Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) and CRISPR-associated (Cas) systems have revolutionized biological and biomedical sciences in many ways. The last few years have also seen tremendous interest in deploying the CRISPR-Cas toolbox for analytical and diagnostic assay development because CRISPR-Cas is one of the most powerful classes of molecular machineries for the recognition and manipulation of nucleic acids. In the short period of development, many CRISPR-enabled assays have already established critical roles in clinical diagnostics, biosensing, and bioimaging. We describe in this review the recent advances and design principles of CRISPR mediated analytical tools with an emphasis on the functional roles of CRISPR-Cas machineries as highly efficient binders and molecular scissors. We highlight the diverse engineering approaches for molecularly modifying CRISPR-Cas machineries and for devising better readout platforms. We discuss the potential roles of these new approaches and platforms in enhancing assay sensitivity, specificity, multiplexity, and clinical outcomes. By illustrating the biochemical and analytical processes, we hope this review will help guide the best use of the CRISPR-Cas toolbox in detecting, quantifying and imaging biologically and clinically important molecules and inspire new ideas, technological advances and engineering strategies for addressing real-world challenges such as the on-going COVID-19 pandemic.

The CRISPR-Cas toolbox for analytical and diagnostic assay development.

The ever-growing global threats to human life caused by the human acute respiratory virus (RV) infections have cost billions of lives, created a significant economic burden, and shaped society for centuries. The timely response to emerging RVs could save human lives and reduce the medical care burden. The development of RV detection technologies is essential for potentially preventing RV pandemic and epidemics. However, commonly used detection technologies lack sensitivity, specificity, and speed, thus often failing to provide the rapid turnaround times. To address this problem, new technologies are devised to address the performance inadequacies of the traditional methods. These emerging technologies offer improvements in convenience, speed, flexibility, and portability of point-of-care test (POCT). Herein, recent developments in POCT are comprehensively reviewed for eight typical acute respiratory viruses. This review discusses the challenges and opportunities of various recognition and detection strategies and discusses these according to their detection principles, including nucleic acid amplification, optical POCT, electrochemistry, lateral flow assays, microfluidics, enzyme-linked immunosorbent assays, and microarrays. The importance of limits of detection, throughput, portability, and specificity when testing clinical samples in resource-limited settings is emphasized. Finally, the evaluation of commercial POCT kits for both essential RV diagnosis and clinical-oriented practices is included.

Advanced Point-of-Care Testing Technologies for Human Acute Respiratory Virus Detection.

Digital CRISPR-based method for the rapid detection and absolute quantification of nucleic acids.

Signal amplification and output of CRISPR/Cas-based biosensing systems: A review.

Interest in CRISPR-Cas12 and CRISPR-Cas13 detection continues to increase as these detection schemes enable the specific recognition of nucleic acids. The fundamental sensitivity limits of these schemes (and their applicability in amplification-free assays) are governed by kinetic rates. However, these kinetic rates remain poorly understood, and their reporting has been inconsistent. We quantify kinetic parameters for several enzymes (LbCas12a, AsCas12a, AapCas12b, LwaCas13a, and LbuCas13a) and their corresponding limits of detection (LoD). Collectively, we present quantification of enzyme kinetics for 14 guide RNAs (gRNAs) and nucleic acid targets for a total of 50 sets of kinetic rate parameters and 25 LoDs. We validate the self-consistency of our measurements by comparing trends and limiting behaviors with a Michaelis-Menten trans-cleavage reaction kinetics model. For our assay conditions, activated Cas12 and Cas13 enzymes exhibit trans-cleavage catalytic efficiencies between order 105 and 106 M-1 s-1. For assays that use fluorescent reporter molecules (ssDNA and ssRNA) for target detection, the kinetic rates at the current assay conditions result in an amplification-free LoD in the picomolar range. The results suggest that successful detection of target requires cleavage (by an activated CRISPR enzyme) of the order of at least 0.1% of the fluorescent reporter molecules. This fraction of reporters cleaved is required to differentiate the signal from the background, and we hypothesize that this required fraction is largely independent of the detection method (e.g., endpoint vs reaction velocity) and detector sensitivity. Our results demonstrate the fundamental nature by which kinetic rates and background signal limit LoDs and thus highlight areas of improvement for the emerging field of CRISPR diagnostics.

Enzyme Kinetics and Detector Sensitivity Determine Limits of Detection of Amplification-Free CRISPR-Cas12 and CRISPR-Cas13 Diagnostics.

Extensive testing is essential to break the transmission of SARS-CoV-2, which causes the ongoing COVID-19 pandemic. Here, we present a CRISPR-based diagnostic assay that is robust to viral genome mutations and temperature, produces results fast, can be applied directly on nasopharyngeal (NP) specimens without RNA purification, and incorporates a human internal control within the same reaction. Specifically, we show that the use of an engineered AsCas12a enzyme enables detection of wildtype and mutated SARS-CoV-2 and allows us to perform the detection step with loop-mediated isothermal amplification (LAMP) at 60-65 °C. We also find that the use of hybrid DNA-RNA guides increases the rate of reaction, enabling our test to be completed within 30 minutes. Utilizing clinical samples from 72 patients with COVID-19 infection and 57 healthy individuals, we demonstrate that our test exhibits a specificity and positive predictive value of 100% with a sensitivity of 50 and 1000 copies per reaction (or 2 and 40 copies per microliter) for purified RNA samples and unpurified NP specimens respectively.

/pdf/an-engineered-crispr-cas12a-variant-and-dna-rna-hybrid-3rtqhmz76z.pdf

An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid COVID-19 testing.

