The objective of this study was to screen for novel quorum-sensing inhibitors (QSIs) from traditional Chinese medicines (TCMs) that inhibit bacterial biofilm formation. Six of 46 active components found in TCMs were identified as putative QSIs based on molecular docking studies. Of these, three compounds inhibited biofilm formation by Pseudomonas aeruginosa and Stenotrophomonas maltophilia at a concentration of 200 mM. A fourth compound (emodin) significantly inhibited biofilm formation at 20 mM and induced proteolysis of the quorum-sensing signal receptor TraR in Escherichia coli at a concentration of 3-30 mM. Emodin also increased the activity of ampicillin against P. aeruginosa. Therefore, emodin might be suitable for development into an antivirulence and antibacterial agent. INTRODUCTIONPseudomonas aeruginosa, an opportunistic human pathogen, may cause acute infections in hospitalized patients, can be isolated from the environment, particularly from soil and water, and it regularly contaminates medical devices (Stover et al., 2000). It is also the predominant cause of chronic lung infection in cystic fibrosis patients (Frederiksen et al., 1997) and has recently been recognized as one of the main causes of chronic wound infections (Gjødsbøl et al., 2006). P. aeruginosa can infect patients by producing a wide range of virulence factors, the expression levels of which are tightly regulated. Key to this regulation is cell density-dependent cell-to-cell signalling, which is termed quorum sensing (QS) (Rumbaugh et al., 2000). As in many other bacteria, QS controls secretion of virulence factors (Mittal et al., 2006), biofilm formation (Waters et al., 2008) and the exchange of DNA (Fuqua & Winans, 1996) in P. aeruginosa. The biofilm mode of growth is recognized as an important bacterial trait that is relevant to infections (Costerton et al., 1994). Many infections involve the formation of bacterial biofilms, which are bacterial communities that settle and proliferate on surfaces and are covered by exopolymers (Lewis, 2007). Once established, biofilms are difficult to eradicate and become a source of secondary infection (Jones et al., 2009). Moreover, bacteria embedded in biofilms are more tolerant than planktonic cells of antibiotics (Donlan & Costerton, 2002;Drenkard, 2003). The dose of antibiotics needed in this situation will often exceed the highest deliverable dose, which makes efficient treatment impossible.QS, as a regulatory mechanism, enables bacteria to make collective decisions with respect to the expression of a specific set of genes that involve the production, release and subsequent detection of chemical signalling molecules, such as N-acylhomoserine lactones (AHLs) that are commonly used by Gram-negative bacteria. When the concentration of AHLs reaches a certain threshold level, binding to a receptor molecule (for example, LuxR) is promoted and the activated LuxR-AHL complex forms dimers or polymers, which, in turn, act as transcriptional regulators of target genes in the QS regulon (Parsek & Greenbe...

Inferring molecular structure from Nuclear Magnetic Resonance (NMR) measurements requires an accurate forward model that can predict chemical shifts from 3D structure. Current forward models are limited to specific molecules...

No abstract

No abstract

Automated input generators are widely used for large-scale dynamic analysis and testing of mobile apps. Such input generators must constantly choose which UI element to interact with and how to interact with it, in order to achieve high coverage with a limited time budget. Currently, most input generators adopt pseudo-random or brute-force searching strategies, which may take very long to find the correct combination of inputs that can drive the app into new and important states.In this paper, we propose Humanoid, a deep learning-based approach to automated Android app testing. Our insight is that if we can learn from human-generated interaction traces, it is possible to generate human-like test inputs based on the visual information in the current UI state and the latest state transitions. We design and implement a deep neural network model to learn how end-users would interact with an app (specifically, which UI elements to interact with and how), and show that we can successfully generate human-like inputs for any new UI based on the learned model. We then apply the model to automated testing of Android apps and demonstrate that it is able to reach higher coverage, and faster as well, than the state-of-the-art test input generators.