A Knowledge-Driven Self-Supervised Approach for Molecular Generation.

Due to the great successes of Graph Neural Networks (GNN) in numerous fields, growing research interests have been devoted to applying GNN to molecular learning tasks. The molecule structure can be naturally represented as graphs where atoms and bonds refer to nodes and edges respectively. However, the atoms are not haphazardly stacked together but combined into various spatial geometries. Meanwhile, since chemical reactions mainly occur in substructures such as functional groups, the substructure plays a decisive role in the molecule's properties. Therefore, directly applying GNN to molecular representation learning could ignore the molecular spatial structure and the substructure properties which in turn degrades the performance of downstream tasks. In this paper, we propose Knowledge-Driven Self-Supervised Model for Molecular Representation Learning (KSMRL) to address above problems. The KSMRL consists of two major pathways: (1) the Spatial Information (SI) based pathway which preserves the spatial information of molecular structure, (2) the Subgraph Constraint (SC) based pathway which retains the properties of substructures into the molecular representation. In this manner, both the atomic level and substructure level information can be included in modeling. According to the experimental results on multiple datasets, the proposed KSMRL can generate discriminative molecular representations. In molecular generation tasks, KSMRL combined with Autoregressive Flow (AF) models or Discrete Flow (DF) models outperforms the state-of-the-art baselines over all datasets. In addition, we demonstrate the effectiveness of KSMRL with property optimization experiments. To indicate the ability of predicting specified potential Drug-Target Interactions (DTIs), a case study for discriminating the interactions between molecule generated by KSMRL and targets is also given.

Can the transformation of trade patterns achieve carbon reduction? Empirical evidence from a panel of 201 Chinese cities during the period 2000-2016.

How the transformation of trade patterns affects carbon emissions is an important topic in the field of trade and the environment. Based on the panel data of Chinese cities from 2000 to 2016, this paper uses fixed effect model, instrumental variable model and threshold estimation model to investigate the impact of the transformation of export trade patterns on carbon emissions and its internal mechanism. We find that during the investigation period, the transformation of export trade pattern has a significant "U"-shaped nonlinear impact on carbon emissions, and the research results still hold after a series of robustness checks and instrumental variable estimation; through the mechanism analysis of the threshold effect, it is found that in the process of the impact of trade patterns transformation on carbon emissions, there are double thresholds for energy structure transition; the nonlinear impact of export trade pattern transformation on carbon emissions has significant heterogeneity in different regions, city categories, and different periods; the mechanism analysis results show that industrial structure upgrading, industrial co-agglomeration, and green technology innovation are the main paths that the transformation of export trade patterns affects carbon emissions. Under the dual goals of fulfilling China's commitment to "carbon peak in 2030 and carbon neutrality in 2060" and implementing corporate social governance to promote green and sustainable development, this paper will provide suggestions for realizing the goal of low-carbon emission reduction from the perspective of the transformation of export trade patterns.

Environmental governance effect of the transformation of export trade mode: empirical evidence from 194 cities in China.

As one of the developing countries, China's export trade mode (ETM) has gradually shifted from processing trade to general trade. Is the deterioration of China's environmental pollution caused by the transformation of ETM? Based on the panel data from 194 cities in China from 2000 to 2016, this paper investigates the impact of ETM transformation on the environmental pollution and its internal mechanism. The results show that the ETM is gradually shifting from processing trade to general trade, and environmental pollution will deteriorate first and then improve, that is, showing a significant "inverted U-shaped" relationship between the transformation of ETM and environmental pollution. Through the robustness test of the threshold, and SYS-GMM model, the results are still valid. The mechanism research shows that the upgrading of industrial structure, energy structure, industrial agglomeration, environmental protection investment, and resource allocation are the main mechanisms that the transformation of ETM affects environmental pollution. The conclusions of this study can provide empirical evidence for the process that the environmental pollution level of developing countries deteriorated and then improved during the process of transforming from processing export trade to general export trade.