JP2015210233A - Massively parallel biomolecule detection method and apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a massively parallel biomolecule detector capable of measuring an amount of biomolecules in specimen solution at high speed and in simultaneously parallel.SOLUTION: According to a massively parallel biomolecule detector 10, when specimen solution is introduced into a plurality of sensor cells CE, a current pulse signal CP which changes pulse width corresponding to electric potential generated by a reaction between biomolecules in the specimen solution and probe molecule is created in a detection circuit (current pulse signal creating part), and an amount of biomolecules is detected based on the current pulse signal CP. Thus, even if the current pulse signal CP having a short signal width required for high-speed measurement is used, the apparatus is not affected by the number of the sensor cells CE and stray capacitance which increases by making a semiconductor integrated circuit substrate 12 larger, and highly accurate detection accuracy is obtained.

Description

  The present invention relates to a massively parallel biomolecule detection method and apparatus for detecting the amount of biomolecules in a specimen in a massively parallel manner for diagnosing diseases, identifying constitutions, identifying individuals, and the like. The present invention relates to a technique that enables a sensor to simultaneously detect many types of biomolecules contained in a specimen.

  What biomolecules in the body and ingestions, such as disease diagnosis, fraudulent food and food poisoning testing, borderline prevention of infections, tracking the effects of sports and medications, providing medicines and cosmetics suitable for the constitution, DNA testing, etc. In order to examine how much is contained, react various types of biomolecules such as sugars, viruses, DNA, and proteins contained in collected samples such as blood, sweat, urine, saliva, and food with the probe. Thus, the amount of a specific biomolecule contained in the specimen is detected.

  For this reason, a probe that reacts only with a specific molecule, such as an enzyme, a single-stranded DNA having a specific sequence, a peptide that reacts with a specific protein, or an antibody is used. When a substance to be detected, i.e., a biomolecule, is clear as in the diagnosis of a specific disease, for example, measurement of blood glucose level in diabetes, a specific probe that reacts only with the biomolecule may be used. However, when it is unclear what kind of biomolecule is contained in the specimen as in preventive diagnosis, it is necessary to use all possible probes. For this reason, a detection apparatus having an enormous number of types of probes is required. For example, DNA bases are composed of 4 types of A, T, C, and G, and 10 bases form 1 million types of sequences. In addition, there are 20 types of amino acids that are constituent molecules of proteins, and 5 amino acids form 3.2 million types of sequences. For this reason, since the kinds of DNA and protein detection are enormous, an apparatus for detecting substances in parallel is desired. Furthermore, if the role of genetic information becomes clearer, a DNA sequencer that reads all the base sequences of DNA, which is 3 billion types of bases in humans, will be an effective tool for preventive medicine.

Conventionally, in biosensor arrays, a method has been used in which a sample is modified with a phosphor and the presence or absence of binding to the probe is detected optically. However, there was a drawback that the size of the device increased. On the other hand, as shown in Non-Patent Document 1, a biosensor using a semiconductor integrated circuit has been proposed. According to this biosensor, there is an advantage that it is not necessary to modify the fluorescent substance on the specimen, and it is easy to handle, small in size, inexpensive, super parallel detection, and information processing is possible. In Non-Patent Document 1, a DNA sequencer in which 10 million sensors are integrated within an area of 1 cm ² has been commercialized.

  Further, as shown in Non-Patent Document 2, an electrochemical measurement type biosensor including a molecule recognition unit and a signal conversion unit has been proposed. This molecular recognition unit utilizes the specificity of a molecular reaction with a predetermined probe when a plurality of types of probes are brought into contact with a specimen, respectively. The signal conversion unit converts the result representing the specificity of the molecular reaction of the molecule recognition unit into a change in an electrical signal such as potential, current, impedance, etc., and performs signal processing.

Japanese Patent No. 4883812 JP 2012-047536 A

Rothenberg et al, “An integrated semiconductor device enabling non-optical genome sequencing” Nature 475, pp.348-352, July 2011 K. Nakazato, “Potentiometric, Amperometric, and Impedimetric CMOS Biosensor Array”, in State of the Art in Biosensors / Book 1, ISBN 980-953-307-695-5, ed. By T. Rinken, In Tech, pp. 163-178, (2012) Y. Liao, H. Yao, B. Parviz, and B. Otis, “A 3-uW Wireless Powered CMOS Glucose Sensor for Active Contact Lens” in Proc. IEEE International Solid-State Circuits Conference (14/18) (ISSCC 2011 ), Feb. 2011, pp. 38-39 Y. Liao, H. Yao, A. Lingley, B. Otis, “A 3-uW CMOS Glucose Sensor for Wireless Contact-Lens Tear Glucose Monitoring,” IEEE J. Solid-State Circuits 47 (1) pp. 335-333 , 2012 Y. Liu, A. Al-Ahdal, P. Georgiou, “Minimal readout scheme for ISFET sensing arrays based on pulse width modulation,” Electronic Letters, vol. 48, no. 10, pp. 548-549, 2012 K. Wang, Y. Liu, C. Toumazou, P. Georgiou, “A TDC Based ISFET Readout for Large-Scale Chemical Sensing Systems”, Biomedical Circuit and Systems Conference (BioCAS), 2012 IEEE, pp. 176-179 N. T. Trung and P. Hafliger, “Time domain ADC for blood glucose implant,” Electronics Letters, vol. 47, no. 26, pp. S18-S20, 2011 N. T. Trung and P. Hafliger, “Inverter Based Readout Circuit for Implanted Glucose Sensor”, Biomedical Circuit and Systems Conference (BioCAS), 2012 IEEE, pp. 252-255 B. Kim, and K. Nakazato, “Dual Data Pulse Width Modulator for Radio Frequency Identification Biosensor Signal Modulation”, Jpn. J. Appl. Phys. 2013, 52, 04CE12

  As shown in Patent Literature 1 and Patent Literature 2, a conventional electrochemical measurement biosensor employs a method of outputting a detection signal as a voltage or current amplitude. However, since this method handles voltage or current analog signals, it has a drawback that noise resistance cannot be sufficiently obtained, and a small detection signal is buried and detection accuracy cannot be obtained.

