Virus Detection and Identification in Minutes Using Single-Particle Imaging and Deep Learning.

The increasing frequency and magnitude of viral outbreaks in recent decades, epitomized by the COVID-19 pandemic, has resulted in an urgent need for rapid and sensitive diagnostic methods. Here, we present a methodology for virus detection and identification that uses a convolutional neural network to distinguish between microscopy images of fluorescently labeled intact particles of different viruses. Our assay achieves labeling, imaging, and virus identification in less than 5 min and does not require any lysis, purification, or amplification steps. The trained neural network was able to differentiate SARS-CoV-2 from negative clinical samples, as well as from other common respiratory pathogens such as influenza and seasonal human coronaviruses. We were also able to differentiate closely related strains of influenza, as well as SARS-CoV-2 variants. Additional and novel pathogens can easily be incorporated into the test through software updates, offering the potential to rapidly utilize the technology in future infectious disease outbreaks or pandemics. Single-particle imaging combined with deep learning therefore offers a promising alternative to traditional viral diagnostic and genomic sequencing methods and has the potential for significant impact.

Viral detection and identification in 20 min by rapid single-particle fluorescence in-situ hybridization of viral RNA.

The increasing risk from viral outbreaks such as the ongoing COVID-19 pandemic exacerbates the need for rapid, affordable and sensitive methods for virus detection, identification and quantification; however, existing methods for detecting virus particles in biological samples usually depend on multistep protocols that take considerable time to yield a result. Here, we introduce a rapid fluorescence in situ hybridization (FISH) protocol capable of detecting influenza virus, avian infectious bronchitis virus and SARS-CoV-2 specifically and quantitatively in approximately 20 min, in virus cultures, combined nasal and throat swabs with added virus and likely patient samples without previous purification. This fast and facile workflow can be adapted both as a lab technique and a future diagnostic tool in enveloped viruses with an accessible genome.

Targeted kinase inhibition relieves slowness and tremor in a Drosophila model of LRRK2 Parkinson's disease.

In a number of Drosophila models of genetic Parkinson's disease (PD) flies climb more slowly than wild-type controls. However, this assay does not distinguish effects of PD-related genes on gravity sensation, "arousal", central pattern generation of leg movements, or muscle. To address this problem, we have developed an assay for the fly proboscis extension response (PER). This is attractive because the PER has a simple, well-identified reflex neural circuit, in which sucrose sensing neurons activate a pair of "command interneurons", and thence motoneurons whose activity contracts the proboscis muscle. This circuit is modulated by a single dopaminergic neuron (TH-VUM). We find that expressing either the G2019S or I2020T (but not R1441C, or kinase dead) forms of human LRRK2 in dopaminergic neurons reduces the percentage of flies that initially respond to sucrose stimulation. This is rescued fully by feeding l-DOPA and partially by feeding kinase inhibitors, targeted to LRRK2 (LRRK2-IN-1 and BMPPB-32). High-speed video shows that G2019S expression in dopaminergic neurons slows the speed of proboscis extension, makes its duration more variable, and increases the tremor. Testing subsets of dopaminergic neurons suggests that the single TH-VUM neuron is likely most important in this phenotype. We conclude the Drosophila PER provides an excellent model of LRRK2 motor deficits showing bradykinesia, akinesia, hypokinesia, and increased tremor, with the possibility to localize changes in neural signaling.