Device for molecular detection, method and system for nucleic acid sequencing

Abstract

Disclosed herein are devices for molecular detection and methods for molecular detection using such devices. The molecular detection device includes: at least one fluid channel, the at least one fluid channel being configured to receive a molecule to be detected; a sensor, the sensor including a spin torque oscillator (STO) and encapsulated by a material separating the sensor from the at least one fluid channel; and a detection circuit coupled to the sensor. At least some of the molecules to be detected are labeled with magnetic nanoparticles (MNPs). The surface of the material provides binding sites for the molecules to be detected. The detection circuit is configured to detect the frequency or frequency noise of a radio frequency (RF) signal generated by the STO in response to the presence or absence of at least one MNP coupled to one or more binding sites associated with the sensor.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 62 / 833,167, filed on April 12, 2019, entitled “SPIN TORQUE OSCILLATOR (STO) SENSORS USED IN NUCLEIC ACID SEQUENCING ARRAYS AND DETECTION SCHEMES FOR NUCLEIC ACID SEQUENCING INCLUDING FREQUENCY-AND PHASE-BASED DETECTION SCHEMES” and U.S. Non-Provisional Application No. 16 / 819,636, filed on March 16, 2020, entitled “DEVICES AND METHODS FOR FREQUENCY-AND PHASE-BASED DETECTION OF MAGNETICALLY-LABELED MOLECULES USING SPIN TORQUE OSCILLATOR (STO) SENSORS”, both of which are hereby incorporated by reference in their entirety for all purposes. This application also incorporates by reference the entire contents of U.S. Application No. 16 / 791,759, filed on February 14, 2020, entitled “SPIN TORQUE OSCILLATOR (STO) SENSORS USEDIN NUCLEIC ACID SEQUENCING ARRAYS AND DETECTION SCHEMES FOR NUCLEIC ACID SEQUENCING” for all purposes. Technical Field

[0003] Embodiments of the present disclosure are generally directed to magnetoresistive (MR) sensor arrays for detecting molecules coupled to magnetic nanoparticles (MNPs), such as for nucleic acid sequencing, such as deoxyribonucleic acid (DNA) sequencing, and methods of molecular detection using such MR sensor arrays. Background Art

[0004] Description of Related Art

[0005] Current state-of-the-art sequencing systems are based on fluorescent signal detection and offer a throughput of 20 billion reads per run (https: / / www.illumina.com / systems / sequencing-platforms / novaseq.html). However, achieving this performance may require large-area flow cells, high-precision free-space imaging optics, and expensive high-power lasers to generate enough fluorescent signals to successfully detect bases.

[0006] A type of nucleic acid sequencing for DNA sequencing is called "synthesis sequencing" (SBS). SBS involves the binding of template DNA for primer hybridization, the incorporation of deoxynucleoside triphosphates (dNTPs), and the detection of the incorporated dNTPs. Two methods have been used to achieve a gradual increase in SBS throughput, the first of which is outward scaling, in which the size and number of flow cells in the sequencer are increased. This method increases the cost of reagents and the price of the sequencing system because higher power lasers and high-precision nanopositioners must be used. The second method involves inward scaling, in which the density of DNA test sites is increased so that the total number of DNA chains sequenced in a fixed-size flow cell is greater. In order to achieve inward scaling, lenses with increasingly higher numerical apertures (NA) must be used to distinguish signals from adjacent fluorophores when the spacing between adjacent fluorophores is reduced. However, this method cannot be implemented indefinitely because the Rayleigh criterion sets the distance between resolvable light sources to 0.61λ / NA, thereby constraining the minimum distance between two sequencing DNA chains to be no less than about 400nm. Similar resolution limits apply to sequencing performed directly on top of an imaging array (similar to a cell phone camera), where the smallest pixel size achieved to date is approximately 1 μm (https: / / www.ephotozine.com / article / complete-guide-to-image-sensor-pixel-size-29652).

[0007] The Rayleigh criterion currently represents a fundamental limit to the inward scaling of optical SBS systems, which can only be overcome by applying super-resolution imaging techniques (see AM Sydor, KJ Czymmek, EMPuchner, and V. Mannella, "Super-Resolution Microscopy: From Single Molecules to Supramolecular Assemblies" Quantitative Cell Biology Special Issue, Vol. 25, p. 730, 2015), and has not yet been achieved in highly multiplexed systems. Therefore, the increase in throughput and the reduction in cost of optical SBS sequencers have been slow due to the need to build larger flow cells and implement more expensive optical scanning and imaging systems.