  On the other hand, in general, it is possible to increase noise resistance by handling time domain signals rather than treating voltage and current amplitude values as analog quantities. Further, the slew rate (response speed) has been improved by the advancement of CMOS integrated circuit technology. In Non-Patent Documents 3 to 9, a biosensor composed of an integrated circuit having a large number of sensor cells with low power consumption and high accuracy is obtained by detecting the amount of biomolecules using time domain signals from these points. Proposed.

  However, as the number of sensor cells increases, the stray capacitance added to the signal transmission line increases and the amount of charge increases. Therefore, if voltage pulses are used, the lower limit of the pulse width (that is, the upper limit of the signal frequency) is used. Since there are significant restrictions, for example, limited to voltage pulses of 100 ns or more, the actual situation consists of an integrated circuit having a large number of sensor cells that simultaneously measure the biomolecular weight at high speed using voltage pulse signals in the time domain. It was difficult to realize a biosensor.

  The present invention has been made against the background of the above circumstances, and its object is to provide a massively parallel biomolecule detection method and apparatus capable of simultaneously measuring the amount of biomolecules in a sample solution at high speed. Is to provide.

  In the biosensor having a large number of sensor cells as the background of the above circumstances, the present inventors have conducted a high-speed measurement when dealing with a signal representing a biomolecular weight in the time width of a current pulse. Even if a pulse signal with a short signal width required for this purpose is used, high detection accuracy can be obtained without being affected by the increase in the number of sensors and the stray capacitance that is increased by increasing the number of integrated circuits. I found it. The present invention has been made based on such knowledge.

  That is, the gist of the first invention for achieving the above object is to provide a plurality of sensor cells each containing one of a plurality of types of probe molecules and into which a sample solution is introduced. A method for detecting a biomolecule in parallel, wherein the amount of the biomolecule is measured in parallel, and the pulse width is changed in accordance with a potential change generated by a reaction between the biomolecule in the sample solution and the probe molecule A current pulse signal is generated, and the amount of the biomolecule is detected based on the current pulse signal.

  Further, the gist of the second invention, which is an apparatus that can suitably carry out the method of the first invention, is that a plurality of probe molecules are accommodated and a plurality of specimen solutions are introduced. And a super-parallel biomolecule detection apparatus that simultaneously measures the amount of biomolecules in the sample solution, which is generated by a reaction between the biomolecules in the sample solution and the probe molecules. A current pulse signal generation unit that generates a current pulse signal that changes a pulse width according to a potential change, and a biomolecule amount calculation that calculates the amount of the biomolecule based on a pulse width of the current pulse signal from a predetermined relationship Part.

  According to the first and second inventions, when a sample solution is introduced into a plurality of sensor cells, the pulse width is changed according to a potential change generated by a reaction between a biomolecule in the sample solution and the probe molecule. A current pulse signal is generated, and the amount of the biomolecule is detected based on the current pulse signal. As a result, even if a pulse signal with a short signal width required for high-speed measurement is used, it is not affected by the stray capacitance that increases due to the increase in the number of sensors and the number of integrated circuits. Is obtained.

  Preferably, the plurality of sensor cells are arranged in a matrix at intersections of the plurality of word lines and the plurality of bit lines intersecting the plurality of word lines in the semiconductor integrated circuit substrate. In addition, a sensor matrix (sensor array circuit) is configured. In this way, since a large number of minute sensor cells can be arranged on the semiconductor integrated circuit substrate, the amount of many kinds of biomolecules in the sample solution can be detected simultaneously and efficiently.

  Preferably, the plurality of sensor cells include a plurality of detection electrodes fixed to the semiconductor integrated circuit substrate, exposed in the plurality of sensor cells, and in contact with the sample solution in a state of being covered with a ferrocene derivative. And probe carrier particles suspended in the sample solution in a state where the probe molecules are fixed on the surface. In this way, the biomolecules in the sample solution introduced into the sensor cell are electrically detected, and the biomolecules can be easily detected.

  Preferably, the plurality of sensor cells include a plurality of detection electrodes fixed to the semiconductor integrated circuit substrate and provided in the plurality of sensor cells, and ions that are provided in the vicinity of the detection electrodes and contact the sample solution. A sensitive membrane and probe carrier particles floating in the sample solution with the probe molecules fixed on the surface are provided. In this way, the biomolecules in the sample solution introduced into the sensor cell are electrically detected, and the biomolecules can be easily detected.

  Preferably, the current pulse signal generation unit is provided between the plurality of detection electrodes and the plurality of word lines and bit lines, and detects the potential change detected by the plurality of detection electrodes. It consists of a plurality of detection circuits that output current pulse signals with corresponding pulse widths. In this way, a current pulse signal having a pulse width corresponding to the potential change detected by the plurality of detection electrodes respectively disposed in the plurality of sensor cells is output, so that it is a short necessary for high-speed measurement. Even when a pulse signal having a signal width is used, high detection accuracy can be obtained without being affected by the number of sensors and the stray capacitance that is increased by increasing the scale of the integrated circuit.

  Preferably, a current / voltage pulse conversion circuit for converting the current pulse signal transmitted through the bit line sequentially selected from the plurality of bit lines into a voltage pulse signal, and a bit string representing a time width of the voltage pulse signal A time-to-digital signal conversion circuit for converting to a digital signal comprising: the biomolecule amount calculation unit, based on an actual potential change represented by the digital signal from a previously stored relationship, a probe molecule in the cell Therefore, there is an advantage that many kinds of biomolecule amounts are detected in parallel.

  Preferably, a short pulse signal generator for outputting a short voltage pulse for outputting a current pulse signal from the current pulse signal output circuit is connected to each of the plurality of word lines. Since the current pulse signal is output from the current pulse signal output circuit in response to the short voltage pulse output from the short pulse signal generator, there is an advantage that many kinds of biomolecular weights can be detected simultaneously.

