ObjectiveTo analyze the supplementary test results of HIV screening positive samples in Jianyang City, Chengdu from 2019 to 2022, to evaluate different HIV testing methods, and to provide a basis for the development of HIV testing strategies in the local area. MethodsWestern blotting (WB) supplementary test was conducted on 1 172 screening positive samples from the HIV confirmatory laboratory in 2019‒2022 according to the national technical specifications. The samples were tested by the rapid test, enzyme-linked immunoassay (ELISA), and chemiluminescence immunoassay (CLIA). The test results of the three HIV screening methods were collected and a database was established for statistical analysis. ResultsA total of 1 172 samples were tested through supplementary test, of which 1 022 samples were tested positive (87.20%), 75 were uncertain (6.40%), and 75 were negative (6.40%). The positive results of the three different HIV screening methods were consistent with the supplementary test. The rapid test had the highest positively supplementary rate of 88.54%, followed by ELISA of 86.98%, and CLIA of 85.92%. The difference was statistically significant (χ²=9.505, P<0.05). The detection rate of WB band patterns in positive samples were the highest at 100.00% for gp160 and gp120, and lowest at 50.68% and 63.41% for p55 and p17, respectively. The WB band patterns of uncertain samples were mainly gp120 (81.33%) and p24 (46.67%). Among the 75 uncertain samples, 39 were followed up and 29 of which turned positive, with a high positive conversion rate of 74.36%. ConclusionIt is necessary to directly add HIV nucleic acid testing to samples with positive WB supplementary test results and samples with uncertain WB supplementary test results in combination with CLIA, so as to avoid the spread of HIV infection caused by missed detections.

Accurate identification of compound-protein interactions (CPIs) in silico may deepen our understanding of the underlying mechanisms of drug action and thus remarkably facilitate drug discovery and development. Conventional similarity-or docking-based computational methods for predicting CPIs rarely exploit latent features from currently available large-scale unlabeled com-pound and protein data and often limit their usage to relatively small-scale datasets. In the present study, we propose DeepCPI, a novel general and scalable computational framework that combines effective feature embedding (a technique of representation learning) with powerful deep learning methods to accurately predict CPIs at a large scale. DeepCPI automatically learns the implicit yet expressive low-dimensional features of compounds and proteins from a massive amount of unla-beled data. Evaluations of the measured CPIs in large-scale databases, such as ChEMBL and Bind-ingDB, as well as of the known drug-target interactions from DrugBank, demonstrated the superior predictive performance of DeepCPI. Furthermore, several interactions among small-molecule compounds and three G protein-coupled receptor targets (glucagon-like peptide-1 recep-tor, glucagon receptor, and vasoactive intestinal peptide receptor) predicted using DeepCPI were experimentally validated. The present study suggests that DeepCPI is a useful and powerful tool for drug discovery and repositioning. The source code of DeepCPI can be downloaded from https://github.com/FangpingWan/DeepCPI.

Pseudouridine(Ψ)is the most prevalent post-transcriptional RNA modification and is widespread in small cellular RNAs and mRNAs.However,the functions,mechanisms,and precise distribution of Ψs(especially in mRNAs)still remain largely unclear.The landscape of Ψs across the transcriptome has not yet been fully delineated.Here,we present a highly effective model based on a convolutional neural network(CNN),called PseudoUridyLation Site Estimator(PULSE),to analyze large-scale profiling data of Ψ sites and characterize the contextual sequence features of pseudouridylation.PULSE,consisting of two alternatively-stacked convolution and pooling layers followed by a fully-connected neural network,can automatically learn the hidden patterns of pseu-douridylation from the local sequence information.Extensive validation tests demonstrated that PULSE can outperform other state-of-the-art prediction methods and achieve high prediction accu-racy,thus enabling us to further characterize the transcriptome-wide landscape ofΨ sites.We fur-ther showed that the prediction results derived from PULSE can provide novel insights into understanding the functional roles of pseudouridylation,such as the regulations of RNA secondary structure,codon usage,translation,and RNA stability,and the connection to single nucleotide vari-ants.The source code and final model for PULSE are available at https://github.com/mlcb-thu/PULSE.

COVID-19 ; Coronavirus Nucleocapsid Proteins ; Humans ; Nucleocapsid Proteins ; Phosphoproteins ; SARS-CoV-2