CN116497095A - Multiple nucleic acid quantification method, device and medium based on sediment bright field image processing - Google Patents
Multiple nucleic acid quantification method, device and medium based on sediment bright field image processing Download PDFInfo
- Publication number
- CN116497095A CN116497095A CN202310028332.XA CN202310028332A CN116497095A CN 116497095 A CN116497095 A CN 116497095A CN 202310028332 A CN202310028332 A CN 202310028332A CN 116497095 A CN116497095 A CN 116497095A
- Authority
- CN
- China
- Prior art keywords
- nucleic acid
- water
- liquid drop
- droplet
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000001921 nucleic acid quantification Methods 0.000 title claims abstract description 30
- 238000012545 processing Methods 0.000 title claims abstract description 28
- 239000013049 sediment Substances 0.000 title description 6
- 108020004707 nucleic acids Proteins 0.000 claims abstract description 131
- 102000039446 nucleic acids Human genes 0.000 claims abstract description 131
- 150000007523 nucleic acids Chemical class 0.000 claims abstract description 131
- 238000001514 detection method Methods 0.000 claims abstract description 99
- 239000007788 liquid Substances 0.000 claims abstract description 74
- 238000003199 nucleic acid amplification method Methods 0.000 claims abstract description 64
- 230000003321 amplification Effects 0.000 claims abstract description 63
- 239000003153 chemical reaction reagent Substances 0.000 claims abstract description 48
- 238000001556 precipitation Methods 0.000 claims abstract description 43
- 238000003062 neural network model Methods 0.000 claims abstract description 8
- 239000012445 acidic reagent Substances 0.000 claims abstract description 7
- 238000004458 analytical method Methods 0.000 claims abstract description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 21
- 239000007850 fluorescent dye Substances 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 9
- 238000003384 imaging method Methods 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 5
- 238000002156 mixing Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 238000004445 quantitative analysis Methods 0.000 abstract description 5
- 239000000523 sample Substances 0.000 description 33
- 238000006243 chemical reaction Methods 0.000 description 30
- 239000002244 precipitate Substances 0.000 description 28
- 108020004414 DNA Proteins 0.000 description 19
- 238000007397 LAMP assay Methods 0.000 description 19
- 239000000047 product Substances 0.000 description 16
- 238000011002 quantification Methods 0.000 description 14
- 238000012549 training Methods 0.000 description 12
- 238000013135 deep learning Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 239000000758 substrate Substances 0.000 description 8
- 238000009826 distribution Methods 0.000 description 7
- 241000588770 Proteus mirabilis Species 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 235000011180 diphosphates Nutrition 0.000 description 6
- 244000052769 pathogen Species 0.000 description 6
- 108090000623 proteins and genes Proteins 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 238000007847 digital PCR Methods 0.000 description 5
- 238000003908 quality control method Methods 0.000 description 5
- 238000011529 RT qPCR Methods 0.000 description 4
- XPPKVPWEQAFLFU-UHFFFAOYSA-J diphosphate(4-) Chemical compound [O-]P([O-])(=O)OP([O-])([O-])=O XPPKVPWEQAFLFU-UHFFFAOYSA-J 0.000 description 4
- 239000000975 dye Substances 0.000 description 4
- 230000001717 pathogenic effect Effects 0.000 description 4
- 230000004568 DNA-binding Effects 0.000 description 3
- JLVVSXFLKOJNIY-UHFFFAOYSA-N Magnesium ion Chemical compound [Mg+2] JLVVSXFLKOJNIY-UHFFFAOYSA-N 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000027455 binding Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000001917 fluorescence detection Methods 0.000 description 3
- 238000002073 fluorescence micrograph Methods 0.000 description 3
- 238000011901 isothermal amplification Methods 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 229910001425 magnesium ion Inorganic materials 0.000 description 3
- 239000003550 marker Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 108091093088 Amplicon Proteins 0.000 description 2
- 206010006187 Breast cancer Diseases 0.000 description 2
- 208000026310 Breast neoplasm Diseases 0.000 description 2
- 102000053602 DNA Human genes 0.000 description 2
- CSNNHWWHGAXBCP-UHFFFAOYSA-L Magnesium sulfate Chemical compound [Mg+2].[O-][S+2]([O-])([O-])[O-] CSNNHWWHGAXBCP-UHFFFAOYSA-L 0.000 description 2
- 206010028980 Neoplasm Diseases 0.000 description 2
- 241000700605 Viruses Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 208000015181 infectious disease Diseases 0.000 description 2
- 238000009830 intercalation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- -1 pyrophosphate ions Chemical class 0.000 description 2
- 125000006853 reporter group Chemical group 0.000 description 2
- 208000002320 spinal muscular atrophy Diseases 0.000 description 2
- 230000001629 suppression Effects 0.000 description 2
- 238000011269 treatment regimen Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- KWIUHFFTVRNATP-UHFFFAOYSA-N Betaine Natural products C[N+](C)(C)CC([O-])=O KWIUHFFTVRNATP-UHFFFAOYSA-N 0.000 description 1
- 208000035473 Communicable disease Diseases 0.000 description 1
- 102100033072 DNA replication ATP-dependent helicase DNA2 Human genes 0.000 description 1
- 102000016928 DNA-directed DNA polymerase Human genes 0.000 description 1
- 108010014303 DNA-directed DNA polymerase Proteins 0.000 description 1
- 102100038595 Estrogen receptor Human genes 0.000 description 1
- 102100031181 Glyceraldehyde-3-phosphate dehydrogenase Human genes 0.000 description 1
- 108010043121 Green Fluorescent Proteins Proteins 0.000 description 1
- 101000927313 Homo sapiens DNA replication ATP-dependent helicase DNA2 Proteins 0.000 description 1
- 101000882584 Homo sapiens Estrogen receptor Proteins 0.000 description 1
- 101001012157 Homo sapiens Receptor tyrosine-protein kinase erbB-2 Proteins 0.000 description 1
- KWIUHFFTVRNATP-UHFFFAOYSA-O N,N,N-trimethylglycinium Chemical compound C[N+](C)(C)CC(O)=O KWIUHFFTVRNATP-UHFFFAOYSA-O 0.000 description 1
- 238000012408 PCR amplification Methods 0.000 description 1
- 102100030086 Receptor tyrosine-protein kinase erbB-2 Human genes 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 229960003237 betaine Drugs 0.000 description 1
- 239000000872 buffer Substances 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- XZTWHWHGBBCSMX-UHFFFAOYSA-J dimagnesium;phosphonato phosphate Chemical compound [Mg+2].[Mg+2].[O-]P([O-])(=O)OP([O-])([O-])=O XZTWHWHGBBCSMX-UHFFFAOYSA-J 0.000 description 1
- 101150088049 dna2 gene Proteins 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 238000000799 fluorescence microscopy Methods 0.000 description 1
- 238000012632 fluorescent imaging Methods 0.000 description 1
- 238000001215 fluorescent labelling Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 108020004445 glyceraldehyde-3-phosphate dehydrogenase Proteins 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 230000007062 hydrolysis Effects 0.000 description 1
- 238000006460 hydrolysis reaction Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000000338 in vitro Methods 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 229910052943 magnesium sulfate Inorganic materials 0.000 description 1
- 235000019341 magnesium sulphate Nutrition 0.000 description 1
- 108020004999 messenger RNA Proteins 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 239000002773 nucleotide Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000008188 pellet Substances 0.000 description 1
- 238000000206 photolithography Methods 0.000 description 1
- 238000003752 polymerase chain reaction Methods 0.000 description 1
- 239000013641 positive control Substances 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 102000003998 progesterone receptors Human genes 0.000 description 1
- 108090000468 progesterone receptors Proteins 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 230000009870 specific binding Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
- C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
- C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
- C12Q1/6844—Nucleic acid amplification reactions
- C12Q1/6851—Quantitative amplification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Organic Chemistry (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Wood Science & Technology (AREA)
- Zoology (AREA)
- Biophysics (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Databases & Information Systems (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Analytical Chemistry (AREA)
- Computational Linguistics (AREA)
- Biotechnology (AREA)
- Immunology (AREA)
- Microbiology (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Biochemistry (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Genetics & Genomics (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention discloses a multiplex nucleic acid quantification method, a device and a medium based on precipitation bright field image processing, wherein the method comprises the following steps: preparing a detection reagent; wherein the detection reagent comprises a LAMP amplification reagent, a nucleic acid sample and a plurality of primers with different concentrations which are uniformly mixed; generating uniform water-in-oil droplets by using a microfluidic chip; amplifying the water-in-oil droplets by using a thermal cycler to enable the water-in-oil droplets to generate a precipitation product; acquiring a liquid drop image of the water-in-oil liquid drop under a bright field, and carrying out classification prediction on the liquid drop image through a mask area convolution neural network model; calculating the nucleic acid concentration of the nucleic acid sample in the detection reagent based on the result of classification prediction, thereby constructing a multiplex nucleic acid quantitative model; and carrying out quantitative detection analysis on the designated nucleic acid reagent by using the multiplex nucleic acid quantitative model. The invention can realize the quantitative analysis of multiple digital nucleic acids under the condition of bright field, thereby reducing the cost of nucleic acid detection.