  Preferably, the current pulse signal output circuit connects a detection transistor having a gate connected to the detection electrode, and a drain of the detection transistor to a first power supply according to a short pulse signal of the word line. Connected between a selection transistor, a switch transistor having a drain connected to the bit line and switching according to a drain voltage of the detection transistor, and a source of the switch transistor and a second power source lower than the first power source A diode. In this way, a current pulse signal having a pulse width corresponding to the potential change detected by the plurality of detection electrodes respectively disposed in the plurality of sensor cells is output.

BRIEF DESCRIPTION OF THE DRAWINGS It is a principal part circuit diagram explaining the structure of the massively parallel biomolecule detection apparatus to which an example of this invention is applied, Comprising: The state by which the sensor array by which the several sensor cell was arrange | positioned in the matrix form was provided in the integrated circuit board is shown. Show. It is a partial circuit diagram explaining the electrical constitution of the sensor matrix of FIG. It is a schematic diagram which expands and shows the cross section in order to demonstrate the principal part structure of the sensor cell of FIG. In a sensor cell, it is a figure which shows the relationship between the density | concentration ratio [Red] / [Ox] of the electric potential Vg, the density | concentration [Red] of a reduced product, and the density | concentration [Ox] of an oxide. Using sensor cell CE containing beads immobilized on the surface with three kinds of enzymes, HK (hexokinase), G6PDH (glucose-6-phosphade dehydrogenase), and Diaphorase as a probe molecule, glucose, maltose with a similar molecular structure, It is a figure explaining the test result which each detects glucose from the sample solution K containing galactose by three types of enzyme reactions. FIG. 2 is a circuit diagram illustrating a configuration example of a detection circuit provided for each sensor cell in the semiconductor integrated circuit substrate of FIG. 1. It is a time chart explaining the action | operation of the detection circuit of FIG. It is a figure explaining the circuit structural example of the current-voltage pulse conversion circuit of FIG. It is a time chart explaining the action | operation of the differential amplifier in the current voltage pulse conversion circuit of FIG. It is a figure which shows the circuit structural example of the differential amplifier in the current voltage pulse conversion circuit of FIG. This is a micrograph of a chip in which a semiconductor integrated circuit including a 128 × 128 sensor matrix, a short pulse generator, a decoder, a current / voltage pulse conversion circuit, and a TDC is formed on a Si substrate by using a process having a minimum line width of 0.6 μm. . It is the microscope picture which expanded the part of (a) containing a part of short pulse generator of FIG. 11, a decoder, and a sensor matrix. 12 is an enlarged micrograph of a part (b) including a part of the TDC, sensor matrix, and current / voltage pulse generation circuit of FIG. In the circuit of FIG. 11, it is the figure which actually measured the operation | movement waveform shown in FIG. 9 with the oscilloscope. It is a figure which shows the relationship between the detection electric potential Vg by the detection electrode and pulse width CPW in 128 * 128 sensor matrix SM of FIG. FIG. 12 is a diagram showing the relationship between the electrostatic capacitance CB of the bit line B and the pulse width CPW of the current pulse signal CP in the semiconductor integrated circuit shown in FIG. 2 is a chart showing a comparison between a pulse width lower limit value and energy consumption per pulse when a voltage pulse signal is used and when a current pulse signal CP is used in the semiconductor integrated circuit substrate shown in FIG. 16 is a chart for comparing the area of the sensor cell CE in the operation in the strong inversion region of FIG. 15 between the case where the voltage pulse signal is used and the case where the current pulse signal is used. 12 is a photomicrograph of a sensor matrix portion in which the detection electrode, base resin, and partition walls shown in FIG. 3 are provided on the semiconductor integrated circuit shown in FIG. FIG. 20 is a photograph in which a solution supply / discharge unit is mounted on the semiconductor integrated circuit shown in FIG. 19 and connected to an arithmetic circuit. FIG. 7 is a circuit diagram for explaining a main part of a detection circuit according to another embodiment of the present invention, corresponding to FIG. 6. FIG. 7 is a circuit diagram for explaining a main part of a detection circuit according to another embodiment of the present invention, corresponding to FIG. 6. FIG. 7 is a circuit diagram for explaining a main part of a detection circuit according to another embodiment of the present invention, corresponding to FIG. 6. FIG. 7 is a circuit diagram for explaining a main part of a detection circuit according to another embodiment of the present invention, corresponding to FIG. 6. FIG. 7 is a circuit diagram for explaining a main part of a detection circuit according to another embodiment of the present invention, corresponding to FIG. 6. It is a schematic diagram which shows the cross section of the other sensor cell of this invention, Comprising: It is a figure equivalent to FIG.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a principal circuit diagram illustrating an example of a configuration of a massively parallel biomolecule detection apparatus 10 to which the present invention is applied, and FIG. 2 is a diagram illustrating an electrical configuration of a sensor matrix SM. 1 and 2, a semiconductor integrated circuit substrate 12, such as a CMOS integrated circuit substrate, has a plurality of word lines W1, W2,... Wm and a plurality of word lines W1, W2,. A plurality of sensor cells CE11, CE12,... CEmn located in the vicinity of intersections with a plurality of bit lines B1, B2,. That is, the word lines W1, W2,... Wm, the bit lines B1, B2,... Bn, and the sensor cells CE11, CE12,.

  A constant frequency clock signal CL output from the clock circuit 14 is supplied to the short pulse signal generator 16 and the decoder 18. The short pulse signal generator 16 outputs a short pulse signal TP having a preset pulse width to the decoder 18 in synchronization with the frequency of the clock signal CL or its decreasing frequency. A plurality of word lines W1, W2,... Wm are connected to the decoder 18, and the decoder 18 outputs a plurality of word lines W1, W2,. When one is selected from Wm and the clock signal CL is received, the word line Wm to which the short pulse signal TP is output is sequentially switched.