[0008]

[0006] Therefore, there is a need for new and improved devices and methods for detecting the presence of molecules, such as nucleic acids, that overcome the limitations of conventional devices and methods. Summary of the invention

[0009] This summary represents non-limiting embodiments of the present invention.

[0010] Improved detection devices, systems and methods for identifying molecules (e.g., nucleic acids) using magnetic nanoparticles (MNPs) are disclosed herein. The disclosure herein includes embodiments of sensors having spin torque oscillators (STOs) that allow for detection of properties that indicate the presence or absence of MNPs near the sensor. Detection methods are also disclosed herein that can be used to detect (e.g., measure or obtain) a property or property change of a radio frequency (RF) signal generated by the sensor, wherein the detected property or property change indicates the presence or absence of MNPs (e.g., in response to a magnetic field or the absence of a magnetic field generated by a magnetic nanoparticle tag).

[0011] In some embodiments, a device for molecular detection includes: at least one fluid channel configured to receive molecules to be detected, wherein at least some of the molecules to be detected are labeled with MNPs; a sensor including an STO and encapsulated by a material separating the sensor from the at least one fluid channel, wherein a surface of the material provides binding sites for the molecules to be detected; and a detection circuit coupled to the sensor and configured to detect a frequency or frequency noise of an RF signal generated by the STO in response to the presence or absence of at least one MNP coupled to one or more binding sites associated with the sensor.

[0012] In some embodiments, the detection circuit is configured to detect the frequency or frequency noise of an RF signal generated by the STO in response to the presence or absence of at least one MNP coupled to one or more binding sites associated with the sensor, in part by applying a DC current to the STO.

[0013] In some embodiments, the molecules to be detected include a first type of molecules and a second type of molecules, the first type of molecules are labeled by a first MNP type, and the second type of molecules are labeled by a second MNP type, the frequency or frequency noise of the RF signal generated by the STO is (a) a first frequency or frequency noise in response to the presence of the first MNP type, or (b) a second frequency or frequency noise in response to the presence of the second MNP type, and the detection circuit is configured to distinguish the first frequency or frequency noise from the second frequency or frequency noise to distinguish the first type of molecules from the second type of molecules.

[0014] In some embodiments, the detection circuit includes a delay line circuit. If present, the delay line circuit may include a power divider configured to split the signal generated by the STO or an amplified version of the signal generated by the STO into a first signal directed to a first path (e.g., a first conductor trace) having a first delay and a second signal directed to a second path (e.g., a second conductor trace) having a second delay, wherein the second delay is greater than the first delay (e.g., if the first path and the second path are conductor traces, the second conductor trace is longer than the first conductor trace); a mixer having a first input coupled to the first path, a second input coupled to the second path, and an output; and a low pass or band pass filter coupled to the output of the mixer.

[0015] In an embodiment including a delay line circuit, the delay line circuit may also include (a) a first amplifier coupled to the power divider and disposed between the STO and the power divider to provide an amplified version of the signal generated by the STO to the power divider, and / or (b) a second amplifier coupled to the output of the low pass or band pass filter.

[0016] In an embodiment including a delay line circuit, the delay line circuit may further include a phase shifter disposed on the first path or the second path between the power divider and the mixer. If present, the phase shifter is configured to adjust the difference between the phase of the first signal and the phase of the second signal such that in the absence of at least one MNP coupled to one or more binding sites associated with the sensor, the DC output of the line delay circuit is below a threshold.

[0017] In some embodiments, the apparatus further comprises a spectrum analyzer coupled to the output of the low pass or band pass filter.If present, the spectrum analyzer may be implemented via non-transitory machine executable instructions for execution by the processor.

[0018] In some embodiments, a system includes an apparatus for molecular detection, the apparatus including a detection circuit having a delay line circuit and a spectrum analyzer coupled to an output of the detection circuit of the apparatus.

[0019] In some embodiments including a delay line circuit, the MNPs used are superparamagnetic, and the delay line circuit includes a spectrum analyzer configured to detect frequency noise of the RF signal generated by the STO based on a comparison of the measured integrated noise from the STO with noise measurements from a reference STO that has not been exposed to any MNPs. If present, the spectrum analyzer may be implemented by the processor.