Description
Technical Field
The invention relates to the technical field of nucleic acid quantification, in particular to a multiplex nucleic acid quantification method, a device and a medium based on sediment bright field image processing.
Background
Nucleic acid detection plays an important role in applications such as identification of pathogen infection, diagnosis of serious diseases, food safety detection, detection of water environment and the like. For example, the carrier of the virus can be found in time through nucleic acid detection, and scientific basis is provided for the establishment of the propagation chain tracking, prevention and control measures of the virus. In the diagnosis of cancer, genetic testing can discover mutated genetic loci, thereby providing a personalized, accurate treatment regimen.
Current nucleic acid detection methods are mostly based on nucleic acid amplification reactions. The target sequence is copied out in the amplification reaction by designing a primer matched with the target sequence, and whether the target sequence of interest exists in the sample is judged by detecting whether the reaction occurs or not. Real-time fluorescent polymerase chain reaction (qPCR) is the most commonly used method for nucleic acid detection, allowing semi-quantification of nucleic acid concentration. For example, the detection technique used for large-scale nucleic acid screening is qPCR. Although qPCR is widely used in academia and healthcare, quantification of qPCR results requires concentration of a reference standard, thus providing only a relative quantification of nucleic acid concentration, low sensitivity, and complicating the test.
Digital PCR (dPCR) is a nucleic acid detection technique that achieves absolute quantification without the need for a standard curve, with digital droplet PCR (ddPCR) being the most common. The technology mainly comprises three links: nucleic acid dispersion, PCR amplification and signal reading, each of which requires a specialized device. In ddPCR, typically, microfluidic technology is used to generate tens of thousands of uniformly sized and defined water-in-oil droplets into which nucleic acids are diluted and dispersed in multiple so that each droplet contains 0 or 1 nucleic acid. Subsequently, each droplet participates in the PCR reaction simultaneously, which can be done on a conventional PCR instrument, the droplet containing the nucleic acid will fluoresce. After amplification, the fluorescent signal of each droplet is read, usually by a flow strategy, and the number of negative and positive droplets is counted. Since the number of nucleic acid molecules encapsulated in each droplet follows a poisson distribution, the number of target nucleic acids in the sample can be estimated. In ddPCR field, related products mostly comprise a droplet generator and a droplet reader, which are used for realizing the links of nucleic acid dispersion and signal reading respectively, but the popularization is limited due to the high cost.
The multi-target nucleic acid detection can detect a plurality of gene fragments simultaneously, so that the detection efficiency is improved. For example, in infectious disease detection, multi-target detection can determine pathogens relatively quickly, guiding the formulation of more scientific treatment regimens; the multi-gene fragments of the same pathogen are detected simultaneously, so that the accuracy of pathogen detection can be improved. Much research has been devoted to achieving multiplex nucleic acid detection based on digital droplet PCR technology. Multiple targets in digital PCR are currently typically achieved in two ways: the fluorescence wavelength and the fluorescence intensity are adjusted. However, these strategies rely on the generation and reading of fluorescent signals, and require fluorescent dyes or fluorescent probes, light sources, light paths, cameras and other modules matched with the fluorescent signals, so that the equipment cost is high, the volume is large, and the implementation of the portable detector is greatly limited. Therefore, how to realize a multi-target digital nucleic acid detection technology without fluorescent labeling, thereby realizing low-cost and high-sensitivity nucleic acid detection is a problem to be solved by those skilled in the art.
Disclosure of Invention
The embodiment of the invention provides a multiple nucleic acid quantitative method, a device, computer equipment and a storage medium based on sediment bright field image processing, which aim to realize multiple digital nucleic acid quantitative analysis under the bright field condition and reduce the nucleic acid detection cost.
In a first aspect, embodiments of the present invention provide a multiplex nucleic acid quantification method based on precipitated bright field image processing, comprising:
preparing a detection reagent; wherein the detection reagent comprises a LAMP amplification reagent, a nucleic acid sample and a plurality of primers with different concentrations which are uniformly mixed;
generating uniform water-in-oil droplets by using a microfluidic chip;
amplifying the water-in-oil droplets by using a thermal cycler to enable the water-in-oil droplets to generate a precipitation product;
acquiring a liquid drop image of the water-in-oil liquid drop under a bright field, and carrying out classification prediction on the liquid drop image through a mask area convolution neural network model;
calculating the nucleic acid concentration of the detection reagent based on the result of classification prediction, thereby constructing a multiplex nucleic acid quantitative model;
and carrying out quantitative detection analysis on the designated nucleic acid reagent by using the multiplex nucleic acid quantitative model.
In a second aspect, embodiments of the present invention provide a multiplex nucleic acid quantification apparatus based on precipitated bright field image processing, comprising:
A first configuration unit configured to configure a detection reagent; wherein the detection reagent comprises a LAMP amplification reagent, a nucleic acid sample and a plurality of primers with different concentrations which are uniformly mixed;
a droplet generation unit for generating uniform water-in-oil droplets of the detection reagent using a microfluidic chip;
a first amplification unit for amplifying the water-in-oil droplets using a thermal cycler to generate a precipitation product from the water-in-oil droplets;
the image prediction unit is used for acquiring a liquid drop image of the water-in-oil liquid drop under a bright field and carrying out classification prediction on the liquid drop image through a mask area convolution neural network model;
a model construction unit for calculating a nucleic acid concentration for the detection reagent based on a result of the classification prediction, thereby constructing a multiplex nucleic acid quantitative model;
and the quantitative detection unit is used for carrying out quantitative detection analysis on the specified nucleic acid reagent by using the multiple nucleic acid quantitative model.