  The current-voltage pulse conversion circuit 20 responds to the selective switching of the plurality of word lines W1, W2,... Wm from the selected sensor cell CE through the bit lines B1, B2,. The output current pulse signal CP is converted into a voltage pulse signal VP and sequentially output to the time digital conversion circuit 22. The time digital conversion circuit 22 converts the voltage pulse signal VP received from the current-voltage pulse conversion circuit 20 into a digital signal DCS composed of a bit signal representing the time width, and sequentially outputs it to the arithmetic circuit 24. In FIG. 1, the current / voltage pulse conversion circuits are provided on the upper and lower sides, but they may be collected on either the upper side or the lower side.

  The arithmetic circuit 24 is constituted by a microcomputer, for example, and based on the pulse width CPW of the current pulse signal CP represented by the digital signal DCS input from the time digital conversion circuit 22 for each sensor cell CE, the digital signal DCS The amount of the biomolecule M in the sample solution K to which the probe molecule 50 in the corresponding sensor cell CE reacts is calculated.

  FIG. 3 is a schematic diagram showing a cross-section for explaining the configuration of one sensor cell CE in the sensor matrix SM on the semiconductor integrated circuit substrate 12. In FIG. 3, a grid-like partition wall 30 having a space of a predetermined interval is erected on a semiconductor integrated circuit substrate 12 via a base resin 32 such as a polyimide resin, and is disposed above the partition wall 30. The opening of the partition wall 30 is covered by the flat cover plate 34 thus formed. As a result, a space of the sensor cell CE surrounded by the substrate 12, the partition wall 30, and the cover plate 34 is formed on the surface of the semiconductor integrated circuit substrate 12. A sample solution passage PK for introducing the sample solution K into the space of each sensor cell CE is provided between the leading edge of the partition wall 30 and the back surface of the cover plate 34. The partition wall 30 is formed from a photosensitive resin such as a photosensitive epoxy resin by using a photolithography technique. The cover plate 34 is made of silicone rubber obtained by vulcanizing, for example, polydimethylsiloxane (PDMS).

  After the circuit shown in FIG. 8 including the circuit of FIG. 6 to be described later is formed on the Si substrate 40 through a well-known standard CMOS integrated circuit process, Cr and Au are sequentially deposited on the semiconductor integrated circuit substrate 12. A plurality of detection electrodes (gold electrodes) 42 formed by patterning by photolithography and etching are provided in a state of being exposed one by one in the space in each sensor cell CE. Al used in the wiring layer 44 easily reacts with the solution, and contact with the solution must be prevented. The step of the opening of the Al wiring layer is 1.5 μm, and the thickness of the detection electrode 42 cannot be completely covered by 0.15 μm. The base resin 32 covers the peripheral edge of the detection electrode 42 to prevent the Al wiring layer 44 from coming into contact with the solution, and the adhesion between the photosensitive resin constituting the partition wall 30 and the detection electrode 42. It also functions as an intermediate layer for improving.

  The detection electrode 42 is immersed for a certain time in ethanol containing a ferrocene derivative such as 11-Ferrocenyl-1-Undercanethiol (11-FUT), and the surface thereof is coated (modified) with a self-assembled monolayer film 46 of the ferrocene derivative. Has been. The ferrocene derivative is a stable electron mediator and contributes to the protection of the detection electrode 42 and the detection range.

  Probe molecules 50 functioning as probes (probing) are fixed on the surface of beads 48 preferably made of magnetic beads. The sample solution K contains a detected biomolecule M such as sugar, virus, DNA, protein, etc., and the probe molecule 50 reacts only with a specific biomolecule among the biomolecules in the sample solution K. A molecule such as an enzyme, a single-stranded DNA having a specific sequence, a peptide that reacts with a specific protein, an antibody, and the like. Different probe molecules 50 are immobilized on the surface of the bead 48 in each sensor cell CE. Preferably, the number of sensor cells CE is a number corresponding to the number of types of probe molecules 50. Alternatively, beads with the same type of probe molecule fixed may be provided in a plurality of sensor cells to provide redundancy and improve accuracy by statistical processing. The beads 48 in each sensor cell CE function as a biomolecule recognition unit.

  In the sensor cell CE, the potential Vg representing the concentration ratio of the oxide and the reduced product is generated by the enzyme reaction between the biomolecule M to be detected and the probe molecule 50 in the sample solution K. Therefore, the potential Vg is the self of the ferrocene derivative. Detection is performed by the detection electrode 42 covered (modified) with the organized monolayer film 46. The relationship between the potential Vg and the concentration ratio [Red] / [Ox] of the reduced product concentration [Red] and the oxide concentration [Ox] is given by the Nernst equation shown in the following equation (1). Reference numeral 4 indicates an experimentally obtained relationship.

FIG. 4 shows a detection electrode 42 in which 50 nm of Cr and 100 nm of Au are sequentially deposited and 11-FUT is modified, and [Fe (CN) ₆ ] ⁻⁴ as a reduction product and [Fe (CN) as an oxide. ₆ ] The relationship obtained using ^-3 is indicated by a circle. Further, the Δ mark indicates the relationship when the detection electrode 42 is not modified with 11-FUT. The relationship indicated by ◯ and the relationship indicated by △ indicate good linearity with a slope of 58 mV / decade in the range of the concentration ratio [Red] / [Ox] from 10 ⁻² to 10 ¹ .

_{^{Vg = Vg O - (k B}} T / q) ln ([Red] / [Ox]) ··· (1)
However, k _B is the Boltzmann constant, T is the absolute temperature, q is the elementary charge quantity.

  Here, the present inventors used a sensor cell CE containing beads 48 immobilized on the surface with three types of enzymes, HK (hexokinase), G6PDH (glucose-6-phosphade dehydrogenase), and Diaphorase as probe molecules 50, Test example in which glucose is detected by three kinds of enzyme reactions shown in Formula (2), Formula (3), and Formula (4) from a sample solution K containing glucose, maltose and galactose having a similar molecular structure. Is described below.

