Search for: All records

Abstract MotivationForecasting the synergistic effects of drug combinations facilitates drug discovery and development, especially regarding cancer therapeutics. While numerous computational methods have emerged, most of them fall short in fully modeling the relationships among clinical entities including drugs, cell lines, and diseases, which hampers their ability to generalize to drug combinations involving unseen drugs. These relationships are complex and multidimensional, requiring sophisticated modeling to capture nuanced interplay that can significantly influence therapeutic efficacy. ResultsWe present a novel deep hypergraph learning method named Heterogeneous Entity Representation for MEdicinal Synergy (HERMES) prediction to predict the synergistic effects of anti-cancer drugs. Heterogeneous data sources, including drug chemical structures, gene expression profiles, and disease clinical semantics, are integrated into hypergraph neural networks equipped with a gated residual mechanism to enhance high-order relationship modeling. HERMES demonstrates state-of-the-art performance on two benchmark datasets, significantly outperforming existing methods in predicting the synergistic effects of drug combinations, particularly in cases involving unseen drugs. Availability and implementationThe source code is available at https://github.com/Christina327/HERMES. 

We explore using LLMs, GPT-4 specifically, to generate draft sentence-level Chinese Uniform Meaning Representations (UMRs) that human annotators can revise to speed up the UMR annotation process. In this study, we use few-shot learning and Think-Aloud prompting to guide GPT-4 to generate sentence-level graphs of UMR. Our experimental results show that compared with annotating UMRs from scratch, using LLMs as a preprocessing step reduces the annotation time by two thirds on average. This indicates that there is great potential for integrating LLMs into the pipeline for complicated semantic annotation tasks. 

We explore using LLMs, GPT-4 specifically, to generate draft sentence-level Chinese Uniform Meaning Representations (UMRs) that human annotators can revise to speed up the UMR annotation process. In this study, we use few-shot learning and Think-Aloud prompting to guide GPT-4 to generate sentence-level graphs of UMR. Our experimental results show that compared with annotating UMRs from scratch, using LLMs as a preprocessing step reduces the annotation time by two thirds on average. This indicates that there is great potential for integrating LLMs into the pipeline for complicated semantic annotation tasks. 

Recent studies have highlighted the underestimated diversity of the genus Diploderma Hallowell, 1861 in the Hengduan Mountain Region in Southwest China, but much of the region remains poorly surveyed for reptile diversity. In this study we describe two new species of Diploderma from the upper Jinsha and middle Yalong River Valley, based on evaluations of morphological, genetic, and distribution data. The two new species are morphologically most similar to D. angustelinea and D. vela, but they can be diagnosed from both recognized taxa and all remaining congeners by a suite of morphological features, particularly the distinct coloration of gular spots. Additionally, both new species either render other recognized species paraphyletic or are allopatric with respect to their morphologically similar congeners. Furthermore, we rediscover D. brevicaudum in the wild for the first time, which was known from historical museum specimens only. We estimate the phylogenetic position of D. brevicaudum within the genus Diploderma based on mitochondrial genealogy, and we provide an expanded diagnosis and comparisons against closely related congeners and provide a detailed description of coloration in life based on newly collected specimens. Our discoveries of the new Diploderma species further highlight the urgent conservation needs of the currently neglected hot-dry valley ecosystems in the Hengduan Mountain Region of China. 

Urban spatial structure is a critical component of urban planning and development, and among the different urban spatial structure strategies, ‘polycentric mega-city region (PMR)’ has recently received great research and public policy interest in China. However, there is a lack of systematic understanding on the spatiality of PMR from a pluralistic perspective. This study aims to fill this gap by investigating the spatiality of PMR in the Yangtze River Delta Urban Agglomeration (YRDUA) using city-level data on gross domestic product (GDP), population share, and urban income growth for the period 2000–2013. The results reveal that economically, the YRDUA is experiencing greater polycentricity, but in terms of social welfare, the region manifests growing monocentricity. We further find that the triple transition framework (marketization, urbanization, and decentralization) can greatly explain the observed patterns. Although the economic goals are accomplished with better spatial linkages and early economic development policies, inequality in spatial distribution of public services and the continuing legacy of central planning remain barriers for the YRDUA to emerge as a successful PMR. The results of this research offer meaningful insights on the impact of polycentric policies in the YRDUA and support policymakers in the implementation of appropriate urban spatial development strategies.