In a third aspect, an embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the multiple nucleic acid quantification method based on the precipitated bright field image processing according to the first aspect when the computer program is executed.
In a fourth aspect, embodiments of the present invention provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a multiplex nucleic acid quantification method based on precipitated bright field image processing as described in the first aspect.
The embodiment of the invention combines the liquid drop micro-flow control, the digital nucleic acid quantification and the deep learning, thereby realizing the absolute quantification technology of multiple digital nucleic acid detection. Specifically, according to ddLAMP (digital liquid drop loop-mediated isothermal amplification) principle, pyrophosphate which falls off in the amplification process of liquid drops containing nucleic acid molecules can be combined with magnesium ions in a system to generate precipitation, so that flocculent precipitation visible in a bright field is formed, and the more the amplification cycle number is, the more the number of generated DNA molecules is, the more the pyrophosphate falls off, and the larger the generated precipitation is; the amplification cycle number can be adjusted by adjusting the primer concentration in the reaction system, thereby changing the size of the generated precipitate. The embodiment of the invention performs liquid drop typing by using an image processing method based on deep learning, thereby realizing absolute quantification of two nucleic acid targets on the premise of only needing bright field imaging. According to the embodiment of the invention, different concentrations are set for primers of different nucleic acid targets, so that the amplification cycle numbers of different nucleic acids are different, different precipitation product amounts are obtained, then droplets are imaged in a bright field, each droplet is divided through deep learning image processing, the targets in the droplet are judged according to the amount of precipitation products in the droplet, and finally the concentration of each target in a sample is calculated by using Poisson distribution, so that multiplex digital nucleic acid quantitative analysis can be carried out under the bright field condition, and the nucleic acid detection cost is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for quantifying multiple nucleic acids based on a precipitated bright field image process according to an embodiment of the present invention;
FIG. 2 is a schematic block diagram of a multiplex nucleic acid quantifying device based on a precipitated bright field image process according to an embodiment of the present invention;
FIG. 3 is a graph showing the relationship between the substrate concentration and the precipitate size in the example of the present invention;
FIG. 4 is a graph showing the relationship between the amplification time and the size of the pellet in the embodiment of the present invention;
FIG. 5 is a graph showing the relationship between the concentration of the primer and the amount of precipitation in the examples of the present invention;
FIG. 6 is a schematic diagram of droplets containing different amounts of precipitation according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a sample reagent generation flow in an embodiment of the present invention;
FIG. 8 is a schematic representation of a fluorescence map in an embodiment of the invention;
FIG. 9 is a network architecture diagram of a mask region convolutional neural network model in an embodiment of the present invention;
FIG. 10 is a graph showing the relationship between the primer concentration and the precipitate size in the example of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Referring to fig. 1, fig. 1 is a flow chart of a method for quantifying multiple nucleic acids based on a precipitated bright-field image processing according to an embodiment of the present invention, which specifically includes: steps S101 to S106.
S101, preparing a detection reagent; wherein the detection reagent comprises a LAMP amplification reagent, a nucleic acid sample and a plurality of primers with different concentrations which are uniformly mixed;
s102, generating uniform water-in-oil droplets by using a microfluidic chip;
s103, amplifying the water-in-oil droplets by using a thermal cycler so as to enable the water-in-oil droplets to generate a precipitation product;
s104, acquiring a liquid drop image of the water-in-oil liquid drop under a bright field, and carrying out classification prediction on the liquid drop image through a mask area convolutional neural network model;
s105, calculating the nucleic acid concentration of the detection reagent based on the classification prediction result, so as to construct a multiplex nucleic acid quantitative model;
s106, carrying out quantitative detection analysis on the designated nucleic acid reagent by using the multiple nucleic acid quantitative model.
The embodiment of the invention combines the liquid drop micro-flow control, the digital nucleic acid quantification and the deep learning, thereby realizing the absolute quantification technology of multiple digital nucleic acid detection. Specifically, according to ddLAMP (digital liquid drop loop-mediated isothermal amplification) principle, pyrophosphate which falls off in the amplification process of liquid drops containing nucleic acid molecules can be combined with magnesium ions in a system to generate precipitation, so that flocculent precipitation visible in a bright field is formed, and the more the amplification cycle number is, the more the number of generated DNA molecules is, the more the pyrophosphate falls off, and the larger the generated precipitation is; the amplification cycle number can be adjusted by adjusting the primer concentration in the reaction system, thereby changing the size of the generated precipitate. The embodiment of the invention performs liquid drop typing by using an image processing method based on deep learning, thereby realizing absolute quantification of two nucleic acid targets on the premise of only needing bright field imaging. According to the embodiment of the invention, different concentrations are set for primers of different nucleic acid targets, so that the amplification cycle numbers of different nucleic acids are different, different precipitation product amounts are obtained, then droplets are imaged in a bright field, each droplet is divided through deep learning image processing, the targets in the droplet are judged according to the amount of precipitation products in the droplet, and finally the concentration of each target in a sample is calculated by using Poisson distribution, so that multiplex digital nucleic acid quantitative analysis can be carried out under the bright field condition, and the nucleic acid detection cost is reduced.
The existing multi-target digital nucleic acid detection technology relies on multi-channel fluorescence detection, and has the disadvantages of high cost and large equipment volume. Therefore, how to realize absolute quantification of multiple nucleic acid targets on the basis of bright field imaging and research a label-free, portable and multi-target digital nucleic acid detection technology are key problems to be solved by the embodiment of the invention. Aiming at the technical requirements of label-free, multi-target and portable digital nucleic acid quantification, the embodiment of the invention provides a method for realizing absolute quantification of multiple targets on the premise of no fluorescent label by taking a precipitation product of isothermal amplification as a marker, taking bright field imaging instead of fluorescent imaging, analyzing the quantity of the precipitation product in liquid drops through deep learning, and judging the yin-yang of each target in each liquid drop. To implement this technique, the present embodiment sets different concentrations for the primers of multiple (e.g., two) nucleic acid targets, makes the amplification cycle number of the multiple nucleic acids different, thereby obtaining precipitation products of different sizes, then images the droplets in the open field, segments each droplet by deep learning image processing, determines the targets in the droplet according to how much of the precipitation product is in the droplet, and finally calculates the concentration of each target in the sample using poisson distribution.
The embodiment of the invention combines the liquid drop micro-flow control, the digital nucleic acid quantification and the deep learning, fully plays the advantages of the micro-flow control and the deep learning in the aspects of small and accurate detection, realizes a low-cost multi-target digital nucleic acid quantification technology, and has important application value for nucleic acid detection in basic research and clinical examination. And the compatibility with in-vitro diagnostic instrument modules is high, and the method has miniaturization potential and better industrialization prospect.