[0020] In some embodiments, the detection circuit includes a phase-locked loop (PLL) configured to provide an error signal output corresponding to the frequency noise of the RF signal generated by the STO in response to the presence of at least one MNP coupled to one or more binding sites associated with the sensor. In embodiments where the detection circuit includes a PLL, the PLL may include a loop filter and a low-pass filter. If present, the loop filter may include an amplifier and at least one resistor.

[0021] In some embodiments where the detection circuit includes a PLL, the detection circuit also includes: a mixer having a first input, a second input, and an output; and a reference oscillator having an input coupled to a tuning input from the PLL and an output coupled to the first input of the mixer. In some such embodiments, the STO is coupled to the second input of the mixer, and the output of the mixer is coupled to the input of the low pass filter.

[0022] In some embodiments, the STO includes a pinned layer, a free layer, and an interlayer disposed between the pinned layer and the free layer. The pinned layer may include one or more ferromagnetic (FM) layers. In some embodiments, the one or more FM layers are the first one or more FM layers, and the free layer includes the second one or more FM layers. The interlayer may include an insulating layer or a metal layer. In some embodiments, in a stationary state of magnetization (when the STO is not oscillating), the magnetic moment of the free layer is oriented to be substantially collinear with the magnetic moment of the pinned layer. In some embodiments, in a stationary state of magnetization, the magnetic moment of the free layer is oriented to be substantially parallel to or antiparallel to the magnetic moment of the pinned layer. In some embodiments, in a stationary state of magnetization, the magnetic moment of the free layer is oriented to be at an angle to the magnetic moment of the pinned layer, wherein the angle is between about 20 degrees and about 60 degrees.

[0023] The present invention also discloses a method for performing nucleic acid sequencing using an apparatus, the apparatus comprising a plurality of STOs and at least one fluid channel, the method comprising: labeling nucleotide precursors with MNPs; adding the labeled nucleotide precursors to at least one fluid channel of the apparatus; detecting the frequency or frequency noise of an RF signal generated by at least one of the plurality of STOs; and determining whether the labeled nucleotide precursor has been detected based at least in part on the detected frequency or frequency noise of the RF signal generated by at least one of the plurality of STOs.

[0024] In some embodiments, detecting the frequency or frequency noise of the RF signal generated by at least one of the plurality of STOs comprises: detecting the amplitude of a DC signal at an output of a delay line circuit coupled to at least one of the plurality of STOs. In some embodiments, detecting the frequency or frequency noise of the RF signal generated by at least one of the plurality of STOs comprises: monitoring an error signal of a detection circuit comprising a phase locked loop.

[0025] In some embodiments, the MNPs are superparamagnetic, and detecting the frequency or frequency noise of an RF signal generated by at least one of a plurality of STOs includes determining a spectral density of the RF signal generated by at least one of a plurality of STOs, integrating the spectral density, and comparing the integrated spectral density with a reference noise associated with a reference STO that is not affected by any MNPs.

[0026] In some embodiments, the method further comprises: binding at least one nucleic acid strand to a binding site in the fluid channel, and adding an extendable primer and a plurality of nucleic acid polymerase molecules to the fluid channel prior to adding the labeled nucleotide precursors to the fluid channel of the device.

[0027] In some embodiments, the method further comprises: in response to determining that the labeled nucleotide precursor has been detected, recording (a) the identity of the nucleotide precursor, or (b) the identity of the base that is complementary to the labeled nucleotide precursor.

[0028] The present invention also describes a method for performing nucleic acid sequencing using a device comprising a plurality of STOs and at least one fluid channel, the method comprising: labeling a first nucleotide precursor with a first MNP type, the first MNP type being selected so that the magnetization of each of the plurality of STOs oscillates at a first frequency; labeling a second nucleotide precursor with a second MNP type, the second MNP type being selected so that the magnetization of each of the plurality of STOs oscillates at a second frequency; adding the labeled first nucleotide precursor and the labeled second nucleotide precursor to the fluid channel of the device; detecting the frequency of a signal generated by at least one of the plurality of STOs using a delay line circuit; and identifying whether the first nucleotide precursor or the second nucleotide precursor has been detected in response to the detected frequency.

