CN117925798A - Multichannel nanopore DNA sequencing detection scheme method and system based on compressed sensing - Google Patents

Abstract

The invention discloses a multichannel nanopore DNA sequencing detection method and system based on compressed sensing, which relate to the technical field of biological microelectronics and solve the problem that a large amount of sensing data generated at high speed in portable nanopore equipment consisting of millions of single nanopores on a chip in the prior art has extremely high design requirements on a traditional data acquisition system following the Nyquist sampling law, and the method comprises the following steps: acquiring a DNA molecular sequence, and converting the DNA molecular sequence into a plurality of sparse signals; multiplying the plurality of sparse signals with the pseudo-random sequence bit by bit to obtain a plurality of modulation sequences; summing the plurality of modulation sequences to obtain an observation vector, and carrying out ADC quantization on the observation vector to obtain a quantized sequence; solving the quantized sequence by using an orthogonal matching pursuit algorithm to obtain a recovery result; the method realizes high-efficiency data compression without losing important information, and greatly reduces the resource consumption of the front end of system data acquisition.

Description

Multichannel nanopore DNA sequencing detection scheme method and system based on compressed sensing

Technical Field

The invention relates to the technical field of biological microelectronics, in particular to a multichannel nanopore DNA sequencing detection method and system based on compressed sensing.

Background

Since the oxford nanopore technology company (ONT) pushed the first nanopore sequencer, mine, in 2014, nanopore sequencing technology and its applications in basic and application research have been significantly developed.

This technique relies on nanoscale protein pores or "nanopores" that serve as biosensors and are embedded in a resistive polymer film. In the electrolytic solution, by applying a constant voltage across the well to generate an ion current through the nanopore, when single-stranded DNA passes through the nanopore, since there are four types of nitrogenous bases constituting the DNA molecule (adenine (a), cytosine (C), guanine (G) and thymine (T)), the ion current flowing through the nanopore can be influenced, and when different bases pass through, the generated characteristic currents are different, the characteristic current signal is detected and read by the integrated circuit, and quantified as a number, so that the determination of the DNA base sequence can be realized. The nanopore-based DNA sequencing technology is mainly attributable to the process of sensing and quantifying weak signals, and has high similarity with the research of relatively mature image sensors. At present, the conventional voltage domain data processing and quantization method of an image sensor is generally adopted in the quantization process of the DNA sequencing technology.

In the existing multichannel data sampling quantization scheme, the time division multiplexing and frequency division multiplexing method reduces the chip area occupied by a quantization circuit to a certain extent through multiplexing a single quantization channel, but because the traditional quantization circuit works based on the shannon-nyquist sampling theorem, the multiplexing is realized by linearly superposing the works of a plurality of channel ADCs on one ADC, the total amount of data to be processed is not changed, so that design indexes in terms of ADC sampling rate and power consumption become more severe along with the increase of the number of channels, and the requirement on the circuit is not reduced in practice. The method of applying compressed sensing technology on each signal channel separately reduces the total data amount, but compared with the traditional nyquist sampling scheme, the more complex circuit architecture occupies larger chip area, and meanwhile, the problem of mismatch between channels exists.

Disclosure of Invention

The invention solves the problem that a large amount of sensing data generated at high speed in portable nanopore equipment consisting of millions of single nanopores on a chip in the prior art has extremely high design requirements on a traditional data acquisition system conforming to the Nyquist sampling law, realizes high-efficiency data compression without losing important information, and greatly reduces the resource consumption of the data acquisition front end of the system.

In a first aspect, the present invention provides a method for detecting multi-channel nanopore DNA sequencing based on compressed sensing, the method comprising:

acquiring a DNA molecular sequence, and converting the DNA molecular sequence into a plurality of sparse signals;

Multiplying a plurality of sparse signals with the pseudo-random sequence bit by bit to obtain a plurality of modulation sequences;

Summing the plurality of modulation sequences to obtain an observation vector, and carrying out ADC quantization on the observation vector to obtain a quantized sequence;

And solving the quantized sequence by using an orthogonal matching pursuit algorithm to obtain a recovery result.