In the prior art, the current research state of multi-target digital nucleic acid detection based on fluorescence and the current research state of digital nucleic acid detection based on bright field image analysis are specifically as follows:
multi-target digital nucleic acid detection based on fluorescent signals
(1) Based on fluorescence wavelength;
based on the advantage of specificity of fluorescent probes, multiple targets can specifically bind to fluorescent probes and emit different fluorescence by designing one or more customized fluorescent probes for each target. The fluorescent probes are complementary to the target sequence, and different fluorescent reporter groups are modified, so that droplets containing different targets emit fluorescence with different wavelengths after amplification, fluorescent signals are detected by using different fluorescent channels, and finally the positive droplet number and the total droplet number of each fluorescent signal are counted and substituted into a poisson distribution-based nucleic acid quantitative formula to realize the quantification of multiple targets.
The prior art aims at the outer bagRare tumor mutations are identified in bubbles, fluorescent probes containing VIC and FAM are respectively designed to specifically identify wild type and mutant type gene sequences wrapped in liquid drops, after amplification is finished, each liquid drop flows in a micro-channel and excites fluorescence of the liquid drops, four different clusters can be observed in a 2D liquid drop diagram according to fluorescence wavelength, and the four clusters respectively represent four types of liquid drops. The prior art also describes a digital PCR technique Crystal Digital PCR that allows for three-color multiplexing TM After sample distribution and amplification, the drop array is transferred to a detection device to capture images under blue, green, and red channels, and different drop populations are partitioned by setting thresholds.
In addition, to simultaneously detect and quantify the number of copies of the various genes associated with spinal muscular atrophy in a single tube, five fluorescent probes were labeled with different fluorophores, followed by a five-color ddPCR assay. These work demonstrate that multi-target digital nucleic acid detection has great potential for pathogen identification, disease diagnosis and prognosis.
However, in this "one color one target" strategy, the design of the fluorescent probe greatly increases the complexity and cost of the design. At the same time, spectral overlap between fluorescent reporter groups limits the multiplicity based on fluorescence wavelength. In terms of detection, multiplexing is also limited by the instrument's fluorescence channel.
(2) Based on the fluorescent signal intensity gradient;
in addition to TaqMan hydrolysis probes, another strategy for fluorescent detection of amplified products is to use DNA binding/intercalating dyes. The fluorescence emitted by the DNA binding/intercalating dye is proportional to the DNA mass, i.e. longer, more amplified products produce a more intense fluorescent response. Thus, another idea of fluorescence-based multi-target detection is to achieve multiple detection by controlling the amount of amplification of different targets, when the dye binds to double-stranded DNA, the droplets containing the different targets exhibit a fluorescence signal intensity gradient. The prior art constructs ddPCR based on DNA binding dyes to achieve quantification of multiple genes of interest in a single reaction. It is verified that the amplification product quality of each target sequence is influenced by changing the design of reverse primer sequences to control the amplicon length and the primer concentration of different targets, thereby indirectly influencing the fluorescence amplitude of positive droplets to enable the positive droplets to have clear division, and successfully realizing 2-fold detection. Based on the above, the difference of the binding amount of EvaGreen fluorescent dye is caused by manipulating the length of the amplicon, so that the quantification of multiple copy number variation is realized.
The fluorescence intensity is often combined with the fluorescence wavelength to provide higher order multiplexing capability, and more kinds of fluorescence intensity combinations are realized in a limited fluorescence channel by adjusting the concentration and the proportion of the fluorescence probe, so that the limitation of 'one color one target' is overcome. In the prior art, only TaqMan probes containing two fluorophores of VIC and FAM are used, and the 5-nucleotide polymorphism detection aiming at spinal muscular atrophy is realized by changing the concentration of different fluorescent probes with the same color. Further, the prior art developed a Rolling Circle Amplification (RCA) based droplet digital ratio fluorescent coding technique, first, when designing the RCA primers, they designed Padlock probes with different ratios of red/green fluorescent probe binding sites for different targets, which cyclize after specific binding to the target into a circular template for RCA amplification, resulting in the fluorescent probe binding sites being replicated hundreds of thousands and keeping their red/green ratio unchanged. Thus, using only two fluorophores, red and green, they successfully achieved 6-fold detection of 6 pathogens by detecting the magnitude of the red and green fluorescence channels and deducing their corresponding ratios. In addition, the prior art utilizes a dual fluorescence detection channel to absolutely quantify sEV-derived PGR, ESR1, ERBB2 and GAPDH mRNA in the plasma of a breast cancer patient by adjusting the concentrations of the CY5/FAM modified Taqman probes of the 4 target sequences. The four target genes are distinguished according to the combination of the two-color fluorescence amplitudes, and then the four target genes are combined with a machine learning model to assist in screening out the optimal marker for breast cancer detection.
However, these methods based on fluorescence detection have the following drawbacks: the reagent cost is high, the equipment is complex and difficult to portable, and the liquid drop reading time is long. Photomultiplier tubes can accomplish flow-through reading of the droplet fluorescence signal, but at high cost. Research on label-free, portable, multi-target digital nucleic acid quantification techniques is a highly desirable problem.
Digital nucleic acid detection based on bright field images
One idea to solve this problem is to use the visible drop characteristics in the bright field as a marker for detection. Although the method realizes digital nucleic acid detection based on bright field images, the multi-target nucleic acid quantification technology is not realized.
In the above work, multiplex nucleic acid detection based on fluorescent signal intensity gradients demonstrates the feasibility of controlling the number of amplification cycles for different targets by varying the primer concentration for the different targets; nucleic acid detection using loop-mediated isothermal amplification instead of PCR in combination with image processing demonstrated the feasibility of achieving digital nucleic acid detection by LAMP based on bright field image observation of precipitation. However, the current multiplex nucleic acid detection method still relies on fluorescence imaging, so that miniaturization of detection equipment is difficult to achieve, and detection cost and complexity are improved. Meanwhile, multiplex nucleic acid detection based on bright field imaging has not been achieved. The multiple nucleic acid quantification based on the sediment bright field image processing provided by the embodiment of the invention can effectively solve the problems, thereby realizing a digital nucleic acid quantification technology without labels, portable and multi-target.
In one embodiment, the step S102 includes:
according to a flow focusing mode, taking the detection reagent as a water phase, and combining an oil phase to generate the water-in-oil liquid drops with different wrapping conditions; the water-in-oil droplets are coated with at least one of no nucleic acid, a first nucleic acid, a second nucleic acid, and a mixture of the first nucleic acid and the second nucleic acid.
In the embodiment, a microfluidic chip is utilized to generate uniform water-in-oil droplets in a flow focusing mode, a water phase is a uniformly mixed detection reagent, and an oil phase is Biorad EvaGreenI oil. Meanwhile, a plurality of droplet wrapping conditions are set, for example, four droplet wrapping conditions are set: no nucleic acid, nucleic acid 1, nucleic acid 2, nucleic acid 1+2.
In one embodiment, the time of the amplification is 30min-60min and the temperature of the amplification is 55 ℃ -70 ℃.
In the embodiment, when the thermal cycler is adopted to amplify the water-in-oil liquid drop, the amplification time is set to be 30-60 min, and the amplification temperature is set to be 55-70 ℃.
In a specific application scenario, the LAMP reaction system suitable for the present embodiment is determined by the following.