. Current deep video quality assessment (VQA) methods are usually with high computational costs when evaluating high-resolution videos. This cost hinders them from learning better video-quality-related representations via end-to-end training. Existing approaches usually con-sider naive sampling to reduce the computational cost, such as resizing and cropping . However, they obviously corrupt quality-related information in videos and are thus not optimal to learn good representations for VQA. Therefore, there is an eager need to design a new quality-retained sampling scheme for VQA. In this paper, we propose Grid Mini-patch Sampling (GMS), which allows consideration of local quality by sampling patches at their raw resolution and covers global quality with contextual relations via mini-patches sampled in uniform grids. These mini-patches are spliced and aligned temporally, named as fragments . We further build the Fragment Attention Network (FANet) specially designed to ac-commodate fragments as inputs. Consisting of fragments and FANet, the proposed FrAgment Sample Transformer for VQA ( FAST-VQA ) enables efficient end-to-end deep VQA and learns effective video-quality-related representations. It improves state-of-the-art accuracy by around 10% while reducing 99 . 5% FLOPs on 1080P high-resolution videos. The newly learned video-quality-related representations can also be transferred into smaller VQA datasets and boost the performance on these scenarios. Extensive experiments show that FAST-VQA has good performance on inputs of various resolutions while retaining high efficiency. We publish our code at https://github.com/timothyhtimothy/FAST-VQA .

FAST-VQA: Efficient End-to-end Video Quality Assessment with Fragment Sampling

We study the problem of image aesthetic assessment (IAA) and aim to automatically predict the image aesthetic quality in the form of discrete distribution, which is particularly important in IAA due to its nature of having possibly higher diversification of agreement for aesthetics. Previous works show the effectiveness of utilizing object-agnostic attention mechanisms to selectively concentrate on more contributive regions for IAA, e.g., attention is learned to weight pixels of input images when inferring aesthetic values. However, as suggested by some neuropsychology studies, the basic units of human attention are visual objects, i.e., the trace of human attention follows a series of objects. This inspires us to predict contributions of different regions at object level for better aesthetics evaluation. With our framework, region-of-interests (RoIs) are proposed by an object detector, and each RoI is associated with a regional feature vector. Then the contribution of each regional feature to the aesthetics prediction is adaptively determined. To the best of our knowledge, this is the first work modeling object-level attention for IAA and experimental results confirm the superiority of our framework over previous relevant methods.

/pdf/object-level-attention-for-aesthetic-rating-distribution-4lyto9z0rw.pdf

Object-level Attention for Aesthetic Rating Distribution Prediction

In this work, we investigate deep learning based solutions to blind quality assessment of stitched panoramic images (SPI). The main problem to tackle is that the ground truth data is usually insufficient. As a result, the learned model can easily overfit data with specific content. Because most distortions of SPIs lie within local regions, the problem cannot be alleviated by commonly-used patch-wise training, which assumes local quality equals global quality. We propose a multi-task learning strategy which encourages learned representation to be less dependent on image content. A siamese network with two weight-shared CNN branches is trained to simultaneously compare the quality of two images of the same scene and predict the quality score of each image. Since two images of the same scene are processed by the same CNN, the CNN tends to find their quality differences instead of content differences under the constraint of the quality ranking objective. Because two tasks share the same representations learned by the CNN, the regression task can be further benefited from the quality-sensitive representations. Extensive experiments demonstrate the effectiveness of the proposed model and its superiority over existing SPI quality assessment methods.

/pdf/content-dependency-reduction-with-multi-task-learning-in-2w20xs9ivx.pdf

Content-Dependency Reduction With Multi-Task Learning In Blind Stitched Panoramic Image Quality Assessment

The temporal relationships between frames and their influences on video quality assessment (VQA) are still under-studied in existing works. These relationships lead to two important types of effects for video quality. Firstly, some temporal variations (such as shaking, flicker, and abrupt scene transitions) are causing temporal distortions and lead to extra quality degradations, while other variations (e.g. those related to meaningful happenings) do not. Secondly, the human visual system often has different attention to frames with different contents, resulting in their different importance to the overall video quality. Based on prominent time-series modeling ability of transformers, we propose a novel and effective transformer-based VQA method to tackle these two issues. To better differentiate temporal variations and thus capture the temporal distortions, we design a transformer-based Spatial-Temporal Distortion Extraction (STDE) module. To tackle with temporal quality attention, we propose the encoder-decoder-like temporal content transformer (TCT). We also introduce the temporal sampling on features to reduce the input length for the TCT, so as to improve the learning effectiveness and efficiency of this module. Consisting of the STDE and the TCT, the proposed Temporal Distortion-Content Transformers for Video Quality Assessment (DisCoVQA) reaches state-of-the-art performance on several VQA benchmarks without any extra pre-training datasets and up to 10% better generalization ability than existing methods. We also conduct extensive ablation experiments to prove the effectiveness of each part in our proposed model, and provide visualizations to prove that the proposed modules achieve our intention on modeling these temporal issues. We will publish our codes and pretrained weights later.

Jingwen Hou

Papers

An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid COVID-19 testing.

FAST-VQA: Efficient End-to-end Video Quality Assessment with Fragment Sampling

Object-level Attention for Aesthetic Rating Distribution Prediction

Content-Dependency Reduction With Multi-Task Learning In Blind Stitched Panoramic Image Quality Assessment

DisCoVQA: Temporal Distortion-Content Transformers for Video Quality Assessment