With reference to the first aspect, in a possible implementation manner, the obtaining a DNA molecule sequence, converting the DNA molecule sequence into a sparse signal includes:

calculating sampling voltage values corresponding to different bases in the DNA molecular sequence by utilizing the nanopore and the detection unit;

determining a sampling voltage value matrix corresponding to the sampling voltage value by using a reference sequence library;

and determining the sparse signal corresponding to the DNA molecular sequence according to the sampling voltage value matrix.

With reference to the first aspect, in one possible implementation manner, the sparse signal includes:

wherein, Representation/>Coefficient vectors under the psi domain; n represents the length of the DNA molecule sequence.

With reference to the first aspect, in one possible implementation manner, the modulation sequence includes:

wherein, Representing sparse signals; /(I)Representing a pseudo-random sequence; n represents the length of the DNA molecule sequence.

In a second aspect, the present invention provides a compressed sensing-based multichannel nanopore DNA sequencing detection system, comprising: m multiplication circuits, a sampling circuit, and a signal recovery circuit; the M multiplication circuits are connected with the input end of the sampling circuit in parallel; the output end of the sampling circuit is connected with the input end of the signal recovery circuit in series;

The M multiplication circuits are respectively used for multiplying the M sparse signals with the pseudo random sequences to obtain M signals to be sampled;

the sampling circuit is used for sampling the M signals to be sampled by utilizing a multi-channel sampling method to obtain multi-channel sampling quantity signals;

The signal recovery circuit is used for separating and recovering the digital codes to obtain M sparse signals.

With reference to the second aspect, in one possible implementation manner, the M multiplication circuits are respectively configured to multiply M sparse signals with a pseudo random sequence to obtain M signals to be sampled, where the multiplying includes:

When the rising edge of the sampling clock signal phi comes, if pseudo random sequence Reverse sequence/>Control signal phipi=1, control signal phini=0, if pseudo random sequence/>Reverse sequence/>The control signal phipi=0 and the control signal phini=1;

And obtaining M signals to be sampled by using the control signals and the multiplication circuit.

With reference to the second aspect, in one possible implementation manner, the M multiplication circuits respectively include: a first set of switching circuits and a second set of switching circuits;

the sampling circuit comprises M groups of binary capacitors;

the M multiplication circuits are respectively connected with the M groups of binary capacitors;

The first group of circuits of each multiplication circuit is correspondingly connected with the upper polar plate of the binary capacitor;

the second group of circuits of each multiplication circuit is correspondingly connected with the lower polar plate of the binary capacitor.

With reference to the second aspect, in one possible implementation manner, the first set of switching circuits includes: a switch K1 and a switch K2;

the second set of switching circuits includes: a switch K3 and a switch K4;

The switch K1 is connected with the switch K2 in parallel;

The switch K3 and the switch K4 are connected in parallel.

With reference to the second aspect, in one possible implementation manner, the sampling circuit is configured to sample the M signals to be sampled by using a multi-channel sampling method, to obtain a multi-channel sampling amount signal, where the method includes:

Dividing the binary capacitors in the sampling circuit into M parts equally to obtain M groups of equally divided binary capacitors;

And respectively utilizing the M groups of binary capacitors to sample the M signals to be sampled on a lower polar plate of the binary capacitors, and obtaining multichannel sampling quantity signals on an upper polar plate of the binary capacitors.

With reference to the second aspect, in one possible implementation manner, the method further includes: a comparator; the input end of the comparator is connected with the upper polar plate of the sampling circuit.

One or more technical schemes provided by the invention have at least the following technical effects or advantages:

(1) Compared with the method of completely acquiring and filtering redundant information in the traditional data compression method, the method disclosed by the invention has the advantages that the compressed sensing technology is utilized to convert the reference sequence library in the nanopore sequencing technology into the sparse representation dictionary in the compressed sensing, and the resource consumption of a front-end data acquisition circuit is reduced by the method of on-chip compressed sampling and off-chip reconstruction signals in the proposed system, so that complex calculation is transferred to a back-end platform with stronger calculation power, and the working efficiency of the sequencing system is improved.

