IMC-MDA: Prediction of miRNA-disease association based on induction matrix completion.

To comprehend the etiology and pathogenesis of many illnesses, it is essential to identify disease-associated microRNAs (miRNAs). However, there are a number of challenges with current computational approaches, such as the lack of "negative samples", that is, confirmed irrelevant miRNA-disease pairs, and the poor performance in terms of predicting miRNAs related with "isolated diseases", i.e. illnesses with no known associated miRNAs, which presents the need for novel computational methods. In this study, for the purpose of predicting the connection between disease and miRNA, an inductive matrix completion model was designed, referred to as IMC-MDA. In the model of IMC-MDA, for each miRNA-disease pair, the predicted marks are calculated by combining the known miRNA-disease connection with the integrated disease similarities and miRNA similarities. Based on LOOCV, IMC-MDA had an AUC of 0.8034, which shows better performance than previous methods. Furthermore, experiments have validated the prediction of disease-related miRNAs for three major human diseases: colon cancer, kidney cancer, and lung cancer.

Data augmentation based semi-supervised method to improve COVID-19 CT classification.

The Coronavirus (COVID-19) outbreak of December 2019 has become a serious threat to people around the world, creating a health crisis that infected millions of lives, as well as destroying the global economy. Early detection and diagnosis are essential to prevent further transmission. The detection of COVID-19 computed tomography images is one of the important approaches to rapid diagnosis. Many different branches of deep learning methods have played an important role in this area, including transfer learning, contrastive learning, ensemble strategy, etc. However, these works require a large number of samples of expensive manual labels, so in order to save costs, scholars adopted semi-supervised learning that applies only a few labels to classify COVID-19 CT images. Nevertheless, the existing semi-supervised methods focus primarily on class imbalance and pseudo-label filtering rather than on pseudo-label generation. Accordingly, in this paper, we organized a semi-supervised classification framework based on data augmentation to classify the CT images of COVID-19. We revised the classic teacher-student framework and introduced the popular data augmentation method Mixup, which widened the distribution of high confidence to improve the accuracy of selected pseudo-labels and ultimately obtain a model with better performance. For the COVID-CT dataset, our method makes precision, F1 score, accuracy and specificity 21.04%, 12.95%, 17.13% and 38.29% higher than average values for other methods respectively, For the SARS-COV-2 dataset, these increases were 8.40%, 7.59%, 9.35% and 12.80% respectively. For the Harvard Dataverse dataset, growth was 17.64%, 18.89%, 19.81% and 20.20% respectively. The codes are available at https://github.com/YutingBai99/COVID-19-SSL.

Changes in stoichiometric characteristics of ambient air pollutants pre-to post-COVID-19 in China.

To prevent the Corona Virus Disease 2019 (COVID-19) spreading, Chinese government takes a series of corresponding measures to restrict human mobility, including transportation lock-down and industries suspension, which significantly influenced the ambient air quality and provided vary rare time windows to assess the impacts of anthropological activities on air pollution. In this work, we divided the studied timeframe (2019/12/24-2020/2/24) into four periods and selected 88 cities from 31 representative urban agglomerations. The indicators of PM2.5/PM10 and NO2/SO2 were applied, for the first time, to analyze the changes in stoichiometric characteristics of ambient air pollutants pre-to post-COVID-19 in China. The results indicated that the ratios of NO2/SO2 presented a responding decline, especially in YRD (-5.01), YH (-3.87), and MYR (-3.84), with the sharp reduction of traffic in post-COVID-19 periods (P3-P4: 2.34 ± 0.94 m/m) comparing with pre-COVID-19 periods (P1-P2: 4.49 ± 2.03 m/m). Whereas the ratios of PM2.5/PM10 increased in P1-P3, then decreased in P4 with relatively higher levels (>0.5) in almost all urban agglomerations. Furthermore, NO2 presented a stronger association with PM2.5/PM10 variation than CO; and PM2.5 with NO2/SO2 variation than PM10. In summary, the economic structure, lockdown measures and meteorological conditions could explain the noteworthy variations in different urban agglomerations. These results would be in great help for improving air quality in the post-epidemic periods.