[0029] In some such embodiments, detecting the frequency of a signal generated by at least one of a plurality of STOs includes: splitting a signal originating from at least one of a plurality of STOs into a first signal and a second signal; directing the first signal to a mixer via a first path (e.g., a first conductor trace) having a first delay; directing the second signal to a mixer via a second path (e.g., a second conductor trace) having a second delay, the second delay being longer than the first delay; the mixer mixing the delayed first signal and the delayed second signal; and a low-pass filter filtering the output from the mixer. In some such embodiments, identifying whether a first nucleotide precursor or a second nucleotide precursor has been detected includes: analyzing the output of a low-pass filter or an amplified version of the output of a low-pass filter. In some such embodiments, detecting the frequency of a signal generated by at least one of a plurality of STOs includes: shifting the phase of the first signal or the second signal. In some embodiments, detecting the frequency of a signal generated by at least one of a plurality of STOs includes: amplifying the output of a low-pass filter.

[0030] In some embodiments, analyzing the output of the low pass filter or an amplified version of the output of the low pass filter includes: accessing a lookup table to determine whether the output of the low pass filter or an amplified version of the output of the low pass filter corresponds to a first expected value for a first nucleotide precursor or a second expected value for a second nucleotide precursor; or determining whether the output of the low pass filter or an amplified version of the output of the low pass filter is within a first range associated with the first nucleotide precursor or a second range associated with the second nucleotide precursor.

[0031] The present invention also discloses a system for performing nucleic acid sequencing, the system comprising: a plurality of STOs; at least one fluid channel; a device for labeling nucleotide precursors with MNPs; a device for adding the labeled nucleotide precursors to at least one fluid channel; a device for detecting the frequency or frequency noise of an RF signal generated by at least one of the plurality of STOs; and a device for determining whether the labeled nucleotide precursor has been detected based at least in part on the detected frequency or frequency noise of the RF signal generated by at least one of the plurality of STOs.

[0032] In some embodiments, the system further comprises a device for binding at least one nucleic acid strand to a binding site in the fluid channel; and a device for adding an extendable primer and a plurality of nucleic acid polymerase molecules to the fluid channel. The system may also comprise a device for recording (a) the identity of the nucleotide precursor, (b) the identity of the base complementary to the labeled nucleotide precursor, or (c) both (a) and (b). BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order that the above-mentioned features of the present disclosure can be understood in detail, the present disclosure is described in more detail with reference to embodiments, some of which are shown in the accompanying drawings. However, it should be noted that the accompanying drawings only illustrate typical embodiments of the present disclosure and are therefore not to be considered as limiting its scope, as the present disclosure may allow for other equally effective embodiments.

[0034] Figure 1A A portion of a sensor is shown according to some embodiments.

[0035] Figure 1B An exemplary sensor that can utilize spin torque oscillations to sense local magnetic fields caused by magnetic particles is shown in accordance with some embodiments.

[0036] Figure 1C An exploded schematic diagram of a reference layer and a free layer of a sensor according to some embodiments is shown.

[0037] Figure 2A , Figure 2B and Figure 2C Shown is how electrons in an electric current interact with a thin film ferromagnetic layer according to some embodiments.

[0038] Figure 3A , Figure 3B and Figure 3C The operating principle of an STO-based sensor according to some embodiments is shown.

[0039] Figure 4A , Figure 4B and Figure 4C An apparatus for molecular detection according to some embodiments is shown.

[0040] Figure 5A , Figure 5B , Figure 5C and Figure 5D Portions of another exemplary device for molecular detection are shown according to some embodiments.

[0041] Figure 5E A sensor selection method according to some embodiments is shown.

[0042] Fig. 5F Another sensor selection method according to some embodiments is shown.

[0043] Fig. 6A , Figure 6B and Figure 6C A cross-point array architecture of sensor elements according to some embodiments is shown.

[0044] Figure 7 Detection circuits according to some embodiments are shown.

[0045] Figure 8 Detection circuits according to some embodiments are shown.

[0046] Fig. 9 Detection circuits according to some embodiments are shown.

[0047] Fig.10 An exemplary sequential binary method suitable for DNA sequencing is shown, according to some embodiments, where a single MNP type is used to label multiple nucleotide precursors.

[0048] Fig.11 is a flow chart illustrating a method suitable for DNA sequencing using multiple nucleotide precursors labeled by different MNP types, according to some embodiments.

[0049] Fig.12 is a flow chart illustrating another method suitable for DNA sequencing using multiple nucleotide precursors labeled by different MNP types, according to some embodiments.