(2) Compared with the prior multi-channel signal processing technology, when the data generated by the nanopore sequencing units of m channels are received, the sampling quantization circuit acquires the nanopore sequencing data at a lower sampling rate under the condition of not losing important information, and the data volume to be processed is only the total data volumeThe required hardware resources and power consumption in the multichannel data sampling and quantizing process are greatly reduced.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments of the present invention or the description of the prior art 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 steps of a method for detecting multi-channel nanopore DNA sequencing based on compressed sensing according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a variable relationship in an observation vector according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a multi-channel nanopore DNA sequencing detection system based on compressed sensing according to an embodiment of the present invention;

FIG. 4A is a schematic diagram showing the analysis of the reconstruction result when the base sequences of the 4 channels are matched with the library according to the embodiment of the invention;

FIG. 4B is a schematic diagram of analysis of the reconstruction result when the base sequences of the 4 channels are not matched with the library according to the embodiment of the invention;

FIG. 5A is a waveform of a characteristic current generated by passing a base sequence of length 16 through a nanopore at a speed of 100 us/min in accordance with an embodiment of the present invention;

FIG. 5B is a graph of RST signal with reset time at the first 2us and integration time at the last 98us of each cycle provided by an embodiment of the present invention;

FIG. 5C is a waveform diagram of the output voltage of the detection unit Cell under RST control provided by an embodiment of the present invention;

FIG. 5D is a graph of discrete sample values obtained by sampling a voltage waveform at 98us per cycle according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a multiplication circuit according to an embodiment of the present invention;

Fig. 7 is a circuit diagram of a multi-channel nanopore DNA sequencing detection system based on compressed sensing according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments 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.

The invention provides a multichannel nanopore DNA sequencing detection method based on compressed sensing, which is shown in fig. 1 and comprises the following steps S101 to S104.

S101, acquiring a DNA molecular sequence, and converting the DNA molecular sequence into a plurality of sparse signals. Here the number of the elements is the number,

Specifically, in step S101, a DNA molecule sequence is acquired, and the DNA molecule sequence is converted into a sparse signal, including the following steps S1011 to S1013.

S1011, calculating sampling voltage values corresponding to different bases in the DNA molecular sequence by utilizing the nanopore and the detection unit.

Illustratively, the DNA molecules are specific, i.e., each DNA molecule has its own specific arrangement of base pairs, which contains specific genetic information, so that when the nanopore sequencing result is aligned with a library of reference sequences to detect the presence of a particular gene, only the base arrangement of interest or significance need be included in the library of reference sequences, and not all possible base arrangements. This inherent structure observed in natural data means that the data can be sparsely represented in an appropriate coordinate system. This means that a corresponding sparse representation dictionary can be directly formed based on the reference sequence library, and further the working efficiency of the data acquisition system can be improved by introducing a compressed sensing technology.

S1012, determining a sampling voltage value matrix corresponding to the sampling voltage values by using the reference sequence library. Here, a procedure for constructing a corresponding sparse representation dictionary based on a library of reference sequences consisting of N known base sequences of length N is described as follows: under the condition of determining the working principle of the used nanopore and the detection unit Cell, the sampling voltage values corresponding to different bases can be directly obtained through calculation. Then, according to the corresponding relation between the base and the sampling voltage value, generating an N sampling voltage value matrix corresponding to the N reference sequence library based on the N reference sequence library (each column corresponds to one base sequence), namely a corresponding sparse representation dictionary, which is recorded as

S1013, determining sparse signals corresponding to the DNA molecular sequences according to the sampling voltage value matrix. Specifically, the sparse signal includes:

wherein, Representation/>Coefficient vectors under the psi domain; n represents the length of the DNA molecule sequence.

For M base sequences of length N, M discrete sample sequences of length N can be obtained by the above-mentioned processingTo/>In the sparse representation dictionary psi, sparse signals of the ith (i is more than or equal to 1 and less than or equal to M) channel signals can be obtained, and the specific sparse signal representation is shown in a formula (1.1).

S102, multiplying the plurality of sparse signals with the pseudo-random sequence bit by bit to obtain a plurality of modulation sequences. Specifically, the modulation sequence includes:

wherein, Representing sparse signals; /(I)Representing a pseudo-random sequence; n represents the length of the DNA molecule sequence.