Glucose + ATP → (HK) → glucose-6-phosphade + ADP (2)
Glucose-6-phosphade + NAD → (G6PDH) → glucose-6-phosphade + NADH (3)
2 [Fe (CN) ₆ ] ^-3 + NADH → (Diaphorase) 2 [Fe (CN) ₆ ] ^-4 (4)

  As shown in FIG. 5, a glucose-only signal was detected, indicating the specificity of not reacting with maltose and galactose each having a molecular structure similar to glucose. This revealed that only glucose in the sample solution K was measured. In this example, the detection of glucose is described as an example, but other biomolecules can be detected by changing the type of the probe molecule 50 immobilized on the bead 48. In the sensor matrix SM in which such sensor cells CE are arranged in a matrix on one chip (semiconductor integrated circuit substrate 12), by providing beads 48 each having a different probe molecule 50 fixed to each sensor cell CE, Parallel detection is possible.

FIG. 6 is a diagram illustrating a basic configuration of a detection circuit DC that functions as a current pulse signal generation unit provided for each sensor cell CE. In FIG. 6, the positive power supply voltage V _DD , the negative power supply voltage V _SS , and the voltage V _B of the bit line B are maintained at constant values. The detection signal of the biomolecule is represented by the current I output from the current source I. The node point n has a capacitance C. From the constant current source CS constant current I _S is outputted. FIG. 7 is a time chart for explaining the operation of the detection circuit DC. As shown in FIG. 7, in the non-selection period of the word line W in which the short pulse signal TP is not applied to the word line W from the short pulse generator TG, which will be described later, the voltage V _W of the word line _W becomes the positive power supply voltage V _DD . The PMOSFET P ₁ is maintained in the off state, the node point n is set to the negative power supply voltage V _SS , the NMOSFET N ₁ is in the off state, and no current flows through the bit line B. However, short pulses at selected applied to the word line W, after the word line voltage V _W is pulled up to a short time the negative supply voltage V _SS, and returned to the positive supply voltage V _DD. During this time, when the word line voltage _{V W} is lowered to the negative supply voltage _{V SS} PMOSFET P ₁ is turned on, the node point voltage _{V n} is increased to the power supply voltage _{V DD.} PMOSFET P ₁ when the voltage _{V W} of the word line W is returned to a positive supply voltage _{V DD} is turned off, but the node point voltage _{V n} is held by the electrostatic capacitance C, gradually decreases by the current I. When a node point voltage _{V n} is between the positive power supply voltage _{V DD} and the threshold voltage _{V th,} NMOSFET N ₁ flows a current -I _S to the bit line B turned on. The threshold voltage V _th is a voltage for switching the NMOSFET N ₁ to the ON state. The width CPW of the current pulse output from the constant current source CS while the NMOSFET N ₁ is in the on state is given by the following equation (5).

CPW = C (V _DD −V _th ) / I (5)
However, since I is a signal having information on biomolecules, information on biomolecules can be transmitted as a pulse width.

FIG. 8 is a circuit diagram illustrating a detection circuit DC provided for each sensor cell CE and a current-voltage pulse conversion circuit 20 provided for each bit line B. In the detection circuit DC of FIG. 8, the current source I of the detection circuit DC of FIG. 6 is configured by an NMOSFET N ₂ , and the constant current source Is is configured by an NMOSFET N ₃ . NMOSFET N ₃ and NMOSFET N ₄ are composed of NMOSFETs of the same size (same characteristics) in order to obtain the same drain voltage when a current is passed, and PMOSFET P is used to pass the same drain current with respect to a common gate current. ₂ and PMOSFET P ₃ are composed of PMOSFETs of the same size (same characteristics), and NMOSFETs N _{6 to} NMOSFET N ₉ have the same size (same characteristics) in order to function as constant current sources for supplying the same drain current determined by the constant current source ISS. ) NMOSFET.

In FIG. 8, the short pulse generator TG outputs a short pulse signal TP in response to the rising of the input clock signal Φ. In the non-selected period of the word line W in which the short pulse signal TP is not applied to the word line W, the voltage V _W of the word line W is maintained at the positive power supply voltage V _DD , the PMOSFET P ₁ is turned off, and the node point n is The negative power supply voltage V _SS is set, the NMOSFET N ₁ is turned off, and no current flows through the bit line B. However, when a short pulse is applied to the word line W, the node point voltage V _n is selected as described above. when there is between the positive supply voltage _{V DD} and the threshold voltage _{V th,} NMOSFET N ₁ is turned on, a current -I _S flows in the bit line B, and corresponds to a potential Vg that has been detected by the sensing electrodes 42 A current pulse CP having a length pulse width CPW is output. The differential amplifier da of FIG. 8 functions as a gain boost of the NMOSFET N ₅ that functions as a source follower, and serves to keep the potential of the bit line B constant. Period of the pulse width of the current pulse signal CP, i.e. in the section NMOSFET N ₁ is in the ON state, a current pulse signal CP is generated in the bit line B having a pulse width CPW represented by equation (5), a current pulse A voltage pulse signal VP having the same pulse width as the signal CP is output from the output terminal Out. The time chart of FIG. 9 shows such an operation. FIG. 10 shows a circuit configuration example of the differential amplifier da.

FIG. 11 shows a process using a minimum line width of 0.6 μm, a Si substrate 40, a 128 × 128 sensor matrix SM, a short pulse generator 16, a decoder 18, a current / voltage pulse conversion circuit 20, a time digital converter TDC. 5 is a micrograph of a chip on which a semiconductor integrated circuit including (time to digital converter) 22 is created. 12 and 13 are enlarged micrographs of parts (a) and (b) of FIG. FIG. 14 is a diagram in which a waveform corresponding to the operation of FIG. 9 is measured with an oscilloscope using this chip. Figure 15 shows the relationship between Vg and the pulse width CPW is a predetermined potential by the detection electrode 42 (the gate voltage of the NMOSFET N _2). In the subthreshold region where the gate voltage Vg of the NMOSFET N ₂ is relatively low, the pulse width CPW changes nonlinearly (linearly with a logarithmic axis) with respect to the gate voltage Vg. In this subthreshold region, the pulse width changes by almost 6 digits, which is effective for identifying the detection signal (detection potential), but realization of a time digital converter TDC that converts the pulse width CPW into a digital bit string signal DCS. Is difficult. On the other hand, in the strong inversion region where the gate voltage Vg is relatively high, the pulse width is as short as 4 ns, but the realization of the time digital converter TDC becomes easy. For this reason, in this embodiment, when the detection information (gate voltage Vg) is converted into the pulse width CPW of the current pulse signal CP, the above strong inversion region is used.

