Search

Many traditional Chinese medicines (TCMs) have been confirmed to have antibacterial activities. However, very few substances have been found to be active against Gram-negative bacteria. This study aimed to identify antimicrobial activity substances against Gram-negative bacteria from fourteen TCMs. Fourteen TCMs with antibacterial potential were chosen for quantitative extraction and antibacterial activity assay, and the plant with the highest activity against Escherichia coli was selected to construct the component-target network. The following virtual screening and enzyme inhibition experiments were performed to analyse the antibacterial mechanisms of the compounds from Galla chinensis. The chemical constituents of Galla chinensis were identified by chemical fingerprinting. 1, 2, 3, 4, 6-Penta-O-galloyl-β-D-glucose (PGG) from Galla chinensis exhibited significant inhibition activity against adenylyl transferase (ATase) of E. coli and antibacterial activity against E. coli. Meanwhile, PGG was identified in the Galla chinensis ethanol extract as the abundant ingredient with a high content of 1.95% (w/w). PGG enriched in Galla chinensis is a promising natural antibiotic with the mode of action inhibiting ATase activity. To our knowledge, this is the first study attributing the antibacterial activity of PGG to its affinity with ATase. [ABSTRACT FROM AUTHOR]

Extreme learning machine (ELM) is a popular learning algorithm for single hidden layer feedforward networks (SLFNs). It was originally proposed with the inspiration from biological learning and has attracted massive attentions due to its adaptability to various tasks with a fast learning ability and efficient computation cost. As an effective sparse representation method, orthogonal matching pursuit (OMP) method can be embedded into ELM to overcome the singularity problem and improve the stability. Usually OMP recovers a sparse vector by minimizing a least squares (LS) loss, which is efficient for Gaussian distributed data, but may suffer performance deterioration in presence of non-Gaussian data. To address this problem, a robust matching pursuit method based on a novel kernel risk-sensitive loss (in short KRSLMP) is first proposed in this paper. The KRSLMP is then applied to ELM to solve the sparse output weight vector, and the new method named the KRSLMP-ELM is developed for SLFN learning. Experimental results on synthetic and real-world data sets confirm the effectiveness and superiority of the proposed method. [ABSTRACT FROM AUTHOR]