Exemplary, ifWhen the corresponding base sequence is in the library of reference sequences, then/>Only one non-zero coefficient, i.e. signal sparsity, is 1. Next,/>+ -1 Pseudo-random sequence/>, corresponding to a channelObtaining modulation sequence/>, after multiplying bitwiseSpecifically expressed as formula (1.2).

S103, summing the plurality of modulation sequences to obtain an observation vector, and carrying out ADC quantization on the observation vector to obtain a quantized sequence.

In the circuit implementation based on SAR ADC proposed towards actual differential system, the process of multiplying differential signal with + -1 pseudo-random sequence is equivalent to the conversion of signal polarity, the multiplication function module is realized by four switches and corresponding control logic, compared with the circuit implementation of multiplication operation by adopting analog multiplier, the circuit structure adopted by the invention is simpler, and the whole process only has dynamic power consumption.

Specifically, the modulation sequences of all channels are summed to obtain an observation vectorVector/>Expressed as:

wherein,

The process is carried out by the steps of, The simplification is as follows:

Fig. 2 shows the relationship between the variables. As can be seen, Θ is a known N x (mxn) matrix, Is an unknown vector of (M×N) 1 (M×N) ×1)/>Is an N x1 vector. /(I)And (3) carrying out ADC quantization to obtain:

wherein, Representing noise introduced by the ADC. This constitutes a system of underdetermined equations under the system architecture of the present invention, which can be solved by a classical linear or convex optimization method, or a greedy method, for which the solution process is directly performed by an orthogonal matching pursuit algorithm (OMP) and the number of iterations is set to M.

Once it isIs accurately solved, and the solved result is recorded as/>The original input signal of the multi-channel can be recovered and the recovered signal is denoted/>, in fig. 2

Wherein the method comprises the steps ofFor/>A vector of length N consisting of [ (i-1) n+1] th element to i N th element.

S104, solving the quantized sequence by using an orthogonal matching pursuit algorithm to obtain a recovery result.

Specifically, the result is reconstructed based on the characteristics of an orthogonal matching pursuit algorithm (OMP)In fact, the sequences/>, to be tested, in the sparse representation dictionary ψThe sequence with the highest matching degree needs to know whether the sequences are completely consistent or not, and the following analysis needs to be carried out:

now respectively define And/>Analysis for reconstruction results

The following describes an analysis method of the signal reconstruction result of the system by taking a four-channel case as an example: of the four channels, the base sequence length of each channel is 1024, respectively recorded asThere is a 1024 x 1024 sparse representation dictionary ψ and four pseudo random binary sequences/>, respectively Can be obtained through the first half of the processing flow of the system shown in FIG. 3

Wherein the method comprises the steps ofNoise introduced by the ADC is modeled here as gaussian random noise with an amplitude within ±2 mV.

Solving with the iteration number of 4 by using OMP algorithm to obtain a four-channel final reconstruction result of the system as followsThereby obtaining

When the base sequences of the 4 channels are all perfectly matched to the library,Noise that will appear to be uniform in amplitude and much smaller than the signal amplitude in the time domain, as shown in FIG. 4A, can be seen/>, if the noise margin is defined to be + -4 mV (as shown by the red dashed line in FIG. 4A)Is within noise margin. This shows that the base sequences of the 4 channels are matched with the library, and the corresponding sequences can be further obtained from the library according to the reconstruction result.

When the base sequences of the 4 channels are not matched with the library,Noise that will appear significant in the time domain and comparable to the signal amplitude, as shown in FIG. 4B, if the noise margin is defined to be + -4 mV (as shown by the red dashed line in FIG. 4B), it can be seen/>A portion of the amplitude of (c) is far beyond the noise margin. This means that at least one of the 4-channel base sequences is not matched with the library, i.e., the point of mismatch corresponds/>The number of unmatched points represents the similarity between the sequence to be tested and the sequences in the library, which provides a direction for subsequent analysis and research.

The invention provides a multichannel nanopore DNA sequencing detection system based on compressed sensing, as shown in FIG. 3, which comprises: m multiplication circuits, a sampling circuit, and a signal recovery circuit; m multiplication circuits are connected in parallel with the input ends of the sampling circuits; the output end of the sampling circuit is connected in series with the input end of the signal recovery circuit.