  The technical significance of transmitting detection information (gate voltage Vg) using the current pulse signal CP of the present embodiment will be described below. In general, the larger the number of sensor cells CE and the larger the number of sensor matrix SM, the greater the bit line capacitance and the number of transistors connected to the bit line. The amount increases. For this reason, if a voltage pulse is used to transmit the detection information, a large restriction occurs in the lower limit of the pulse width (that is, the upper limit of the signal frequency), and the pulse width is limited to, for example, 100 ns or more. Detection cannot be performed in the strong inversion region of FIG. In order to use the strong inversion region, it is necessary to provide a capacitance C of 10 pF or more in the sensor cell. In this case, the sensor cell CE has a capacitor of 100 μm × 100 μm or more (0.6 μm process). (In the case of the used polycrystalline silicon two-layer capacitor), the area of the sensor cell CE is increased by two orders of magnitude. On the other hand, when the current pulse signal CP of this embodiment is used, the capacitance C can be realized by the parasitic capacitance of the transistor, so that it is not necessary to actively provide a capacitor, and a semiconductor integrated circuit having the same chip size. The substrate 12 can be provided with 100 times the number of sensor cells, and various types of biomolecules can be simultaneously measured at high speed. The semiconductor integrated circuit board 12 on which the large-scale sensor matrix SM is mounted can function as the massively parallel biomolecule detection apparatus 10.

  FIG. 16 shows the relationship between the capacitance CB of the bit line B and the pulse width CPW of the current pulse signal CP. According to this relationship, it is shown that even if the capacitance CB of the bit line B increases to 15 pF, it is possible to sufficiently conduct detection information using a pulse width CPW current pulse signal CP of about 4 ns. Has been.

Next, the lower limit of the pulse width and the energy consumption per pulse when the voltage pulse signal is used and when the current pulse signal CP is used will be described. In actual semiconductor integrated circuit substrate 12, considering the drain capacitance and the wiring capacitance of the NMOSFET N _1, the capacitance of each bit line B is for example of the order of 10 pF. When driving at a drain current of 100 μA, when trying to transmit detection information with a voltage pulse with an amplitude of 1 V, the lower limit value of the width of the transmission pulse is 10 pF × 1 V / 100 μA = 100 ns, and the energy per pulse is 10 pF. * 1V * 1V = 10 pJ. On the other hand, when the current pulse signal CP is used, the lower limit value of the pulse width CPW is 4 ns, and the energy per pulse is 4 ns × 100 μA × 1 V = 0.4 pJ. FIG. 17 is a chart showing these values in comparison.

Next, the area of the sensor cell CE when operated in the strong inversion region of FIG. 15 will be compared. When importance is attached to the linearity between the gate voltage Vg and the pulse width CPW, it is necessary to operate the detection transistor (NMOSFET N ₂ ) in the strong inversion region. When the minimum transistor is used, the drain current for operating in the strong inversion region is approximately 0.5 to 100 μA. In order to convert this region into a pulse width equal to or greater than the pulse width lower limit value, the capacitance of the node point n in the detection circuit DC of the sensor cell CE is 100 μA × 100 ns / 1 V = 10 pF in the voltage pulse signal, and in the current pulse signal. 100 μA × 4 ns / 1 V = 0.4 pF is required. The largest capacitance per unit area obtained by the analog / digital mixed process of 0.6 μm is a polycrystalline silicon capacitor of 0.0009 pF / μm ² . Therefore, the capacitor region in the case of using the voltage pulse signal needs a region of 100 μm × 100 μm, whereas the capacitor region in the case of using the current pulse signal needs a region of 20 μm × 20 μm. Further, since the parasitic capacitance associated with the node point n is approximately 0.02 pF, when a current pulse is used, a strong inversion region of 0.5 to 5 μA is covered without intentionally forming a capacitor. be able to. FIG. 18 is a chart for comparing the areas of the sensor cells CE in this strong inversion region operation.

  FIG. 19 is a photomicrograph of a sensor matrix portion in which the detection electrodes 42, the base resin 32, and the partition walls 30 of FIG. 3 are provided on a semiconductor integrated circuit. The detection electrode is shifted from the center of the partition opening so that the detection electrode is not covered by the beads 48. FIG. 20 is a photograph in which the solution supply / discharge unit is mounted on the semiconductor integrated circuit and connected to the arithmetic circuit 24. The arithmetic circuit includes a microcontroller unit, a memory, a digital input / output circuit, a liquid crystal display, and the like. As described above, according to the massively parallel biomolecule detection apparatus 10 of the present embodiment, when the sample solution K is introduced into the plurality of sensor cells CE, the sample solution K is detected in the detection circuit (current pulse signal generation unit) DC. A current pulse signal CP that changes the pulse width CPW according to the potential Vg generated by the reaction between the biomolecule M in the probe and the probe molecule 50 is generated, and the amount of the biomolecule M is detected based on the current pulse signal CP. The As a result, even if a current pulse signal CP having a short signal width required for high-speed measurement is used, it is affected by the number of sensor cells CE and the stray capacitance that is increased by increasing the size of the semiconductor integrated circuit board 12. Therefore, high detection accuracy can be obtained.

  In addition, according to the massively parallel biomolecule detection apparatus 10 of the present embodiment, the plurality of sensor cells CE include the plurality of word lines W and the plurality of word lines W intersecting the plurality of word lines W in the semiconductor integrated circuit substrate 12. The sensor matrix SM (sensor array circuit) is configured by being arranged in a matrix in the vicinity of the intersection with the bit line B. Thereby, since a large number of minute sensor cells CE can be arranged on the semiconductor integrated circuit substrate 12, the amount of many types of biomolecules M in the sample solution K can be detected simultaneously and efficiently.