(1) Substrate concentration in LAMP reaction System
Besides the primer and the sample to be detected, the LAMP reaction system also comprises a reaction substrate consisting of buffers such as magnesium sulfate, betaine and the like, dNTPs, bst DNA polymerase and fluorescent dye. When the substrate is insufficient in the reaction system, the amount of the produced precipitate is limited by the amount of the substrate, and the amount of the precipitate cannot be increased with increasing the primer concentration. As shown in FIG. 3, the amount of precipitate generated at a relatively high primer concentration (55.8 ng/. Mu.L) can be significantly increased by increasing the substrate concentration in the reaction system to 1.5 times. The increase of the precipitation amount generated under the high primer concentration is beneficial to enlarging the difference between the generation of precipitation when the primer concentration is low and the primer concentration is high, so that the deep learning model is beneficial to more accurately carrying out liquid drop typing through the precipitation amount. The present example thus uses a reaction system comprising a higher concentration of substrate.
(2) LAMP reaction amplification time
The amplification time of LAMP reaction is generally between 30min and 60min, and the amplification time required for detecting different samples is slightly different. Too short amplification time can lead to insufficient reaction, unobvious reaction results (too small difference between fluorescent brightness and background noise or too small precipitation amount to be visible), high false negative rate of detection results, lower true value of detection results and unfavorable generation of differential precipitation amount. The amplification time is too long, so that the non-specific amplification probability can be increased, and the detection result is inaccurate.
The present example explores the effect of LAMP amplification time on the amount of precipitate generated, and thus determines the optimal amplification period.
This example uses Proteus mirabilis primers and positive quality control for experiments. The specific operation is as follows: the reaction system was prepared using 18.68 ng/. Mu.L and 62.28 ng/. Mu.L primers, two sets of droplets were photographed after different amplification times, and the optical density values of the precipitates in the droplets were measured. As a result, as shown in FIG. 4, the reaction system of the primer with low concentration produced a precipitate with high concentration in total, and as the amplification time increases, the amount of precipitate in both groups showed a tendency to rise first and then to settle. The experimental results show that: (1) Within the amplification time of 70min, no obvious non-specific amplification phenomenon appears; (2) After 45min of reaction, both groups of precipitation amounts tended to be gentle, so that the amplification time was determined to be 45min in this example; (3) It was again demonstrated how much precipitate was formed after amplification by adjusting the primer concentration.
In a specific embodiment, two nucleic acid labeling methods based on adjusting primer concentration by differentiating the amount of precipitate generated are constructed.
As shown in fig. 5, in order to achieve absolute quantitative detection of two nucleic acids, in this example, primers corresponding to two different nucleic acids are added to the reaction system, and the concentrations of the two primers are different, so that after droplets are generated and isothermal amplification is performed, the amount of precipitate generated by droplets containing DNA corresponding to a low concentration primer is the smallest, the amount of precipitate generated by droplets containing DNA corresponding to a high concentration primer is the largest, and the amount of precipitate generated by droplets containing two kinds of DNA is the largest.
In order to determine the concentrations of the two primers, the following experiments were performed using Proteus mirabilis, and buffalo-derived primers and two positive controls, respectively: preparing a reaction system, wherein two primers are simultaneously added into the reaction system, and the concentrations of the two primers are different; and dividing the reaction system into three parts, respectively adding positive quality control of Proteus mirabilis, positive quality control of buffalo origin and excess of the two positive quality control, wherein the excess of the two positive quality control in the third part is used for ensuring that each liquid drop contains two kinds of DNA at the same time. And carrying out isothermal amplification after generating liquid drops in three groups, shooting three groups of liquid drops after amplification, and observing the precipitation condition in each group of liquid drops. The concentration of the two primers in the system is adjusted to maximize the degree of precipitation in the three sets of droplets. The experimental results show that when the concentration of the Proteus mirabilis primer is 30.03 ng/. Mu.L and the concentration of the buffalo source primer is 65.89 ng/. Mu.L, the three groups of precipitation amounts are better in differentiation degree, and the effect is shown in figure 6.
In one embodiment, the step S104 includes:
selecting target liquid drops from the amplified water-in-oil liquid drops;
a micro-channel imaging method is adopted to read the droplet diameter of the target droplet, and the target droplet is injected into the micro-fluidic chip based on the droplet diameter;
and shooting the target liquid drop in the microfluidic chip through a bright field microscope to obtain the liquid drop image.
In this example, a small amount of droplets (i.e., the target droplets) were taken on a slide glass after the amplification was completed, and the droplet diameter was measured. The target liquid drop is read by adopting a method of imaging in a micro-channel. Specifically, this embodiment uses photolithography and soft etching techniques to process microchannels having a height of about 46.5 μm and a width of about 800 μm, the channels being designed with an inlet and an outlet. Then, the target droplets are injected from the chip inlet, a small amount of oil phase can be injected if the droplets in the chip are too crowded, and a 10x bright field microscope is used to capture the image of the droplets.
In an embodiment, the step S104 further includes:
scanning the liquid drop image through a region proposal network in the mask region convolution neural network model, and generating anchor frames with different sizes and shapes at each pixel point in the liquid drop image;
Screening anchor frames which contain objects and have position accuracy and size reaching preset thresholds as proposals;
generating a feature map for the drop image by a convolutional neural network in the mask region convolutional neural network model, and aligning the feature map with the proposal;
the proposal of full convolution layer alignment is adopted for classification output.
In this embodiment, as shown in fig. 7, the Mask R-CNN (Mask region convolutional neural network) has good performance in the example segmentation task, and the network structure can be simply divided into: convolutional Neural Network (CNN), region proposal network (RPN, regionproposal network), region of interest alignment layer (ROI alignment), full convolutional layer (FCN, fully convolutional network). The network can predict not only the class and bounding box of each region of interest (ROI, region ofinterest), but also generate a mask (mask) at the pixel level location of the object. The Mask R-CNN framework is divided into two phases: the first stage is to scan the image and generate proposal (possibly containing the area of the object), which is completed by RPN, firstly, generating anchor frames with different sizes and shapes at each pixel point, screening out the anchor frames with more accurate positions and sizes and containing the object as proposal; the second stage aligns the proposal with the feature map, classifies it and generates bounding boxes and masks, mainly done by CNNs.
In one embodiment, the multiplex nucleic acid quantification method based on the precipitation bright field image processing further comprises:
preparing a plurality of sample reagents, and respectively mixing different fluorescent dyes with each sample reagent to form different sample liquid drops;
amplifying the sample liquid drop, and collecting a bright field and a fluorescent chart of the sample liquid drop after the amplification is finished;
confirming the boundary of the sample liquid drop in the fluorescent chart and marking the category of the sample liquid drop;
and learning the mask region convolutional neural network model by using the fluorescence map with the marks.
Before the classification prediction is performed on the droplet image through the mask area convolutional neural network model, the mask area convolutional neural network model is constructed and trained in advance, so that the mask area convolutional neural network model is more suitable for the scene corresponding to the embodiment. The specific construction training process is as follows:
(1) Generating training sets and validation sets
In order to learn the content contained in each type of drop and thus give the correct label information when labeling next, the present embodiment uses the fluorescent color as an indication of the true type of each type of drop when generating the training set and the validation set. As shown in fig. 8, the specific operation is: four sets of droplets were generated with different fluorochromes, each of the other three sets containing one fluorescence, one DNA target combination (DNA 1, DNA2, DNA 1+dna2), and two primers with different concentrations, were removed from the non-fluorescent, non-DNA droplets, respectively, with the reagents. After amplification, mixing the four droplets, collecting a bright field and a fluorescence image, and knowing the content wrapped by the droplets through the fluorescence fusion image.