Multichannel nanopore DNA sequencing detection system based on compressed sensing, still includes: a comparator; the input end of the comparator is connected with the upper polar plate of the sampling circuit.

Exemplary, a system as shown in fig. 3, wherein I ₁ to I _M are respectively the characteristic currents generated by the M nanopores. The characteristic currents are sensed and detected by a weak current detection unit (Cell), the Cell is an integrating and amplifying module, the function of the Cell is to integrate the characteristic currents on an integrating capacitor C _int to obtain output voltage signals V ₁ to V _M, one working period of the Cell is the same as the time of a single base passing through a nanopore, and the output voltages V ₁ to V _M are reset by a RST signal before the integration of each period starts. Then, the M channels of integrated voltages V ₁ -V _M are sampled at the end of the period at the frequency of f _s (same as the operating frequency of the Cell), resulting in a discrete M sampling sequencesTo/>

Taking one channel as an example, the characteristic current amplitude generated by passing different bases through the nanopore is in the pA level, and here, it is assumed that the characteristic currents corresponding to A, T, C, G bases are 120pA, 60pA, 90pA and 100pA respectively, and the characteristic current waveform generated by passing a base sequence with a length of 16 through the nanopore at a speed of 100 us/min is shown in FIG. 5A. Then, it is assumed that the detection unit resets the output to 0 when the control signal rst=1, integrates at rst=0, and specifies the first 2us of each cycle as the reset time and the last 98us as the integration time, such RST signal being shown in fig. 5B. Under RST control, the output voltage waveform of the detection unit Cell is as shown in fig. 5C. The discrete sample values obtained after sampling the voltage waveform at 98us per cycle are shown in fig. 5D. Because different bases correspond to different characteristic currents, voltage slopes generated by integration on Cint are different, and further amplitudes obtained by sampling are also different, in conventional nanopore sequencing detection, the corresponding base sequence is often obtained by taking the voltage slopes as the basis of analysis.

The M multiplication circuits are respectively used for multiplying the M sparse signals with the pseudo-random sequence to obtain M signals to be sampled; specifically, the M multiplication circuits are respectively configured to multiply the M sparse signals with the pseudo random sequence to obtain M signals to be sampled, and the method includes: when the rising edge of the sampling clock signal phi comes, if pseudo random sequenceReverse sequence/>Control signal phipi=1, control signal phini=0, if pseudo random sequence/>Reverse sequence/>The control signal phipi=0 and the control signal phini=1; and obtaining M signals to be sampled by using the control signals and the multiplication circuit.

The M multiplication circuits respectively include: as shown in fig. 6, a first set of switching circuits and a second set of switching circuits; the sampling circuit comprises M groups of binary capacitors; the M multiplication circuits are respectively connected with M groups of binary capacitors; the first group of circuits of each multiplication circuit is correspondingly connected with the upper polar plate of the binary capacitor; the second group of circuits of each multiplication circuit is correspondingly connected with the lower polar plate of the binary capacitor.

Specifically, the first set of switching circuits includes: a switch K1 and a switch K2; a second set of switching circuits comprising: a switch K3 and a switch K4; the switch K1 is connected in parallel with the switch K2; switch K3 and switch K4 are connected in parallel.

The sampling circuit is used for sampling M signals to be sampled by utilizing a multichannel sampling method to obtain multichannel sampling quantity signals, and specifically comprises the following steps:

(1) And equally dividing the binary capacitors in the sampling circuit into M parts to obtain M groups of equally divided binary capacitors.

(2) And respectively utilizing M groups of binary capacitors to sample M signals to be sampled on a lower polar plate of the binary capacitors, and obtaining multichannel sampling quantity signals on an upper polar plate of the binary capacitors.

The signal recovery circuit is used for separating and recovering the digital codes to obtain M sparse signals.

In a specific embodiment provided by the present invention, m=4. In fig. 7, REF, VCM, GND are the positive reference voltage, the common mode voltage, and the negative reference voltage during the ADC quantization. V1 to V4 are output signals of 4 detection units,To/>Is a + -1 pseudo-random sequence corresponding to four channels, and phi is the sampling clock of the ADC.

