feature vectors for predicting drug–target interactions (DTIs). We applied the metropolis–hasting random walk (MHRW) algorithm to improve the performance of the random walk with restart (RWR) algorithm, forming the basis by which the self-loop probability of the current node is removed. Additionally, the propagation efficiency of the MHRW was improved using the improved metropolis–hasting random walk (IMRWR) algorithm, facilitating network deep sampling. Finally, we proposed a correction of the transfer probability of the entire network after increasing the self-loop rate of isolated nodes to form the ISLRWR algorithm. Notably, the ISLRWR algorithm improved the area under the receiver operating characteristic curve (AUROC) by 7.53 and 5.72%, and the area under the precision-recall curve (AUPRC) by 5.95 and 4.19% compared to the RWR and MHRW algorithms, respectively, in predicting DTIs performance. Moreover, after excluding the interference of homologous proteins (popular drugs or targets may lead to inflated prediction results), the ISLRWR algorithm still showed a significant performance improvement.

Large-scale population movements can turn local diseases into widespread epidemics. Grasping the characteristic of the population flow in the context of the COVID-19 is of great significance for providing information to epidemiology and formulating scientific and reasonable prevention and control policies. Especially in the post-COVID-19 phase, it is essential to maintain the achievement of the fight against the epidemic. Previous research focuses on flight and railway passenger travel behavior and patterns, but China also has numerous suburban residents with a not-high economic level; investigating their travel behaviors is significant for national stability. However, estimating the impacts of the COVID-19 for suburban residents’ travel behaviors remains challenging because of lacking apposite data. Here we submit bus ticketing data including approximately 26,000,000 records from April 2020–August 2020 for 2705 stations. Our results indicate that Suburban residents in Chinese Southern regions are more likely to travel by bus, and travel frequency is higher. Associated with the economic level, we find that residents in the economically developed region more likely to travel or carry out various social activities. Considering from the perspective of the traveling crowd, we find that men and young people are easier to travel by bus; however, they are exactly the main workforce. The indication of our findings is that suburban residents’ travel behavior is affected profoundly by economy and consistent with the inherent behavior patterns before the COVID-19 outbreak. We use typical regions as verification and it is indeed the case.