This work is supported by the National Natural Science Foundation of China 31970752, Science, Technology, Innovation Commission of Shenzhen Municipality JCYJ20190809180003689, JSGG20200225150707332, JSGG20191129110812708, WDZC20200820173710001; Shenzhen Bay Laboratory Open Funding, SZBL2020090501004; China Postdoctoral Science Foundation 2020M680023; and General Administration of Customs of the People&#39;s Republic of China 2021HK007.

1. Processing of samples
<ol>
	<li>Take Frog Virus 3 (FV3, genera Ranavirus, family Iridoviridae), a double-stranded DNA virus. Select the major capsid (mcp) gene as the target for the detection of FV3 since it is highly conserved and usually regarded as the target for ranavirus detection. The target sequence selected is shown in Table 1. 
	NOTE: Frog Virus 3 is taken as an example in this protocol.</li>
	<li>For target DNA fragments preparation, use DNA...

<ol>
	<li>Gire, S. K., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Genomic+surveillance+elucidates+Ebola+virus+origin+and+transmission+during+the+2014+outbreak.">Genomic surveillance elucidates Ebola virus origin and transmission during the 2014 outbreak.</a> Science. 345 (6202), 1369-1372 (2014).</li><li>The Ebola Outbreak Epidemiology Team. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Outbreak+of+Ebola+virus+disease+in+the+Democratic+Republic+of+the+Congo,+April-May+2018:+an+epidemiological+study.">Outbreak of Ebola virus disease in the Democratic Republic of the Congo, April-May 2018: an epidemiological study.</a> Lancet. 392 (10143), 213-221 (2018).</li><li>Zumla, A., Hui, D. S., Perlman, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Middle+East+respiratory+syndrome.">Middle East respiratory syndrome.</a> Lancet. 386 (9997), 995-1007 (2015).</li><li>Plourde, A. R., Bloch, E. M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+literature+review+of+Zika+virus.">A literature review of Zika virus.</a> Emerg Infect Dis. 22 (7), 1185-1192 (2016).</li><li>Yuan, X., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Current+and+perspective+diagnostic+techniques+for+COVID-19.">Current and perspective diagnostic techniques for COVID-19.</a> ACS Infect Dis. 6 (8), 1998-2016 (2020).</li><li>Minhaj, F. S., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Monkeypox+outbreak+-+nine+states,+May+2022.">Monkeypox outbreak - nine states, May 2022.</a> MMWR Morb Mortal Wkly Rep. 71 (23), 764-769 (2022).</li><li>Bao, M., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Challenges+and+opportunities+for+clustered+regularly+interspaced+short+palindromic+repeats+based+molecular+biosensing.">Challenges and opportunities for clustered regularly interspaced short palindromic repeats based molecular biosensing.</a> ACS Sens. 6 (7), 2497-2522 (2021).</li><li>Broughton, J. P., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=CRISPR-Cas12-based+detection+of+SARS-CoV-2.">CRISPR-Cas12-based detection of SARS-CoV-2.</a> Nat Biotechnol. 38 (7), 870-874 (2020).</li><li>Chen, J. S., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=CRISPR-Cas12a+target+binding+unleashes+indiscriminate+single-stranded+DNase+activity.">CRISPR-Cas12a target binding unleashes indiscriminate single-stranded DNase activity.</a> Science. 360 (6387), 436-439 (2018).</li><li>Gootenberg, J. S., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Nucleic+acid+detection+with+CRISPR-Cas13a/C2c2.">Nucleic acid detection with CRISPR-Cas13a/C2c2.</a> Science. 356 (6336), 438-442 (2017).</li><li>Kellner, M. J., Koob, J. G., Gootenberg, J. S., Abudayyeh, O. O., Zhang, F. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=SHERLOCK:+nucleic+acid+detection+with+CRISPR+nucleases.">SHERLOCK: nucleic acid detection with CRISPR nucleases.</a> Nat Protoc. 14 (10), 2986-3012 (2019).</li><li>Mukama, O., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=An+ultrasensitive+and+specific+point-of-care+CRISPR/Cas12+based+lateral+flow+biosensor+for+the+rapid+detection+of+nucleic+acids.">An ultrasensitive and specific point-of-care CRISPR/Cas12 based lateral flow biosensor for the rapid detection of nucleic acids.</a> Biosens Bioelectron. 159, 112143 (2020).</li><li>Schwank, G., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Functional+repair+of+CFTR+by+CRISPR/Cas9+in+intestinal+stem+cell+organoids+of+cystic+fibrosis+patients.">Functional repair of CFTR by CRISPR/Cas9 in intestinal stem cell organoids of cystic fibrosis patients.</a> Cell Stem Cell. 13 (6), 653-658 (2013).