(2) Marking a training set and a verification set:
the droplet boundaries were circled with Labelme and the droplet categories were noted, as shown in FIG. 9. If no fluorescent droplet exists, a negative (representing negative) label is given; blue fluorescent droplets, then the bright field droplets are given to low_positive (representing small precipitates) labels; red fluorescent droplets, then the bright field droplets are given to medium_positive (representing medium precipitation) labels; purple fluorescent droplets are given their bright field droplets to a high_positive (representing large precipitates) tag.
(3) Model training
The precipitation identification and droplet classification uses Mask R-CNN network. The embodiment uses the Mask R-CNN as an image segmentation model, and uses Mask R-CNN codes of Abdula for Tensorflow2 implementation. First, we select Resnet-50 as the backbone model. In order to accelerate the model training speed, the present embodiment reduces the size of the video frame, which is originally 1608×1608 pixels, to 768×768 pixels. In addition, the present embodiment sets the number of regions of interest (ROIs) per training image to 5000, and sets the upper limit of the number of detection of the gold standard example object in training to 1000. The non-maximum suppression threshold in the RPN is set to 0.95, and the maximum number of proposed areas after non-maximum suppression is set to 9000 to produce a sufficient number of proposed areas. The anchor frame is set to 32, 64, 128, 256, 512 pixels in steps of 4,8, 16, 32, 64 pixels in length and aspect ratio of 0.5,1,2. According to the specific condition of the experiment, the upper limit of the object detection quantity of each image in the detection is consistent with the upper limit of the object detection quantity in the training, and the non-maximum inhibition threshold value in the RPN is set to be 0.1.
In model training, this embodiment uses a random gradient descent optimization algorithm with a learning rate of 0.005. The raw image and associated label and Mask information form training (80%) and test (20%) data for the Mask R-CNN model. The marked data are put into a deep learning network to train weights, and weight files with ideal effects can be obtained by training about 1500 epochs. After the detection is finished, the model outputs the category of the detected liquid drop and the quantity of each category, so that the concentration quantification of the two nucleic acids can be realized through the nucleic acid concentration calculation formula.
In one embodiment, the step S105 includes:
the nucleic acid concentration C was calculated according to the following formula:
C＝-ln P-/V
wherein P represents the proportion of negative droplets, and V represents the volume of each of the water-in-oil droplets.
In this embodiment, loop-mediated isothermal amplification (LAMP) is a nucleic acid amplification method similar to PCR, and has the advantage of being capable of generating visible precipitates in the process of nucleic acid amplification at a constant temperature. When DNA is amplified, free dNTPs are combined with a nucleic acid single strand, and the released pyrophosphate ions are combined with magnesium ions in the LAMP reagent to generate magnesium pyrophosphate precipitate. In digital droplet nucleic acid detection, whether a droplet contains a DNA molecule can be determined by observing whether a precipitate is generated in the droplet under an open field. Studies have demonstrated that the probability of containing k nucleic acid molecules per droplet follows a poisson distribution. For a uniform volume droplet, assuming that the volume of each droplet is V and the concentration to be found is C, C satisfies the following equation:
C＝-ln P-/V
Wherein P-represents the proportion of negative droplets. By the above formula, we can obtain the nucleic acid concentration C in the sample by counting the ratio of the number of droplets containing the precipitate to the total number of droplets.
According to the LAMP principle, the more the amplification cycles are in the amplification process, the more the number of DNA molecules finally amplified, the more pyrophosphate ions are dropped, and the more the generated precipitation amount is caused. On the other hand, in the case where the reaction substrate is sufficient, theoretically, the larger the number of primers, the larger the number of amplification cycles, and the larger the amount of precipitate generated. Therefore, if two primers corresponding to different DNAs are added to the reaction system and the concentrations of the two primers are made different, the kind of DNA contained in the droplet can be judged by the amount of precipitation.
This example demonstrates that the size of the precipitate formed can be adjusted by adjusting the primer concentration. Specifically, first, 11 kinds of total concentration of Proteus mirabilis primers from 0 ng/. Mu.L to 62 ng/. Mu.L were used, respectively, and after droplets were generated using a reaction reagent containing a Proteus mirabilis positive DNA sample, a loop-mediated isothermal amplification reaction was performed. The optical density value of the precipitate in each droplet was then measured, and the result was that the amount of precipitate was in an ascending trend with an increase in the primer concentration in the range of 0 ng/. Mu.L to 55.8 ng/. Mu.L, as shown in FIG. 10; the primer concentration is continuously increased, and the precipitation amount is stable due to the lack of reaction raw materials such as dNTPs and the like, and is not increased along with the increase of the primer concentration. This demonstrates a high correlation between the amount of precipitate formed and the primer concentration, demonstrating the feasibility of this example to achieve labeling of DNA species by adjusting the amount of precipitate formed by adjusting the primer concentration.
FIG. 2 is a schematic block diagram of a multiplex nucleic acid quantification apparatus 200 based on a bright field image processing of a sediment, the apparatus 200 comprising:
a first configuration unit 201 for configuring a detection reagent; wherein the detection reagent comprises a LAMP amplification reagent, a nucleic acid sample and a plurality of primers with different concentrations which are uniformly mixed;
a droplet generation unit 202 for generating uniform water-in-oil droplets of the detection reagent using a microfluidic chip;
a first amplification unit 203 for amplifying the water-in-oil droplets using a thermal cycler to generate a precipitation product from the water-in-oil droplets;
an image prediction unit 204, configured to obtain a droplet image of the water-in-oil droplet in the open field, and perform classification prediction on the droplet image through a mask area convolutional neural network model;
a model construction unit 205 for calculating a nucleic acid concentration for the detection reagent based on the result of the classification prediction, thereby constructing a multiplex nucleic acid quantitative model;
a quantitative detection unit 206 for performing quantitative detection analysis on the specified nucleic acid reagent using the multiplex nucleic acid quantitative model.
In one embodiment, the droplet generation unit 202 includes:
The flow focusing unit is used for taking the detection reagent as a water phase according to a flow focusing mode and combining an oil phase to generate the water-in-oil liquid drops with different wrapping conditions; the water-in-oil droplets are coated with at least one of no nucleic acid, a first nucleic acid, a second nucleic acid, and a mixture of the first nucleic acid and the second nucleic acid.
In one embodiment, the time of the amplification is 30min-60min and the temperature of the amplification is 55 ℃ -70 ℃.
In an embodiment, the image prediction unit 204 includes:
a target selecting unit for selecting target liquid drops from the amplified water-in-oil liquid drops;
the reading and injecting unit is used for reading the droplet diameter of the target droplet by adopting a micro-channel imaging method and injecting the target droplet into the micro-fluidic chip based on the droplet diameter;
and the image shooting unit is used for shooting the target liquid drop in the microfluidic chip through a bright field microscope to obtain the liquid drop image.