  Further, according to the massively parallel biomolecule detection apparatus 10 of the present embodiment, the plurality of sensor cells CE are fixed to the semiconductor integrated circuit substrate 12 and exposed to the plurality of sensor cells CE, respectively, and are self-assembled monolayers. A plurality of detection electrodes 42 that are in contact with the sample solution K in a state of being covered with a membrane (ferrocene derivative) 46, and beads (probe carrier particles) 48 that are suspended in the sample solution K in a state where the probe molecules 50 are fixed on the surface. Is provided. For this reason, the biomolecule M in the sample solution K introduced into the sensor cell CE is electrically detected, and the biomolecule M can be easily detected.

  In addition, according to the massively parallel biomolecule detection apparatus 10 of the present embodiment, a detection circuit (current pulse signal generation unit) DC is provided for each sensor cell CE, and a plurality of detection electrodes 42 and a plurality of word lines W and A current pulse signal CP having a pulse width CPW corresponding to the potential Vg provided between the bit lines B and detected by the plurality of detection electrodes 42 is output. For this reason, since the current pulse signal CP having the pulse width CPW corresponding to the potential detected by the plurality of detection electrodes 42 respectively disposed in the plurality of sensor cells CE is output, it is a short necessary for high-speed measurement. Even when the current pulse signal CP having the signal width is used, high detection accuracy can be obtained without being affected by the number of sensors and the stray capacitance that is increased by increasing the scale of the integrated circuit.

  Further, according to the massively parallel biomolecule detection apparatus 10 of the present embodiment, the current voltage for converting the current pulse signal CP transmitted through the bit lines B sequentially selected from the plurality of bit lines B into the voltage pulse signal VP. The calculation circuit (biomolecular weight calculation unit) 24 includes a pulse conversion circuit 20 and a time digital signal conversion circuit TDC that converts a digital signal DCS composed of a bit string representing the time width of the voltage pulse signal VP. From the relationship, the amount of the biomolecule M that has reacted with the probe molecule 50 in the sensor cell CE is calculated based on the actual potential Vg represented by the digital signal DCS. For this reason, there exists an advantage by which the quantity of many types of biomolecule M is detected simultaneously.

  Further, according to the massively parallel biomolecule detection apparatus 10 of the present embodiment, a short pulse signal for outputting the current pulse signal CP from the detection circuit (current pulse signal generation unit) DC to the plurality of word lines W. Short pulse signal generators 16 for outputting TP are connected to each other. In response to the short pulse signal TP output from the short pulse signal generator 16, the current pulse signal CP is output from the detection circuit (current pulse signal generation unit) DC. There is an advantage of being detected in parallel.

Further, according to the massively parallel biomolecule detection device 10 of the present embodiment, the detection circuit (current pulse signal generation unit) DC includes an NMOSFET N ₂ (detection transistor) having a gate connected to the detection electrode 42, and PMOSFET P ₁ (selection transistor) that connects the drain of the detection transistor to the first power supply V _DD according to the short pulse signal TP of the word line W and the drain connected to the bit line B, the drain voltage of the detection transistor the NMOSFET N ₁ to switch the (switch transistor), and a NMOSFET N ₃ is connected between the source and lower than the first power supply _{V DD} second power supply _{V SS} of the switch transistor (diode) provided. In this way, a current pulse signal CP having a pulse width CPW corresponding to the potential Vg detected by the plurality of detection electrodes 42 respectively disposed in the plurality of sensor cells CE is output.

FIG. 21 shows another configuration example of the detection circuit DC. The detection circuit DC shown in FIG. 21, with respect to the detection circuit DC of FIG. 8, between the node point n and the second power supply V _SS, and differs in that a capacitor C1 is provided, other configurations are similar It is. In the detection circuit DC of FIG. 21, since the capacitance of the capacitor C1 is added in addition to the parasitic capacitance of the node point n, there is an advantage that the operation is stabilized. In a state where a current pulse is being output, the node point n is a voltage higher than the threshold value of the NMOSFET, and therefore a capacitor using an NMOSFET can be used for C1.

FIG. 22 also shows another configuration example of the detection circuit DC. The detection circuit DC shown in FIG. 22 differs from the detection circuit DC shown in FIGS. 6 and 8 in that a capacitor C2 is provided between the node point n and the word line W. It is the same. In the detection circuit DC of FIG. 22, since the capacitance of the capacitor C2 is added in addition to the parasitic capacitance of the node point n, there is an advantage that the operation is stabilized. Since the voltage at the node point n is boosted when the voltage of the word line W returns from V _SS to V _{DD, the} current pulse time can be lengthened to improve accuracy.

FIG. 23 also shows another configuration example of the detection circuit DC. In the detection circuit DC shown in FIG. 23, a resistor R _{S is provided} between the source of the NMOSFET N ₂ functioning as a detection transistor and the second power supply V _SS with respect to the detection circuit DC shown in FIGS. Is different, and the other configurations are the same. The detection circuit DC of FIG. 23 has an advantage that the linearity between the potential Vg and the pulse width CPW of the current pulse signal CP is improved. A diode-connected MOSFET (with the gate and drain short-circuited) may be used instead of the resistor _RS .

FIG. 24 also shows another configuration example of the detection circuit DC. The detection circuit DC shown in FIG. 24, FIG. 6, with respect to the detection circuit DC of Fig. 8, instead of the NMOSFET N _3, and differs in that a bipolar pn diode are provided, other configurations are similar. In the detection circuit DC of FIG. 24, the voltage across the bipolar pn diode is set as low as about 0.6 V, and there is an advantage that the power supply voltage can be reduced.

FIG. 25 also shows another configuration example of the detection circuit DC. The detection circuit DC shown in FIG. 25 is different from the detection circuit DC shown in FIGS. 6 and 8 in that an NMOSFET N ₁₀ and a POSFET P ₁₀ are provided, and the other configurations are the same. It is possible to increase the charging time of the node point n, the POSFET P ₁ is a select transistor can be configured with a minimum transistor, it is possible to reduce the area of the sensor cell.