</li><li>Yin, L., Man, S., Ye, S., Liu, G., Ma, L. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=CRISPR-Cas+based+virus+detection:+Recent+advances+and+perspectives.">CRISPR-Cas based virus detection: Recent advances and perspectives.</a> Biosens Bioelectron. 193, 113541 (2021).</li><li>Dronina, J., Bubniene, U. S., Ramanavicius, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+application+of+DNA+polymerases+and+Cas9+as+representative+of+DNA-modifying+enzymes+group+in+DNA+sensor+design+(review).">The application of DNA polymerases and Cas9 as representative of DNA-modifying enzymes group in DNA sensor design (review).</a> Biosens Bioelectron. 175, 112867 (2021).</li><li>Fozouni, P., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Amplification-free+detection+of+SARS-CoV-2+with+CRISPR-Cas13a+and+mobile+phone+microscopy.">Amplification-free detection of SARS-CoV-2 with CRISPR-Cas13a and mobile phone microscopy.</a> Cell. 184 (2), 323-333.e9 (2021).</li><li>Kumar, M., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=FnCas9-based+CRISPR+diagnostic+for+rapid+and+accurate+detection+of+major+SARS-CoV-2+variants+on+a+paper+strip.">FnCas9-based CRISPR diagnostic for rapid and accurate detection of major SARS-CoV-2 variants on a paper strip.</a> eLife. 10, e67130 (2021).</li><li>Lee, R. A., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Ultrasensitive+CRISPR-based+diagnostic+for+field-applicable+detection+of+Plasmodium+species+in+symptomatic+and+asymptomatic+malaria.">Ultrasensitive CRISPR-based diagnostic for field-applicable detection of Plasmodium species in symptomatic and asymptomatic malaria.</a> Proc Natl Acad Sci U S A. 117 (41), 25722-25731 (2020).</li><li>Ganguli, A., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Hands-free+smartphone-based+diagnostics+for+simultaneous+detection+of+Zika,+Chikungunya,+and+Dengue+at+point-of-care.">Hands-free smartphone-based diagnostics for simultaneous detection of Zika, Chikungunya, and Dengue at point-of-care.</a> Biomed Microdevices. 19 (4), 73 (2017).</li><li>Yeo, S. J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Smartphone-based+fluorescent+diagnostic+system+for+highly+pathogenic+H5N1+viruses.">Smartphone-based fluorescent diagnostic system for highly pathogenic H5N1 viruses.</a> Theranostics. 6 (2), 231-242 (2016).</li><li>von Chamier, L., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Democratising+deep+learning+for+microscopy+with+ZeroCostDL4Mic.">Democratising deep learning for microscopy with ZeroCostDL4Mic.</a> Nat Commun. 12 (1), 2276 (2021).</li><li>Shiaelis, N., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Virus+detection+and+identification+in+minutes+using+single-particle+imaging+and+deep+learning.">Virus detection and identification in minutes using single-particle imaging and deep learning.</a> ACS Nano. 17 (1), 697-710 (2020).</li><li>Liu, Y., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Mixed-UNet:+Refined+class+activation+mapping+for+weakly-supervised+semantic+segmentation+with+multi-scale+inference.">Mixed-UNet: Refined class activation mapping for weakly-supervised semantic segmentation with multi-scale inference.</a> Front. Comput. Sci. 4, 1036934 (2022).</li><li>Lawrimore, J., Doshi, A., Walker, B., Bloom, K. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=AI-assisted+forward+modeling+of+biological+structures.">AI-assisted forward modeling of biological structures.</a> Front Cell Dev Biol. 7, 279 (2019).</li><li>Yang, Y., Hu, Y., Zhang, X., Wang, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Two-stage+selective+ensemble+of+CNN+via+deep+tree+training+for+medical+image+classification.">Two-stage selective ensemble of CNN via deep tree training for medical image classification.</a> IEEE Trans Cybern. 52 (9), 9194-9207 (2022).</li><li>Zhang, R., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=RCMNet:+A+deep+learning+model+assists+CAR-T+therapy+for+leukemia.">RCMNet: A deep learning model assists CAR-T therapy for leukemia.</a> Comput Biol Med. 150, 106084 (2022).</li><li>Xie, Y., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Stroke+prediction+from+electrocardiograms+by+deep+neural+network.">Stroke prediction from electrocardiograms by deep neural network.</a> Multimed Tools Appl. 80, 17291-17297 (2021).</li><li>Wang, J., Zhu, H., Wang, S., Zhang, Y. -. D. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+review+of+deep+learning+on+medical+image+analysis.">A review of deep learning on medical image analysis.