Unraveling the influence of CRISPR/Cas13a reaction components on enhancing trans-cleavage activity for ultrasensitive on-chip RNA detection.

The CRISPR/Cas13 nucleases have been widely documented for nucleic acid detection. Understanding the intricacies of CRISPR/Cas13's reaction components is pivotal for harnessing its full potential for biosensing applications. Herein, we report on the influence of CRISPR/Cas13a reaction components on its trans-cleavage activity and the development of an on-chip total internal reflection fluorescence microscopy (TIRFM)-powered RNA sensing system. We used SARS-CoV-2 synthetic RNA and pseudovirus as a model system. Our results show that optimizing Mg2+ concentration, reporter length, and crRNA combination significantly improves the detection sensitivity. Under optimized conditions, we detected 100 fM unamplified SARS-CoV-2 synthetic RNA using a microtiter plate reader. To further improve sensitivity and provide a new amplification-free RNA sensing toolbox, we developed a TIRFM-based amplification-free RNA sensing system. We were able to detect RNA down to 100 aM. Furthermore, the TIRM-based detection system developed in this study is 1000-fold more sensitive than the off-coverslip assay. The possible clinical applicability of the system was demonstrated by detecting SARS-CoV-2 pseudovirus RNA. Our proposed sensing system has the potential to detect any target RNA with slight modifications to the existing setup, providing a universal RNA detection platform.

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning.

We report a fast, easy-to-implement, highly sensitive, sequence-specific, and point-of-care (POC) DNA virus detection system, which combines recombinase polymerase amplification (RPA) and CRISPR/Cas12a system for trace detection of DNA viruses. Target DNA is amplified and recognized by RPA and CRISPR/Cas12a separately, which triggers the collateral cleavage activity of Cas12a that cleaves a fluorophore-quencher labeled DNA reporter and generalizes fluorescence. For POC detection, portable smartphone microscopy is built to take fluorescent images. Besides, deep learning models for binary classification of positive or negative samples, achieving high accuracy, are deployed within the system. Frog virus 3 (FV3, genera Ranavirus, family Iridoviridae) was tested as an example for this DNA virus POC detection system, and the limits of detection (LoD) can achieve 10 aM within 40 min. Without skilled operators and bulky instruments, the portable and miniature RPA-CRISPR/Cas12a-SPM with artificial intelligence (AI) assisted classification shows great potential for POC DNA virus detection and can help prevent the spread of such viruses.

Detection of Frog Virus 3 by Integrating RPA-CRISPR/Cas12a-SPM with Deep Learning.

A fast, easy-to-implement, highly sensitive, and point-of-care (POC) detection system for frog virus 3 (FV3) is proposed. Combining recombinase polymerase amplification (RPA) and CRISPR/Cas12a, a limit of detection (LoD) of 100 aM (60.2 copies/µL) is achieved by optimizing RPA primers and CRISPR RNAs (crRNAs). For POC detection, smartphone microscopy is implemented, and an LoD of 10 aM is achieved in 40 min. The proposed system detects four positive animal-derived samples with a quantitation cycle (Cq) value of quantitative PCR (qPCR) in the range of 13 to 32. In addition, deep learning models are deployed for binary classification (positive or negative samples) and multiclass classification (different concentrations of FV3 and negative samples), achieving 100 and 98.75% accuracy, respectively. Without temperature regulation and expensive equipment, the proposed RPA-CRISPR/Cas12a combined with smartphone readouts and artificial-intelligence-assisted classification showcases the great potential for FV3 detection, specifically POC detection of DNA virus.

RCMNet: A deep learning model assists CAR-T therapy for leukemia.

Acute leukemia is a type of blood cancer with a high mortality rate. Current therapeutic methods include bone marrow transplantation, supportive therapy, and chemotherapy. Although a satisfactory remission of the disease can be achieved, the risk of recurrence is still high. Therefore, novel treatments are demanding. Chimeric antigen receptor-T (CAR-T) therapy has emerged as a promising approach to treating and curing acute leukemia. To harness the therapeutic potential of CAR-T cell therapy for blood diseases, reliable cell morphological identification is crucial. Nevertheless, the identification of CAR-T cells is a big challenge posed by their phenotypic similarity with other blood cells. To address this substantial clinical challenge, herein we first construct a CAR-T dataset with 500 original microscopy images after staining. Following that, we create a novel integrated model called RCMNet (ResNet18 with Convolutional Block Attention Module and Multi-Head Self-Attention) that combines the convolutional neural network (CNN) and Transformer. The model shows 99.63% top-1 accuracy on the public dataset. Compared with previous reports, our model obtains satisfactory results for image classification. Although testing on the CAR-T cell dataset, a decent performance is observed, which is attributed to the limited size of the dataset. Transfer learning is adapted for RCMNet and a maximum of 83.36% accuracy is achieved, which is higher than that of other state-of-the-art models. This study evaluates the effectiveness of RCMNet on a big public dataset and translates it to a clinical dataset for diagnostic applications.

Pigment analysis based on a line-scanning fluorescence hyperspectral imaging microscope combined with multivariate curve resolution.

A rapid and cost-effective system is vital for the detection of harmful algae that causes environmental problems in terms of water quality. The approach for algae detection was to capture images based on hyperspectral fluorescence imaging microscope by detecting specific fluorescence signatures. With the high degree of overlapping spectra of algae, the distribution of pigment in the region of interest was unknown according to a previous report. We propose an optimization method of multivariate curve resolution (MCR) to improve the performance of pigment analysis. The reconstruction image described location and concentration of the microalgae pigments. This result indicated the cyanobacterial pigment distribution and mapped the relative pigment content. In conclusion, with the advantage of acquiring two-dimensional images across a range of spectra, HSI conjoining spectral features with spatial information efficiently estimated specific features of harmful microalgae in MCR models.