Deep Learning Solution for Quantification of Fluorescence Particles on a Membrane.

The detection and quantification of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) virus particles in ambient waters using a membrane-based in-gel loop-mediated isothermal amplification (mgLAMP) method can play an important role in large-scale environmental surveillance for early warning of potential outbreaks. However, counting particles or cells in fluorescence microscopy is an expensive, time-consuming, and tedious task that only highly trained technicians and researchers can perform. Although such objects are generally easy to identify, manually annotating cells is occasionally prone to fatigue errors and arbitrariness due to the operator's interpretation of borderline cases. In this research, we proposed a method to detect and quantify multiscale and shape variant SARS-CoV-2 fluorescent cells generated using a portable (mgLAMP) system and captured using a smartphone camera. The proposed method is based on the YOLOv5 algorithm, which uses CSPnet as its backbone. CSPnet is a recently proposed convolutional neural network (CNN) that duplicates gradient information within the network using a combination of Dense nets and ResNet blocks, and bottleneck convolution layers to reduce computation while at the same time maintaining high accuracy. In addition, we apply the test time augmentation (TTA) algorithm in conjunction with YOLO's one-stage multihead detection heads to detect all cells of varying sizes and shapes. We evaluated the model using a private dataset provided by the Linde + Robinson Laboratory, California Institute of Technology, United States. The model achieved a mAP@0.5 score of 90.3 in the YOLOv5-s6.

auts2 Features and Expression Are Highly Conserved during Evolution Despite Different Evolutionary Fates Following Whole Genome Duplication.

The AUTS2 gene plays major roles during brain development and is associated with various neuropathologies including autism. Data in non-mammalian species are scarce, and the aim of our study was to provide a comprehensive analysis of auts2 evolution in teleost fish, which are widely used for in vivo functional analysis and biomedical purposes. Comparative genomics in 78 species showed that auts2a and auts2b originate from the teleost-specific whole genome duplication (TGD). auts2a, which is highly similar to human AUTS2, was almost systematically retained following TGD. In contrast, auts2b, which encodes for a shorter protein similar to a short human AUTS2 isoform, was lost more frequently and independently during evolution. RNA-seq analysis in 10 species revealed a highly conserved profile with predominant expression of both genes in the embryo, brain, and gonads. Based on protein length, conserved domains, and expression profiles, we speculate that the long human isoform functions were retained by auts2a, while the short isoform functions were retained by auts2a and/or auts2b, depending on the lineage/species. auts2a showed a burst in expression during medaka brain formation, where it was expressed in areas of the brain associated with neurodevelopmental disorders. Together, our data suggest a strong conservation of auts2 functions in vertebrates despite different evolutionary scenarios in teleosts.

Combining quantitative ADF STEM with SiNx membrane-based MEMS devices: A simulation study with Pt nanoparticles.

Computer simulations are used to assess the influence of a 20-nm-thick SiNx membrane on the quantification of atomic-resolution annular dark-field (ADF) scanning transmission electron microscopy images of Pt nanoparticles. The discussions include the effect of different nanoparticle/membrane arrangements, accelerating voltage, nanoparticle thickness and the presence of adjacent atomic columns on the accuracy with which the number of Pt atoms in each atom column can be counted. The results, which are based on the use of ADF scattering cross-sections, show that an accuracy of better than a single atom is attainable at 200 and 300 kV. At 80kV, the scattering in a typical SiNx membrane is sufficiently strong that the best possible atom counting accuracy is reduced to +/- 2 atoms. The implications of the work for quantitative studies of Pt nanoparticles imaged through SiNx membranes are discussed.