</a> Mobile Netw Appl. 26, 351-380 (2021).</li><li>Artoni, P., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Deep+learning+of+spontaneous+arousal+fluctuations+detects+early+cholinergic+defects+across+neurodevelopmental+mouse+models+and+patients.">Deep learning of spontaneous arousal fluctuations detects early cholinergic defects across neurodevelopmental mouse models and patients.</a> Proc Natl Acad Sci U S A. 117 (38), 23298-23303 (2020).</li><li>Li, J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=DeepLearnMOR:+a+deep-learning+framework+for+fluorescence+image-based+classification+of+organelle+morphology.">DeepLearnMOR: a deep-learning framework for fluorescence image-based classification of organelle morphology.</a> Plant Physiol. 186 (4), 1786-1799 (2021).</li><li>Yosinski, J., Clune, J., Bengio, Y., Lipson, H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=How+transferable+are+features+in+deep+neural+networks.">How transferable are features in deep neural networks.</a> Proceedings of the 27th International Conference on Neural Information Processing Systems. 2, 3320-3328 (2014).</li><li>He, Q., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=High-throughput+and+all-solution+phase+African+Swine+Fever+Virus+(ASFV)+detection+using+CRISPR-Cas12a+and+fluorescence+based+point-of-care+system.">High-throughput and all-solution phase African Swine Fever Virus (ASFV) detection using CRISPR-Cas12a and fluorescence based point-of-care system.</a> Biosens Bioelectron. 154, 112068 (2020).</li><li>Krizhevsky, A., Sutskever, I., Hinton, G. E. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=ImageNet+classification+with+deep+convolutional+neural+networks.">ImageNet classification with deep convolutional neural networks.</a> Commun. ACM. 60 (6), 84-90 (2017).</li><li>Lawton, S., Viriri, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Detection+of+COVID-19+from+CT+lung+scans+using+transfer+learning.">Detection of COVID-19 from CT lung scans using transfer learning.</a> Comput Intell Neurosci. 2021, 5527923 (2021).</li><li>Lei, Z., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Detection+of+frog+virus+3+by+integrating+RPA-CRISPR/Cas12a-SPM+with+deep+learning.">Detection of frog virus 3 by integrating RPA-CRISPR/Cas12a-SPM with deep learning.</a> ACS Omega. 8 (36), 32555-32564 (2023).</li><li>Chen, Z., Huang, J., Zhang, F., Zhou, Y., Huang, H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Detection+of+shrimp+hemocyte+iridescent+virus+by+recombinase+polymerase+amplification+assay.">Detection of shrimp hemocyte iridescent virus by recombinase polymerase amplification assay.</a> Mol Cell Probes. 49, 101475 (2020).</li><li>Fu, X., Sun, J., Ye, Y., Zhang, Y., Sun, X. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+rapid+and+ultrasensitive+dual+detection+platform+based+on+Cas12a+for+simultaneous+detection+of+virulence+and+resistance+genes+of+drug-resistant+Salmonella.">A rapid and ultrasensitive dual detection platform based on Cas12a for simultaneous detection of virulence and resistance genes of drug-resistant Salmonella.</a> Biosens Bioelectron. 195, 113682 (2022).</li><li>Habimana, J. D., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Mechanistic+insights+of+CRISPR/Cas+nucleases+for+programmable+targeting+and+early-stage+diagnosis:+A+review.">Mechanistic insights of CRISPR/Cas nucleases for programmable targeting and early-stage diagnosis: A review.</a> Biosens Bioelectron. 203, 114033 (2022).</li><li>Liang, Y., Lin, H., Zou, L., Deng, X., Tang, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Rapid+detection+and+tracking+of+Omicron+variant+of+SARS-CoV-2+using+CRISPR-Cas12a-based+assay.">Rapid detection and tracking of Omicron variant of SARS-CoV-2 using CRISPR-Cas12a-based assay.</a> Biosens Bioelectron. 205, 114098 (2022).</li><li>Sivaraman, D., Biswas, P., Cella, L. N., Yates, M. V., Chen, W. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Detecting+RNA+viruses+in+living+mammalian+cells+by+fluorescence+microscopy.">Detecting RNA viruses in living mammalian cells by fluorescence microscopy.</a> Trends Biotechnol. 29 (7), 307-313 (2011).</li><li>Wang, I. H., Burckhardt, C. J., Yakimovich, A., Greber, U. F. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Imaging,+tracking+and+computational+analyses+of+virus+entry+and+egress+with+the+cytoskeleton.">Imaging, tracking and computational analyses of virus entry and egress with the cytoskeleton.</a> Viruses. 10 (4), 166 (2018).</li></ol>