[0050] Fig.13 Methods suitable for DNA sequencing using MNP-labeled nucleotide precursors and a tunable reference oscillator are shown according to some embodiments.

[0051] Fig.14A A method of making a detection device according to some embodiments is shown.

[0052] Fig. 14B According to some embodiments, Fig.14A The results of the steps of the manufacturing method.

[0053] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. It is contemplated that elements disclosed in one embodiment may be beneficially utilized on other embodiments without specific recitation. DETAILED DESCRIPTION

[0054] Improved detection devices, systems and methods for identifying molecules (e.g., nucleic acids) using magnetic nanoparticles (MNPs) are disclosed herein. The disclosure herein includes embodiments of sensors having spin torque oscillators (STOs) that allow for the detection of characteristics that indicate the presence or absence of MNPs near the sensor. Detection methods are also disclosed herein that can be used to detect (e.g., measure or obtain) characteristics or characteristic changes of radio frequency (RF) signals generated by the sensor, wherein the detected characteristics or characteristic changes indicate the presence or absence of MNPs (e.g., in response to a magnetic field or the absence of a magnetic field generated by a magnetic nanoparticle tag). For example, the devices and methods can determine whether the sensor is generating a signal having a frequency at a specific frequency or within a specified frequency range, and based on this, determine whether the sensor detects one or more MNPs. For another example, the devices and methods can detect the frequency / phase noise of the signal generated by the sensor, and based on this, determine whether the sensor detects one or more MNPs and / or the type of MNPs detected.

[0055] In some embodiments, the MNP is coupled to a molecule to be detected. For example, in a DNA sequencing application, the MNP can label a nucleotide precursor, which is then incorporated into a target DNA strand attached to a binding site near the sensor. As a result of incorporating the MNP-labeled nucleotide precursor, at least one MNP is near the sensor, and its presence can have an effect on the behavior of the STO. This effect can be detected, thereby detecting the presence of the MNP. The presence of the MNP can then be used to determine that a specific nucleotide precursor has been incorporated into the target DNA strand.

[0056] As used herein, the term "spin torque oscillator" and the acronym "STO" refer to any device that utilizes the precession of magnetization due to spin torque induced by a spin-polarized current.

[0057] The terms "frequency noise", "phase noise" and "frequency / phase noise" are used interchangeably herein. These terms are used to refer to the random fluctuations in the instantaneous frequency of an oscillating signal.

[0058] In the following description, reference is made to embodiments of the present disclosure. However, it should be understood that the present disclosure is not limited to the embodiments specifically described. On the contrary, any combination of the following features and elements (whether or not related to different embodiments) is considered to realize and practice the present disclosure. In addition, although the embodiments of the present disclosure can achieve advantages that are superior to other possible solutions and / or superior to the prior art, whether a specific advantage is achieved by a given embodiment is not a limitation to the present disclosure. Therefore, the following aspects, features, embodiments and advantages are merely illustrative and are not considered to be elements or limitations of the appended claims unless explicitly stated in a claim (or in multiple claims). Similarly, reference to "the present disclosure" should not be interpreted as a summary of any inventive subject matter disclosed herein, and should not be considered to be elements or limitations of the appended claims unless explicitly stated in a claim (or in multiple claims).

[0059] It should be understood that, at the outset, the disclosure herein can be used to detect any type of molecule to which a magnetic particle can be attached or coupled. The present invention assumes that the particles attached to the molecules to be detected are magnetic nanoparticles, but this assumption is exemplary and is not intended to be limiting. Therefore, the term "magnetic nanoparticle" and the acronym "MNP" include all types of magnetic particles that can be attached or coupled to molecules to be detected.

[0060] Therefore, the apparatus and methods disclosed herein can be used to detect any molecule type that can be labeled by magnetic nanoparticles. Such molecule types can be biomolecule types, such as proteins, antibodies, etc. For example, the disclosure herein can be used to detect nucleic acids (for example, in DNA sequencing). The disclosure herein can also be used to detect non-biological (inorganic or inactive) molecules, such as pollutants, minerals, compounds, etc. The presentation of the part of the disclosure in the context of nucleic acid sequencing is merely exemplary and is not intended to limit the scope of the present invention. Therefore, although some of the disclosure herein is provided in the context of nucleic acid sequencing, and specifically DNA sequencing, it should be understood that the embodiments herein can be generally used to detect any type of molecule to which magnetic nanoparticles can be attached.