In an embodiment, the image prediction unit 204 further comprises:
an anchor frame generating unit, configured to scan the droplet image through a region proposal network in the mask region convolutional neural network model, and generate anchor frames with different sizes and shapes at each pixel point in the droplet image;
A proposal screening unit for screening an anchor frame containing an object and having both position accuracy and size reaching a preset threshold as a proposal;
a proposal alignment unit for generating a feature map for the droplet image by a convolutional neural network in the mask region convolutional neural network model, and aligning the feature map with the proposal;
and the classification output unit is used for performing classification output by adopting the proposal of the full convolution layer alignment.
In an embodiment, the model construction unit 205 includes:
a concentration calculation unit for calculating a nucleic acid concentration C according to the following formula:
C＝-ln P_/V
wherein P represents the proportion of negative droplets, and V represents the volume of each of the water-in-oil droplets.
In one embodiment, the multiplex nucleic acid quantification apparatus 200 based on the precipitation bright field image processing further comprises:
a second configuration unit for configuring a plurality of sample reagents, and mixing different fluorescent dyes with each sample reagent into different sample droplets;
the fluorescence image collecting unit is used for amplifying the sample liquid drops and collecting the bright field and the fluorescence image of the sample liquid drops after the amplification is finished;
a category marking unit for confirming the boundary of the sample droplet in the fluorescence map and marking the category of the sample droplet;
And the model learning unit is used for learning the mask region convolution neural network model by using the fluorescence map with the mark.
Since the embodiments of the apparatus portion and the embodiments of the method portion correspond to each other, the embodiments of the apparatus portion are referred to the description of the embodiments of the method portion, and are not repeated herein.
The embodiment of the present invention also provides a computer readable storage medium having a computer program stored thereon, which when executed can implement the steps provided in the above embodiment. The storage medium may include: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The embodiment of the invention also provides a computer device, which can comprise a memory and a processor, wherein the memory stores a computer program, and the processor can realize the steps provided by the embodiment when calling the computer program in the memory. Of course, the computer device may also include various network interfaces, power supplies, and the like.
In the description, each embodiment is described in a progressive manner, and each embodiment is mainly described by the differences from other embodiments, so that the same similar parts among the embodiments are mutually referred. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section. It should be noted that it would be obvious to those skilled in the art that various improvements and modifications can be made to the present application without departing from the principles of the present application, and such improvements and modifications fall within the scope of the claims of the present application.
It should also be noted that in this specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. A multiplex nucleic acid quantification method based on a precipitated bright field image processing, comprising:
preparing a detection reagent; wherein the detection reagent comprises a LAMP amplification reagent, a nucleic acid sample and a plurality of primers with different concentrations which are uniformly mixed;
Generating uniform water-in-oil droplets by using a microfluidic chip;
amplifying the water-in-oil droplets by using a thermal cycler to enable the water-in-oil droplets to generate a precipitation product;
acquiring a liquid drop image of the water-in-oil liquid drop under a bright field, and carrying out classification prediction on the liquid drop image through a mask area convolution neural network model;
calculating the nucleic acid concentration of the detection reagent based on the result of classification prediction, thereby constructing a multiplex nucleic acid quantitative model;
and carrying out quantitative detection analysis on the designated nucleic acid reagent by using the multiplex nucleic acid quantitative model.
2. The method for quantitative determination of multiple nucleic acids based on the processing of a sedimentary bright field image according to claim 1, wherein the step of generating uniform water-in-oil droplets of the detection reagent using a microfluidic chip comprises:
according to a flow focusing mode, taking the detection reagent as a water phase, and combining an oil phase to generate the water-in-oil liquid drops with different wrapping conditions; the water-in-oil droplets are coated with at least one of no nucleic acid, a first nucleic acid, a second nucleic acid, and a mixture of the first nucleic acid and the second nucleic acid.
3. The method for quantitative determination of multiple nucleic acids based on the image processing of a sedimented bright-field according to claim 1, wherein the amplification time is 30min to 60min and the amplification temperature is 55 ℃ to 70 ℃.
4. The method for multiplex nucleic acid quantification based on precipitated bright-field image processing of claim 1, wherein the acquiring a droplet image of the water-in-oil droplet in bright-field and classifying and predicting the droplet image by a mask region convolutional neural network model comprises:
selecting target liquid drops from the amplified water-in-oil liquid drops;
a micro-channel imaging method is adopted to read the droplet diameter of the target droplet, and the target droplet is injected into the micro-fluidic chip based on the droplet diameter;
and shooting the target liquid drop in the microfluidic chip through a bright field microscope to obtain the liquid drop image.
5. The method for multiplex nucleic acid quantification based on precipitated bright-field image processing of claim 1, wherein the acquiring a droplet image of the water-in-oil droplet in bright-field and classifying and predicting the droplet image by a mask region convolutional neural network model further comprises:
scanning the liquid drop image through a region proposal network in the mask region convolution neural network model, and generating anchor frames with different sizes and shapes at each pixel point in the liquid drop image;
Screening anchor frames which contain objects and have position accuracy and size reaching preset thresholds as proposals;
generating a feature map for the drop image by a convolutional neural network in the mask region convolutional neural network model, and aligning the feature map with the proposal;
the proposal of full convolution layer alignment is adopted for classification output.
6. The method for multiplex nucleic acid quantification based on the precipitation bright-field image processing according to claim 1, wherein the calculating the nucleic acid concentration for the detection reagent based on the result of the classification prediction, thereby constructing a multiplex nucleic acid quantification model, comprises:
the nucleic acid concentration C was calculated according to the following formula:
C＝-lnP-/V
wherein P represents the proportion of negative droplets, and V represents the volume of each of the water-in-oil droplets.
7. The method for multiplex nucleic acid quantification based on the processing of a sedimentary bright-field image of claim 1, further comprising:
preparing a plurality of sample reagents, and respectively mixing different fluorescent dyes with each sample reagent to form different sample liquid drops;
amplifying the sample liquid drop, and collecting a bright field and a fluorescent chart of the sample liquid drop after the amplification is finished;
confirming the boundary of the sample liquid drop in the fluorescent chart and marking the category of the sample liquid drop;
And learning the mask region convolutional neural network model by using the fluorescence map with the marks.