The authors have nothing to disclose.

In this method, we develop a fast, easy-to-implement, highly sensitive, sequence-specific, and POC DNA virus detection system with AI assistance. After obtaining samples, RPA is applied to amplify the target sequence, and then CRISPR/Cas12a can recognize the target DNA and release fluorescence, which enlarges the detection signal. Portable smartphone microscopy is built to take fluorescence images, and deep learning models with transfer learning are used for binary classification of the positive and negative samples' ima...

In recent years, epidemics of infectious diseases caused by different viruses have occurred frequently, including the Ebola virus disease (EVD) epidemic in 20141 and 20182, the Middle East Respiratory Syndrome (MERS) in 20153, the Zika virus disease epidemic in 20154, the Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)5 and the continuing Monkeypox caused by Monkeypox virus (MKPV) in 20226. These sudden outbreaks of epidemic infectious diseases cause a large number of deaths and bring huge economic losses and social unrest. A rapid and accurate detection system is urgently required to quickly diagnose the infection and prevent the virus&#39;s further spread.
Recently, clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated (Cas) proteins have gained worldwide attention and have shown promising results in nucleic acid detection7,8,9,10,11,12,13,14,15. The CRISPR/Cas12a protein, guided by CRISPR RNA (crRNA), binds to and cleaves the target DNA. This activity leads to the release of nonspecific single-stranded DNA (ssDNA), known as trans-cleavage, and can be utilized to enhance the detection signal for nucleic acid detection. Some traditional detection methods like polymerase chain reaction (PCR), quantitative real-time PCR (qPCR), and enzyme-linked immunosorbent assay (ELISA) are complicated, time-consuming, and costly for point-of-care (POC) detection. Our previous work successfully developed an automated, integrated, and cost-effective detection system for the African swine fever virus (ASFV) based on CRISPR/Cas12a technology. In this system, we achieved a detection limit of 1 pM within a 2-h time frame without the need for amplification. The CRISPR/Cas12a system and recombinase polymerase amplification (RPA) are combined to improve the sensitivity and specificity for trace DNA detection. Compared with other isothermal amplification techniques, RPA is simple in design and convenient in operation since it has a shorter reaction time without sophisticated temperature control equipment.
For POC detection of pathogens, instruments such as smartphone microscopy (SPM), handheld fluorimeter, or lateral flow strips are developed for the results readouts16,17,18. SPM captures images through a camera and uploads them to some mobile applications for fast data analysis. Such microscopy makes a portable, cheap, and miniatured signal acquisition system with high sensitivity and has shown advantages in detecting pathogens such as H5N1, the Zika virus, and SARS-CoV-219,20.&#160;Therefore, we build a portable SPM to catch the fluorescence signals triggered by RPA-CRISPR/Cas12a detection of the target DNA virus. The ssDNA reporter probe linking a fluorophore and a quencher will be cleaved when CRISPR/Cas12a recognizes the target DNA virus, and the fluorescence emitted by the fluorophore can be captured by SPM.
Compared to the professional software usually used to obtain the results information from the fluorescence images from SPM21,&#160;some experts use machine learning and deep learning to quantify the concentrations of virus DNA after attaining fluorescence images22, which is more time-consuming. When it comes to classifying medical images, conventional neural networks (CNNs) are often used to learn features from the raw pixelated images in an end-to-end manner23,24,25,26. Popular CNN-based deep learning models like AlexNet, DenseNet-121, and EfficientNet-B7 have been successfully applied in this field27,28. However, obtaining large data sets in specific domains can be challenging, necessitating transfer learning29,30. This approach pre-trains a deep learning model with a large dataset, and the pre-trained model is used as the starting point for a new task with a small dataset. This technique can reduce the need for large data sets, combat overfitting, and reduce training time31. Herein, we use deep learning models with transfer learning for the binary classification of the positive and negative samples&#39; fluorescence images.
In this method, we combine RPA and the CRISPR/Cas12a system for the 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. We build a portable SPM to take the fluorescent images for POC detection and develop deep learning models for binary classification. The schematic of the built POC detection system is shown in Figure 1.&#160;Without skilled operators and bulky instruments, the RPA-CRISPR/Cas12a-SPM with artificial intelligence (AI) assisted classification shows great potential for POC DNA virus detection.
<img alt="Figure 1" class="xfigimg" src="/files/ftp_upload/64833/64833fig01.jpg" /> 
Figure 1: The schematic of the RPA-CRISPR/Cas12-SPM detection system along with AI classification for collected images. The nucleic acids of animal-derived samples are released by PINDBK. The Target DNA of the virus is amplified and recognized specifically by the RPA-CRISPR/Cas12a system. CRISPR/Cas12a bonds with crRNA and the Cas12a-crRNA complex bonds with target DNA, which triggers the collateral cleavage of CRISPR/Cas12a on the ssDNA reporter probes. The fluorophore on the reporter is released, and the fluorescence is detected by a commercialized plate reader or the SPM we build. Three different deep learning models, including AlexNet, DenseNet-121, and EfficientNet-B7 with transfer learning, are used to classify the fluorescence images. This figure is reused with permission from Lei et al.35. <a href="https://www.jove.com/files/ftp_upload/64833/64833fig01large.jpg" target="_blank">Please click here to view a larger version of this figure.</a>