[0061] In addition, although the description herein focuses on DNA as an exemplary nucleic acid, the various embodiments described are generally applicable to nucleic acid sequencing. Similarly, although SBS is used for exemplary purposes in the following description, the various embodiments are not limited to SBS sequencing schemes (e.g., dynamic sequencing may be used alternatively).

[0062] Conventional nucleic acid sequencing, such as that used for DNA sequencing, generally relies on the detection of fluorescence. Specifically, fluorescence-based techniques (e.g., in fluorescence-based nucleic acid sequencing techniques) for distinguishing different bases in a sample rely on, for example, the quality of the signal generated by a detection moiety associated with a particular type of nucleotide. For example, conventional fluorescence sequencing techniques utilize identifiably different fluorescent moieties that are each attached to one of the four nucleotides A, T, C, and G utilized in the sequencing reaction.

[0063] A conventional DNA sequencing method involves adapting single-stranded DNA (ssDNA) to attach to a solid support of a sequencing device and using a technique such as the polymerase chain reaction to amplify the amount of ssDNA to produce many DNA molecules with a short leader sequence. Oligonucleotides complementary to the short leader sequence can then be added so that a short segment of double-stranded DNA (dsDNA) is present at the leader sequence. The double-stranded portion of the bound molecule is a primer for a suitable DNA polymerase (such as Taq polymerase) that is effective at high temperatures.

[0064] Sequencing can then be performed using one of several methods. For example, sequencing can use a mixture of four fluorescently labeled 3'-blocking dNTPs (fluorescently labeled dideoxynucleotide terminators), where the fluorescent label is part of the 3'-blocking group. The fluorescent label is used as a "reversible terminator" for the polymerization reaction. Each of the NTPs is labeled with a different label (i.e., each of the A, G, C, and T nucleotides has a different fluorescent label), and the different labels can be distinguished by fluorescence spectroscopy or by other optical means.

[0065] Four fluorescently labeled nucleotide precursors can be used to sequence millions of clusters of parallel DNA chains. DNA polymerase catalyzes the incorporation of fluorescently labeled dNTPs into DNA template chains during the sequential cycles of DNA synthesis. In each sequencing cycle, the bound double-stranded DNA molecules are exposed to a mixture of DNA polymerase and four fluorescently labeled 3'-blocking NTPs. The polymerase adds one of the four dNTPs to the growing oligonucleotide chain (regardless of which dNTP is complementary to the next unpaired base in ssDNA). Then, by washing at a temperature that prevents free dNTPs from binding to ssDNA but not high enough to dehybridize dsDNA, unincorporated dNTPs and other impurities left unreacted or generated during the reaction are separated from the vicinity of the DNA bound to the support.

[0066] Because only one of the four types of dNTPs will be added to the oligonucleotide, and the four fluorescent labels are distinguishable, the identity of the incorporated dNTP can be identified by laser excitation and imaging. Specifically, each of the four filters is used to determine whether light of a specific wavelength (e.g., color) is emitted. The fluorescent label can then be enzymatically cleaved to allow the next round of incorporation. Because each base type can be paired with only one other base type, the identity of the base that just pairs in the unknown ssDNA sequence can be known from the identity of the incorporated dNTP (which is known from the wavelength of the emitted light). Therefore, the base is identified directly from the fluorescence measurement during each cycle.

[0067] One disadvantage of the above methods is that complex optical systems are required to filter out light of different wavelengths to detect the fluorescent labels of the incorporated dNTPs and to distinguish between different emission colors (wavelengths). Other methods have been developed to simplify the optical systems, but their sequencing is slower and requires intermediate chemical steps within each sequencing cycle. Therefore, these methods have been introduced into smaller, cheaper entry-level sequencing systems, but have not been introduced into more advanced systems that require fast throughput.