Unlike the order of the signal processing function blocks in the system, the process of multiplying the 4-channel input signal with the pseudo-random binary sequence (+ -1 PRBS) is performed first, followed by the process of sampling and summing.

For fully differential signals, multiplying by + -1 has the same effect as exchanging the polarity of the differential signal, which can be accomplished by a multiplication function consisting of four switches as shown in fig. 7. In FIG. 7, pseudo-random sequences with + -1And its inverted sequence/>The control signals phi pi and phi ni of the sampling switches are generated together through an AND gate. When the rising edge of the sampling clock comes, if/>Then there are phipi=1 and phini=0, the original differential signal passes through a multiplication module composed of four switches to obtain the signal/>, to be sampled by the ADCMultiplication of the signal with "+1" is achieved; if it isThen there are phipi=0 and phini=1, the original differential signal passes through a multiplication module composed of four switches to obtain the signal/>, to be sampled by the ADCMultiplication of the signal with "-1" is achieved.

In the invention, the multiplication process of the differential signal and the + -1 pseudo-random sequence is realized by only four switches and corresponding control logic, and compared with the multiplication operation realized by adopting an analog multiplier, the circuit structure adopted by the invention is simpler, and the whole process only has dynamic power consumption.

The binary DAC array is equally divided into 4 parts, where the MSB capacitor 2 ⁹ C is split into two equal 2 ⁸ C, receiving the inputs of channel 1 and channel 2 respectively, the next highest capacitor 2 ⁸ C is unchanged, receiving the input of channel 3, all the remaining capacitors together form a capacitor of 2 ⁸ C, receiving the input of channel 4. In sampling, the input signals of 4 channels are sampled simultaneously with the same weight through the form of sampling of the lower polar plate. When sampling is completed, the signal coming to the polar plate on the capacitor array, i.e. the input end of the comparator, is

The summation process is automatically implemented in the sampling process without additional arithmetic circuitry. Unlike the original flow, there is one in the observation vectorThis is easily compensated during signal reconstruction-the reconstruction result is amplified by a factor of four.

Finally, a single ADC converts the average result to a digital sequence from which the reconstruction algorithm recovers and separates the input signals for all channels. The SAR ADC can simultaneously sample and quantize the multichannel sparse signal at the Nyquist rate of one channel. Compared with the prior art, the architecture not only saves power consumption and hardware cost, but also avoids the problems of time sequence deflection, offset mismatch, gain mismatch and the like among channels.

The invention uses a single DAC array to compress and sample multi-channel signals at the same time, which is realized by equally dividing the DAC array according to the number of signal channels and adopting a method of sampling by a lower polar plate.

Compared with a method for realizing compressed sensing for each channel independently, the method expands a compressed sensing algorithm to multi-channel application and embeds the multi-channel application into a single SAR ADC, the SAR ADC simultaneously compresses, samples and quantifies multi-channel sparse signals at the Nyquist rate of one channel, an internal multiplication module is realized by adopting a passive circuit with only dynamic power consumption, and the averaging operation is automatically realized by utilizing the characteristics of a DAC capacitor array without an additional circuit structure, so that the circuit complexity and the chip area occupied by a quantification circuit are greatly reduced, and meanwhile, the problems of time sequence deflection, offset mismatch, gain mismatch and the like among channels under the condition of multipath quantification are avoided as the multi-channel sampling quantification is realized on the single ADC.

In the multichannel signal sampling and quantizing scheme provided by the invention, the compressed sensing theory is firstly expanded to multichannel application, only a single ADC is used in circuit realization, and the multichannel input signals are simultaneously compressed and sampled at the sampling and quantizing rates of a single channel, so that the power consumption and the area of a quantizing circuit are greatly saved.

In the circuit implementation based on SAR ADC proposed towards actual differential system, the multi-channel signal is compressed and sampled simultaneously by a single DAC array, which is realized by dividing the DAC array equally according to the number of signal channels and adopting a method of sampling by a lower polar plate.

To illustrate the technical solution of the invention, but not to limit the invention; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced with equivalents; such modifications and substitutions do not depart from the spirit of the invention.