8. A multiplex nucleic acid quantification device based on a precipitation bright field image processing, comprising:
a first configuration unit configured to configure a detection reagent; wherein the detection reagent comprises a LAMP amplification reagent, a nucleic acid sample and a plurality of primers with different concentrations which are uniformly mixed;
a droplet generation unit for generating uniform water-in-oil droplets of the detection reagent using a microfluidic chip;
a first amplification unit for amplifying the water-in-oil droplets using a thermal cycler to generate a precipitation product from the water-in-oil droplets;
the image prediction unit is used for acquiring a liquid drop image of the water-in-oil liquid drop under a bright field and carrying out classification prediction on the liquid drop image through a mask area convolution neural network model;
a model construction unit for calculating a nucleic acid concentration for the detection reagent based on a result of the classification prediction, thereby constructing a multiplex nucleic acid quantitative model;
and the quantitative detection unit is used for carrying out quantitative detection analysis on the specified nucleic acid reagent by using the multiple nucleic acid quantitative model.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of multiplex nucleic acid quantification based on the processing of a sedimented bright-field image according to any one of claims 1 to 7 when the computer program is executed by the processor.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the multiplex nucleic acid quantification method based on the precipitated bright-field image processing according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310028332.XA CN116497095B (en) | 2023-01-09 | 2023-01-09 | Dual nucleic acid quantification method, device and medium based on sediment bright field image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310028332.XA CN116497095B (en) | 2023-01-09 | 2023-01-09 | Dual nucleic acid quantification method, device and medium based on sediment bright field image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116497095A true CN116497095A (en) | 2023-07-28 |
| CN116497095B CN116497095B (en) | 2024-07-12 |
Family
ID=87315512
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310028332.XA Active CN116497095B (en) | 2023-01-09 | 2023-01-09 | Dual nucleic acid quantification method, device and medium based on sediment bright field image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116497095B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117523559A (en) * | 2024-01-08 | 2024-02-06 | 深圳赛陆医疗科技有限公司 | Base recognition method and device, gene sequencer and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180373921A1 (en) * | 2017-06-27 | 2018-12-27 | The Regents Of The University Of California | Label-free digital brightfield analysis of nucleic acid amplification |
| CN112813152A (en) * | 2021-04-02 | 2021-05-18 | 深圳市博瑞生物科技有限公司 | Digital PCR (polymerase chain reaction) liquid drop fluorescence detection method based on image recognition |
| CN113337583A (en) * | 2021-06-09 | 2021-09-03 | 湖南大学 | Micro-drop digital detection method for multiple target nucleic acids |
| CN113999764A (en) * | 2021-10-20 | 2022-02-01 | 郑州大学 | Integrated micro-droplet digital LAMP nucleic acid detection chip and detection method |
| CN114540469A (en) * | 2022-01-11 | 2022-05-27 | 深圳大学 | Digital nucleic acid quantification method based on non-uniform volume liquid drops and image processing |
| CN115093961A (en) * | 2022-07-29 | 2022-09-23 | 华中农业大学 | Multi-volume liquid drop digital LAMP nucleic acid absolute quantitative detection device and method and application |
| US20220356520A1 (en) * | 2019-05-27 | 2022-11-10 | Centre National De La Recherche Scientifique (Cnrs) | Method of digital multiplex detection and/or quantification of biomolecules and use thereof |
-
2023
- 2023-01-09 CN CN202310028332.XA patent/CN116497095B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180373921A1 (en) * | 2017-06-27 | 2018-12-27 | The Regents Of The University Of California | Label-free digital brightfield analysis of nucleic acid amplification |
| US20220356520A1 (en) * | 2019-05-27 | 2022-11-10 | Centre National De La Recherche Scientifique (Cnrs) | Method of digital multiplex detection and/or quantification of biomolecules and use thereof |
| CN112813152A (en) * | 2021-04-02 | 2021-05-18 | 深圳市博瑞生物科技有限公司 | Digital PCR (polymerase chain reaction) liquid drop fluorescence detection method based on image recognition |
| CN113337583A (en) * | 2021-06-09 | 2021-09-03 | 湖南大学 | Micro-drop digital detection method for multiple target nucleic acids |
| CN113999764A (en) * | 2021-10-20 | 2022-02-01 | 郑州大学 | Integrated micro-droplet digital LAMP nucleic acid detection chip and detection method |
| CN114540469A (en) * | 2022-01-11 | 2022-05-27 | 深圳大学 | Digital nucleic acid quantification method based on non-uniform volume liquid drops and image processing |
| CN115093961A (en) * | 2022-07-29 | 2022-09-23 | 华中农业大学 | Multi-volume liquid drop digital LAMP nucleic acid absolute quantitative detection device and method and application |
Non-Patent Citations (6)
| Title |
|---|
| CUI WU等: "TriD-LAMP: A pump-free microfluidic chip for duplex droplet digital loop-mediated isothermal amplification analysis", 《ANALYTICA CHIMICA ACTA》, vol. 1233, pages 340513 * |
| HECTOR E MUÑOZ等: "Fractal LAMP: Label-Free Analysis of Fractal Precipitate for Digital Loop-Mediated Isothermal Nucleic Acid Amplification", 《ACS SENS》, vol. 28, no. 5, pages 385 - 394 * |
| LINZHE CHEN等: "Deep-dLAMP: Deep Learning-Enabled Polydisperse Emulsion-Based Digital Loop-Mediated Isothermal Amplification", 《ADV SCI(WEINH)》, vol. 9, no. 9, 24 January 2022 (2022-01-24), pages 2105450 * |
| TUSHAR D RANE等: "Microfluidic continuous flow digital loop-mediated isothermal amplification (LAMP)", 《LAB CHIP》, vol. 15, no. 3, pages 776 - 782, XP055850523, DOI: 10.1039/C4LC01158A * |
| 夏赟: "微流控等温扩增核酸定性与定量检测新方法研究", 《中国博士学位论文全文数据库 基础科学辑》, no. 8, pages 006 - 54 * |
| 陈基耘: "基于环介导等温扩增的数字核酸分析技术及其应用研究", 《中国博士学位论文全文数据库 基础科学辑》, no. 6, pages 006 - 28 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117523559A (en) * | 2024-01-08 | 2024-02-06 | 深圳赛陆医疗科技有限公司 | Base recognition method and device, gene sequencer and storage medium |
| CN117523559B (en) * | 2024-01-08 | 2024-03-29 | 深圳赛陆医疗科技有限公司 | Base recognition method and device, gene sequencer and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116497095B (en) | 2024-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210382056A1 (en) | Method for Rapid Accurate Dispensing, Visualization and Analysis of Single Cells | |
| US20230122742A1 (en) | System and method for determining copies-per-unit-volume using pcr and flow control of droplets | |
| US8831887B2 (en) | Absolute PCR quantification | |
| US20120088691A1 (en) | Highly multiplexed real-time pcr using encoded microbeads | |
| US10767218B2 (en) | Method for reducing quantification errors caused by reaction volume deviations in digital polymerase chain reaction | |
| Velez et al. | Massively parallel digital high resolution melt for rapid and absolutely quantitative sequence profiling | |
| JP7545511B2 (en) | Digital PCR measurement method and measurement device | |
| CN116497095B (en) | Dual nucleic acid quantification method, device and medium based on sediment bright field image processing | |
| CN104039978A (en) | Hydrolysis probes | |
| JP2019213513A (en) | Measuring method and measuring apparatus of digital pcr | |
| US20180230515A1 (en) | Method for reducing quantification errors caused by an optical artifact in digital polymerase chain reaction | |
| CN118871594A (en) | Methods, systems, and apparatus for qPCR and array-based analysis of targets in a sample volume | |
| CN113215325B (en) | Reaction system, method and kit for detecting multiple HPV subtypes by two-dimensional PCR single tube closed tube | |
| WO2022259334A1 (en) | Dna detection method and dna detection system | |
| EP4092134A1 (en) | Method and system for dna detection | |
| CN118703602A (en) | Multi-target detection kit based on droplet color gradation coding and application thereof | |
| WO2023067077A1 (en) | High stokes shift fluorescence dyes for multiplex detection | |
| CN108107031A (en) | Method for detecting virus based on DNA microarray technology | |
| CN1314495A (en) | Method for producing polyase chain reaction gene chip |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |