GNAEMDA: Microbe-Drug Associations Prediction on Graph Normalized Convolutional Network.

The importance of microbe-drug associations (MDA) prediction is evidenced in research. Since traditional wet-lab experiments are both time-consuming and costly, computational methods are widely adopted. However, existing research has yet to consider the cold-start scenarios that commonly seen in real-world clinical research and practices where data of confirmed microbe-drug associations are highly sparse. Therefore, we aim to contribute by developing two novel computational approaches, the GNAEMDA (Graph Normalized Auto-Encoder to predict Microbe-Drug Associations), and a variational extension of the GNAEMDA (called VGNAEMDA), to provide effective and efficient solutions for well-annotated cases and cold-start scenarios. Multi-modal attribute graphs are constructed by collecting multiple features of microbes and drugs, and then input into a graph normalized convolutional network, where a l2-normalization is introduced to avoid the norm-towards-zero tendency of isolated nodes in embedding space. Then the reconstructed graph output by the network is used to infer undiscovered MDA. The difference between the proposed two models lays in the way to generate the latent variables in network. To verify the effectiveness of the two proposed models, we conduct a series of experiments on three benchmark datasets in comparison with six state-of-the-art methods. The comparison results indicate that both GNAEMDA and VGNAEMDA have strong prediction performances in all cases, especially in identifying associations for new microbes or drugs. In addition, we conduct case studies on two drugs and two microbes and find that more than 75% of the predicted associations have been reported in PubMed. The comprehensive experimental results validate the reliability of our models in accurately inferring potential MDA.

Multi-residue detection of pesticides using a sensitive immunochip assay based on nanogold enhancement.

This paper describes the development of a new multiplex immunoassay for simultaneous detection of seven pesticides (triazophos, methyl-parathion, fenpropathrin, carbofuran, thiacloprid, chlorothalonil, and carbendazim). Sixteen pairs of pesticide antibodies and antigens were screened for reactivity and cross-reaction. A microarray chip consisting of seven antigens immobilized on a nitrocellulose membrane was then constructed. Nanogold was employed for labeling and signal amplification to obtain a sensitive colorimetric immunoassay. The direct and indirect detection formats were further compared using primary antibody-gold and secondary antibody-gold conjugates as tracers. An integrated 7-plex immunochip assay based on the indirect model was established and optimized. The detection limits for the pesticides were 0.02-6.45 ng mL(-1), which meets detection requirements for pesticide residues. Naked-eye assessment showed the visual detection limits of the assay ranged from 1 to 100 ng mL(-1). Spiked recovery results demonstrated that the immunochip assay had potential for multi-analysis of pesticide residues in vegetables and fruits. The proposed microarray methodology is a flexible and versatile tool, which can be applied to other competitive multiplex immunoassays for small molecular compounds.

Simultaneous determination of saflufenacil and its two metabolites in soil samples by ultra-performance liquid chromatography with tandem mass spectrometry.

Saflufenacil is a new protoporphyrinogen-IX-oxidase inhibitor herbicide. When used, it can enter the soil and has a high risk to reach and contaminate groundwater and aquatic systems. A rapid and sensitive method of ultra-performance LC with MS/MS was developed for the simultaneous determination of saflufenacil and its two metabolites in soil samples. A modified quick, easy, cheap, effective, rugged, and safe method was applied as the pretreatment procedure. The method was validated by five types of soil samples collected from several regions of China, which all showed good linearity (R(2) ≥ 0.9914) and precision (RSD ≤ 26.2%). The average recoveries of the three analytes ranged between 74.1 and 118.9% at spiking levels of 3-300 µg/kg. The method limits of detection (S/N 3:1) and method limits of quantification (S/N 10:1) achieved are in the ranges of 0.25-2.75 and 0.83-9.16 µg/kg, respectively. This indicated that the developed ultra-performance LC with MS/MS method is a promising analytical tool for monitoring the environmental risks posed by saflufenacil.