  In the above embodiment, the potential is detected by the gold electrode coated with the ferrocene derivative. However, a method of detecting the electric charge in the vicinity of the detection electrode may be used. The schematic diagram showing the cross section of the sensor cell shown in FIG. 26 is different from the schematic diagram of FIG. 3 in that an ion sensitive film 60 is provided on the detection electrode and a change in the amount of interfacial charge is detected as a potential change. However, other configurations are the same. For example, a change in pH due to a molecular reaction is detected as a change in the amount of interfacial charge in a silicon nitride film or tantalum oxide that is a proton sensitive film. In this case, the detection electrode is insulated from the solution, and no electrons are exchanged between the detection electrode and the solution.

  As mentioned above, although one Example of this invention was described based on drawing, this invention is applied also in another aspect.

  For example, in the above-described embodiment, the massively parallel biomolecule detection apparatus 10 including the semiconductor integrated circuit board 12 on which the sensor matrix SM composed of the sensor cells CE arranged vertically and horizontally is used. However, the sensor cells are not necessarily aligned vertically and horizontally. The sensor matrix SM made of CE may not be used, and the sensor matrix SM may not be provided in one plane.

  In the semiconductor integrated circuit substrate 12 of the above-described embodiment, 128 × 128 detection circuits DC, 128 word lines W and bit lines B, and 128 × 128 sensor cells CE are formed. In addition to that, a part or all of the decoder 18, the 128 current-voltage pulse conversion circuits 20, and the time digital conversion circuit 22 may be provided.

  In addition, although not illustrated one by one, the present invention can be variously modified without departing from the spirit of the present invention.

10: Massively parallel biomolecule detection device 12: Semiconductor integrated circuit board 14: Clock circuit 16: Short pulse signal generator 18: Decoder 20: Current / voltage pulse conversion circuit DC: Detection circuit (current pulse signal generator)
CE: sensor cell 22: time digital conversion circuit 24: arithmetic circuit 42: detection electrode (current pulse signal output circuit)
46: Self-assembled monolayer film 48: Beads (probe carrier particles)
50: Probe molecule 60: Ion sensitive film W: Word line B: Bit line TP: Short pulse signal DCS: Digital signal CP: Current pulse signal CPW: Pulse width CL: Clock signal SM: Sensor matrix Vg: Detection potential (gate Voltage)
M: Biomolecule K: Sample solution P ₁ : PMOSFET (selection transistor)
N _1: NMOSFET (switch transistor)
N ₂ : NMOSFET (detection transistor)
N _3: NMOSFET (diode)

Claims (9)

In a massively parallel biomolecule detection method comprising a plurality of sensor cells each containing one of a plurality of types of probe molecules and having a plurality of sensor cells into which a sample solution is introduced, and simultaneously measuring the amount of biomolecules in the sample There,
Generating a current pulse signal that changes a pulse width in accordance with a potential change generated by a reaction between a biomolecule in the sample solution and the probe molecule, and detecting the amount of the biomolecule based on the current pulse signal; A method for detecting biomolecules in parallel. A massively parallel biomolecule detection apparatus that includes a plurality of sensor cells each containing one of a plurality of types of probe molecules and into which a sample solution is introduced, and that simultaneously measures the amount of biomolecules in the sample solution Because
A current pulse signal generation unit that generates a current pulse signal that changes a pulse width according to a potential change generated by a reaction between the biomolecule in the sample solution and the probe molecule;
A massively parallel biomolecule detection device comprising: a biomolecule amount calculation unit that calculates the amount of the biomolecule based on a pulse width of the current pulse signal from a predetermined relationship. The plurality of sensor cells are arranged in a matrix at intersections of a plurality of word lines and a plurality of bit lines intersecting with the plurality of word lines on a semiconductor integrated circuit substrate to form a sensor matrix. The massively parallel biomolecule detection apparatus according to claim 2, wherein: The plurality of sensor cells include a plurality of detection electrodes fixed to the semiconductor integrated circuit substrate and exposed in the plurality of sensor cells, respectively, and in contact with the analyte solution in a state of being covered with a ferrocene derivative, and the probe molecules. The massively parallel biomolecule detection device according to claim 3, wherein probe carrier particles are provided that float on the sample solution in a state of being fixed to the surface. In the plurality of sensor cells, a plurality of detection electrodes fixed to the semiconductor integrated circuit substrate and provided in the plurality of sensor cells, an ion sensitive film provided in the vicinity of the detection electrodes and in contact with the analyte solution, The massively parallel biomolecule detection device according to claim 3, wherein probe carrier particles are provided that float on the sample solution in a state in which the probe molecules are fixed on the surface. The current pulse signal generation unit is provided between the plurality of detection electrodes and the plurality of word lines and bit lines, and has a pulse width corresponding to the potential detected by the plurality of detection electrodes. The massively parallel biomolecule detection device according to claim 4 or 5, comprising a plurality of detection circuits for outputting signals. A current-voltage pulse conversion circuit that converts the current pulse signal transmitted from the plurality of bit lines sequentially on the bit line selected to a voltage pulse signal, and a digital signal composed of a bit string representing the time width of the voltage pulse signal Including a time digital signal conversion circuit,
The biomolecular weight calculating unit calculates the amount of biomolecules that have reacted with the probe in the cell based on an actual potential represented by the digital signal from a previously stored relationship. Item 7. The massively parallel biomolecule detection device according to Item 6. The short pulse signal generator for outputting a short voltage pulse for outputting a current pulse signal from the current pulse signal output circuit is connected to each of the plurality of word lines. 8. The massively parallel biomolecule detection device according to 7. The current pulse signal output circuit is
A detection transistor having a gate connected to the detection electrode;
A selection transistor for connecting a drain of the detection transistor to a first power supply in response to a short pulse signal of the word line;
A switch transistor having a drain connected to the bit line and switching according to a drain voltage of the detection transistor;
The massively parallel biomolecule detection device according to claim 8, further comprising: a diode connected between a source of the switch transistor and a second power source lower than the first power source.