<table border="1" fo:keep-together.within-page="1" fo:keep-with-next.within-page="always">
	<tbody>
		<tr>
			<td>20x amplification</td>
			<td>OLYMPUS</td>
			<td>OPLN20X</td>
			<td></td>
		</tr>
		<tr>
			<td>532 nm green laser</td>
			<td>Thorlabs</td>
			<td>PL201</td>
			<td>with 0.9 mW output power</td>
		</tr>
		<tr>
			<td>535 nm cutoff wavelength</td>
			<td>chrome</td>
			<td>AT535</td>
			<td></td>
		</tr>
		<tr>
			<td>6x DNA loading buffer</td>
			<td>Thermo scientific</td>
			<td>R0611</td>
			<td></td>
		</tr>
		<tr>
			<td>96-well black microplate</td>
			<td>Corning Incorporated</td>
			<td>3603</td>
			<td>Black with flat clear bottom</td>
		</tr>
		<tr>
			<td>Aspherical lens</td>
			<td>Lubang</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>Bandpass filter</td>
			<td>SEMROCK</td>
			<td>FF01-542/27-25</td>
			<td></td>
		</tr>
		<tr>
			<td>Bsu DNA Polymerase</td>
			<td>ATG Biotechnology</td>
			<td>M103</td>
			<td>Large Fragment</td>
		</tr>
		<tr>
			<td>crRNA</td>
			<td>Sangon Biotech</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>DNA fragments</td>
			<td>Sangon Biotech</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>Dichroic holders</td>
			<td>Ruicage</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>Dichroic mirror</td>
			<td>SEMROCK</td>
			<td>FF555-Di03-25x36</td>
			<td>with a cutoff wavelength of 535 nm</td>
		</tr>
		<tr>
			<td>E.Z.N.A Gel Extraction Kit</td>
			<td>Omega Biotek</td>
			<td>D2500-02</td>
			<td></td>
		</tr>
		<tr>
			<td>EnGen Lba Cas12a (Cpf1)</td>
			<td>New England Biolabs (Beijing) LTD</td>
			<td>M0653T</td>
			<td></td>
		</tr>
		<tr>
			<td>Filter holders</td>
			<td>Ruicage</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>Fluorophore-ssDNA-Quencher reporter probes</td>
			<td>Sangon Biotech</td>
			<td>N/A</td>
			<td>TAMRA (carboxy tetramethylrhodamine) as the fluorophore at the 5 ends; BHQ2 (Black Hole Quencher-2) as the quencher at the 3 ends</td>
		</tr>
		<tr>
			<td>GP32</td>
			<td>ATG Biotechnology</td>
			<td>M104</td>
			<td></td>
		</tr>
		<tr>
			<td>ImageJ</td>
			<td>Open-source</td>
			<td>Version 1.53t 24</td>
			<td>Downloaded from https://imagej.nih.gov/ij/&#160;</td>
		</tr>
		<tr>
			<td>Microplate reader</td>
			<td>SPARK, TECAN</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>Multi-Block thermal Cycler PCR instrument</td>
			<td>LongGene</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>NanoDrop 2000/2000c Spectrophotometers</td>
			<td>Thermo Scientific</td>
			<td>ND-2000</td>
			<td></td>
		</tr>
		<tr>
			<td>NEBuffer r2.1</td>
			<td>New England Biolabs (Beijing) LTD</td>
			<td>B6002S</td>
			<td>10x CRISPR/Cas12a Reaction buffer</td>
		</tr>
		<tr>
			<td>Oxygen plasma treatment&#160;</td>
			<td>Electro-Technic&#160; Products</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>Pathogen Inactivate, Nucleic acid extraction-free, Direct-to-PCR Buffer with Proteinase K (PINDBK)</td>
			<td>Ebio</td>
			<td>PINDBK -25mL</td>
			<td></td>
		</tr>
		<tr>
			<td>PCR primer pairs</td>
			<td>Sangon Biotech</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>PDMS</td>
			<td>Dow Corning</td>
			<td>Sylgard 184</td>
			<td></td>
		</tr>
		<tr>
			<td>RPA primer pairs</td>
			<td>Sangon Biotech</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>Smartphone</td>
			<td>Huawei</td>
			<td>Mate10</td>
			<td></td>
		</tr>
		<tr>
			<td>Translation stages</td>
			<td>Ruicage</td>
			<td>N/A</td>
			<td></td>
		</tr>
		<tr>
			<td>Transmitted neutral density filters</td>
			<td>Thorlabs</td>
			<td>ND40A</td>
			<td></td>
		</tr>
		<tr>
			<td>Triplet achromatic lenses</td>
			<td>Thorlabs</td>
			<td>TRH127-020-A</td>
			<td></td>
		</tr>
		<tr>
			<td>UvsX</td>
			<td>ATG Biotechnology</td>
			<td>M105</td>
			<td></td>
		</tr>
		<tr>
			<td>UvsY</td>
			<td>ATG Biotechnology</td>
			<td>M106</td>
			<td></td>
		</tr>
	</tbody>
</table>