Claims (10)

1. A multichannel nanopore DNA sequencing detection method based on compressed sensing is characterized by comprising the following steps:

acquiring a DNA molecular sequence, and converting the DNA molecular sequence into a plurality of sparse signals;

Multiplying a plurality of sparse signals with the pseudo-random sequence bit by bit to obtain a plurality of modulation sequences;

Summing the plurality of modulation sequences to obtain an observation vector, and carrying out ADC quantization on the observation vector to obtain a quantized sequence;

And solving the quantized sequence by using an orthogonal matching pursuit algorithm to obtain a recovery result.

2. The compressed sensing-based multichannel sampling quantization circuit of claim 1, wherein the acquiring a sequence of DNA molecules, converting the sequence of DNA molecules into sparse signals, comprises:

calculating sampling voltage values corresponding to different bases in the DNA molecular sequence by utilizing the nanopore and the detection unit;

determining a sampling voltage value matrix corresponding to the sampling voltage value by using a reference sequence library;

and determining the sparse signal corresponding to the DNA molecular sequence according to the sampling voltage value matrix.

3. The compressed sensing-based multichannel sample quantization circuit of claim 1, wherein the sparse signal comprises:

wherein, Representation/>Coefficient vectors under the psi domain; n represents the length of the DNA molecule sequence.

4. The compressed sensing-based multichannel sample quantization circuit of claim 1, wherein the modulation sequence comprises:

wherein, Representing sparse signals; /(I)Representing a pseudo-random sequence; n represents the length of the DNA molecule sequence.

5. A compressive sensing-based multichannel nanopore DNA sequencing detection system, comprising: m multiplication circuits, a sampling circuit, and a signal recovery circuit; the M multiplication circuits are connected with the input end of the sampling circuit in parallel; the output end of the sampling circuit is connected with the input end of the signal recovery circuit in series;

The M multiplication circuits are respectively used for multiplying the M sparse signals with the pseudo random sequences to obtain M signals to be sampled;

the sampling circuit is used for sampling the M signals to be sampled by utilizing a multi-channel sampling method to obtain multi-channel sampling quantity signals;

The signal recovery circuit is used for separating and recovering the digital codes to obtain M sparse signals.

6. The compressed sensing-based multichannel sampling quantization circuit of claim 5, wherein the M multiplication circuits are configured to multiply M sparse signals with a pseudo random sequence, respectively, to obtain M signals to be sampled, and include:

When the rising edge of the sampling clock signal phi comes, if pseudo random sequence Reverse sequence-/>Control signal phipi=1, control signal phini=0, if pseudo random sequence/>Reverse sequence-/>The control signal phipi=0 and the control signal phini=1;

And obtaining M signals to be sampled by using the control signals and the multiplication circuit.

7. The compressed sensing-based multichannel sample quantization circuit of claim 5, wherein the M multiplication circuits each comprise: a first set of switching circuits and a second set of switching circuits;

the sampling circuit comprises M groups of binary capacitors;

the M multiplication circuits are respectively connected with the M groups of binary capacitors;

The first group of circuits of each multiplication circuit is correspondingly connected with the upper polar plate of the binary capacitor;

the second group of circuits of each multiplication circuit is correspondingly connected with the lower polar plate of the binary capacitor.

8. The compressed sensing based multichannel sample quantization circuit of claim 5, wherein the first set of switching circuits comprises: a switch K1 and a switch K2;

the second set of switching circuits includes: a switch K3 and a switch K4;

The switch K1 is connected with the switch K2 in parallel;

The switch K3 and the switch K4 are connected in parallel.

9. The compressed sensing-based multichannel sampling quantization circuit of claim 5, wherein the sampling circuit is configured to sample the M signals to be sampled by using a multichannel sampling method to obtain a multichannel sample size signal, and comprises:

Dividing the binary capacitors in the sampling circuit into M parts equally to obtain M groups of equally divided binary capacitors;

And respectively utilizing the M groups of binary capacitors to sample the M signals to be sampled on a lower polar plate of the binary capacitors, and obtaining multichannel sampling quantity signals on an upper polar plate of the binary capacitors.

10. The compressed sensing-based multichannel sampling quantization circuit of claim 5, wherein the sampling circuit further comprises: a comparator; the input end of the comparator is connected with the upper polar plate of the sampling circuit.