Improving Chemical Reaction Prediction with Unlabeled Data.

Predicting products of organic chemical reactions is useful in chemical sciences, especially when one or more reactants are new organics. However, the performance of traditional learning models heavily relies on high-quality labeled data. In this work, to utilize unlabeled data for better prediction performance, we propose a method that combines semi-supervised learning with graph convolutional neural networks for chemical reaction prediction. First, we propose a Mean Teacher Weisfeiler-Lehman Network to find the reaction centers. Then, we construct the candidate product set. Finally, we use an Improved Weisfeiler-Lehman Difference Network to rank candidate products. Experimental results demonstrate that, with 400k labeled data, our framework can improve the top-5 accuracy by 0.7% using 35k unlabeled data. When the proportion of unlabeled data increases, the performance gain can be larger. For example, with 80k labeled data and 35k unlabeled data, the performance gain with our framework can be 1.8%.

Circadian protein CLK suppresses transforming growth factor-ß expression in peripheral B cells of nurses with day-night shift rotation.

BACKGROUND AND AIMS: The mechanism of dysfunction of regulatory B cells is unclear. The circadian locomotor output cycles kaput (CLK) regulates immune responses. CLK expression can be increased by alteration of the circadian rhythm. This study tests a hypothesis that alteration of the circadian rhythm, such as engaging the day-night shift rotation (DNSR), interferes with the expression of transforming growth factor (TGF)-ß in B cells (TGFbB cell). METHODS: Peripheral blood samples were collected from DNSR nurses and persons with the regular circadian clock life style (RC). The frequency of TGFbB cells in the blood samples was assessed by flow cytometry. The expression of TGF-beta in B cells was assessed with real time RT-PCR. RESULTS: We observed that the frequency of peripheral TGFbB cells was less in DNSR nurses as compared to RC subjects. The expression of CLK and histone deacetylase 11 in peripheral B cells was higher, the TGF-ß expression was lower, in peripheral B cells of DNSR nurses. Over-expression of CLK repressed the expression of TGF-ß in B cells, which was mediated by HDAC11. CONCLUSIONS: The CLK expression in peripheral B cells is higher in DNSR nurses, which suppresses the expression of TGF-ß in B cells. To regulate the expression of CLK during the circadian clock alteration needs to be further investigated.

Collective human mobility pattern from taxi trips in urban area.

We analyze the passengers' traffic pattern for 1.58 million taxi trips of Shanghai, China. By employing the non-negative matrix factorization and optimization methods, we find that, people travel on workdays mainly for three purposes: commuting between home and workplace, traveling from workplace to workplace, and others such as leisure activities. Therefore, traffic flow in one area or between any pair of locations can be approximated by a linear combination of three basis flows, corresponding to the three purposes respectively. We name the coefficients in the linear combination as traffic powers, each of which indicates the strength of each basis flow. The traffic powers on different days are typically different even for the same location, due to the uncertainty of the human motion. Therefore, we provide a probability distribution function for the relative deviation of the traffic power. This distribution function is in terms of a series of functions for normalized binomial distributions. It can be well explained by statistical theories and is verified by empirical data. These findings are applicable in predicting the road traffic, tracing the traffic pattern and diagnosing the traffic related abnormal events. These results can also be used to infer land uses of urban area quite parsimoniously.