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.

This method focuses on a fast, easy-to-implement, highly sensitive, and point-of-care (POC) detection system for DNA viruses. Primer pairs design for the RPA reaction and crRNA design for CRISPR/Cas12a reaction are two of the essential parts since they will affect the efficiency of the RPA-CRISPR/Cas12a reaction and influence the subsequent detection and classification.
In this method, FV3 is regarded as an example of DNA virus detection. Some RPA primer pairs for FV3 are...

Watch this Scientific Journal Video about DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning at JoVE.com

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

We present a protocol that combines recombinase polymerase amplification with a CRISPR/Cas12a system for trace detection of DNA viruses and builds portable smartphone microscopy with an artificial intelligence-assisted classification for point-of-care DNA virus detection.

We report a fast, easy-to-implement, highly sensitive, sequence-specific, and point-of-care (POC) DNA virus detection system, which combines recombinase ...

dna-virus-detection-system-based-on-rpa-crisprcas12a-spm-deep

Research

JoVE Journal

Biology

393 조회수.  Tsinghua-Berkeley Shenzhen Institute. 당사의 프로토콜은 개구리 바이러스 3에 대한 빠르고 매우 민감한 검출 시스템을 도입합니다. 중요한 것은 이 방법을 통해 DNA 바이러스의 현장 진단을 가능하게 하여 전염병 발생을 예방하는 데 중요한 역할을 한다는 것입니다. 이 기술의 주요 장점은 RPA와 CRISPR/Cas12a 시스템의 통합에 있으며, 휴대용 스마트폰 기반 현미경 및 AI 분류로 보완됩니다.이 통합은 감지 효율성...