Particle size distributions in the range of 0.01–10 µm were measured in urban Shanghai in the summer of 2013 using a Wide‐range Particle Spectrometer (WPS). Size‐segregated aerosol samples were collected concurrently using a Micro‐Orifice Uniform Deposit Impactor (MOUDI), which aided our in‐depth understanding of the new particle formation (NPF) mechanism in the polluted Yangtze River Delta area. During the observations, 16 NPF events occurred at high temperatures (~34.7°C) on clear and sunny days. In the ammonium‐poor PM1.0 (particulate matter less than 1.0 µm), sulfate and ammonium accounted for 92% of the total water‐soluble inorganic species. Six aminiums were detected in these MOUDI samples, among which the group of diethylaminium and trimethylaminium (DEAH+ + TMAH+) was the most abundant. The very high level of aminiums (average concentration up to 86.4 ng m−3 in PM1.8), together with highly acidic aerosols, provided insight into the frequent NPF events. The high mass ratio of total aminiums to NH4+ (>0.2 for PM0.056) further highlighted the important role of amines in promoting NPF. The concentration of DEAH+ + TMAH+ in new particles below 180 nm was strongly correlated with aerosol phase acidity, indicating that acid‐base reactions dominated the aminium formation in NPF events. The unexpected enhancement of DEAH+ + TMAH+ on a nonevent day was attributed to the transportation of an SO2 plume. Our results reveal that the heterogeneous uptake of amines is dominated by the acid‐base reaction mechanism, which can effectively contribute to particle growth in NPF events.

No abstract

Abstract. We characterize a representative particulate matter (PM) episode that occurred in Shanghai during winter 2014. Particle size distribution, hygroscopicity, effective density, and single particle mass spectrometry were determined online, along with offline analysis of water-soluble inorganic ions. The mass ratio of SNA / PM 1.0 (sulfate, nitrate, and ammonium) fluctuated slightly around 0.28, suggesting that both secondary inorganic compounds and carbonaceous aerosols contributed substantially to the haze formation, regardless of pollution level. Nitrate was the most abundant ionic species during hazy periods, indicating that NO x contributed more to haze formation in Shanghai than did SO 2 . During the representative PM episode, the calculated PM was always consistent with the measured PM 1.0 , indicating that the enhanced pollution level was attributable to the elevated number of larger particles. The number fraction of the near-hydrophobic group increased as the PM episode developed, indicating the accumulation of local emissions. Three "banana-shaped" particle evolutions were consistent with the rapid increase of PM 1.0 mass loading, indicating that the rapid size growth by the condensation of condensable materials was responsible for the severe haze formation. Both hygroscopicity and effective density of the particles increased considerably with growing particle size during the bananashaped evolutions, indicating that the secondary transformation of NO x and SO 2 was one of the most important contributors to the particle growth. Our results suggest that the accumulation of gas-phase and particulate pollutants under stagnant meteorological conditions and subsequent rapid particle growth by secondary processes were primarily responsible for the haze pollution in Shanghai during wintertime.

No abstract

The discovery of disease-causing genes is a critical step towards understanding the nature of a disease and determining a possible cure for it. In recent years, many computational methods to identify disease genes have been proposed. However, making full use of disease-related (e.g., symptoms) and gene-related (e.g., gene ontology and protein-protein interactions) information to improve the performance of disease gene prediction is still an issue. Here, we develop a heterogeneous disease-gene-related network (HDGN) embedding representation framework for disease gene prediction (called HerGePred). Based on this framework, a low-dimensional vector representation (LVR) of the nodes in the HDGN can be obtained. Then, we propose two specific algorithms, namely, an LVR-based similarity prediction and a random walk with restart on a reconstructed heterogeneous disease-gene network (RW-RDGN), to predict disease genes with high performance. First, to validate the rationality of the framework, we analyze the similarity-based overlap distribution of disease pairs