[0068] As previously described, the disclosure herein can be used to detect any type of molecule (e.g., biomolecules, organic molecules, inorganic molecules, or inactive molecules) to which magnetic particles (e.g., MNPs) can be attached. The devices and methods disclosed herein use MNPs and sensors to perform detection of molecules, such as in nucleic acid sequencing (e.g., DNA sequencing using SBS chemistry). Specifically, embodiments of the present disclosure include sensors with STOs that can be used to detect magnetic fields (or changes in magnetic fields) emitted by MNPs, and specifically for distinguishing the presence and absence of magnetic fields emitted or not emitted by MNPs near the sensor. Embodiments using the same MNP type for all molecules to be detected are disclosed, as well as embodiments using multiple MNP types, each type labeling a different type of molecule. Embodiments disclosed herein allow for different types of molecules to be distinguished.

[0069] Embodiments of the present disclosure also include various detection methods to obtain or determine (e.g., measure) a characteristic of a sensor or output from a sensor (e.g., a change in oscillation frequency and / or frequency noise) caused by an MNP used as a tag near the sensor. Knowledge of the specific type of molecule (e.g., in a DNA sequencing application, the type of nucleotide precursor) to which a specific MNP tag is attached can then be used to identify the specific molecule type (e.g., in a DNA sequencing application, the last paired base of a ssDNA strand that is complementary to the identified nucleotide precursor).

[0070] STO Sensor

[0071] In some embodiments disclosed herein, a spin torque oscillation magnetoresistive sensor is provided to sense the magnetic field caused by MNPs coupled to the molecules being detected. The sensor is configured to detect the precession oscillation frequency or the noise of the precession oscillation frequency of the magnetization of the magnetic layer to sense the magnetic field of the MNPs. The sensor may include a magnetic free layer, a magnetic pinned layer, and a non-magnetic layer located between the free layer and the pinned layer. In operation, a detection circuit coupled to these layers induces an electrical (DC) current through these layers. The spin polarization of electrons traveling through the sensor causes precession due to the spin torque of the magnetization of one or more layers. The frequency and / or noise of the oscillation changes in response to the magnetic field generated by the MNPs near the sensor.

[0072] In some embodiments, understanding how the presence of a particular type of MNP is expected to change the sensor's oscillation frequency or the noise in the oscillation frequency (frequency noise) can be used to detect the presence or absence of a magnetic field, and thus the presence or absence of MNPs. In some embodiments, the expected effect of a particular type of MNP on the sensor's oscillation frequency is known. For example, the presence of a particular type of MNP can cause the sensor to oscillate at a frequency f1, where f1 is different from the frequency f0 at which the sensor oscillates in the absence of the particular type of MNP, and the measured frequency can be used to detect the presence or absence of a particular type of MNP near the sensor. Specifically, if the measured oscillation frequency is f1, it can be inferred that there is at least one MNP near the sensor, and if the measured oscillation frequency is f0, it can be inferred that there are no MNPs near the sensor. As another example, a particular type of MNP can cause noise in the oscillation frequency (called phase noise or frequency noise), and detecting the presence or increase of this noise can allow the presence of MNPs to be detected. Similarly, the absence of frequency noise in the sensor oscillation or the frequency noise being below a threshold can be used to determine that the sensor is not affected by any MNP.

[0073] Figure 1A A three-layer structure of sensor 105 is shown according to some embodiments. Figure 1A The exemplary sensor 105 has a bottom 108 and a top 109. The sensor 105 includes an STO, which is a patterned magnetic device having an active area including three layers, Figure 1A Shown in FIG. 1 are two ferromagnetic (FM) layers 106A, 106B separated by a nonmagnetic spacer layer 107.

[0074] In some embodiments where the STO is a thin film device, the FM layers 106A, 106B are designed so that their magnetic moments are oriented substantially in the plane of the film or substantially perpendicular to the plane of the film. Suitable materials for the FM layers 106A, 106B include, for example, alloys of Co, Ni, and Fe (sometimes mixed with other elements). The above exemplary materials are merely exemplary and are not intended to be limiting. Suitable materials for the FM layers 106A, 106B are known to those of ordinary skill in the art.

[0075] The non-magnetic spacer 107 may be, for example, a metal material or a combination of metal materials, such as, for example, copper or silver, in which case the structure is referred to as a spin valve (SV). Alternatively, the non-magnetic spacer 107 may be an insulator material, such as, for example, aluminum oxide (also referred to as alumina in the art) or magnesium oxide, in which case the structure is referred to as a magnetic tunnel junction (MTJ). Distinguishing between materials for insulator materials is merely exemplary and is not intended to be limiting. Suitable materials for the non-magnetic layer 107 are known to those of ordinary skill in the art.

[0076] The active area of ​​sensor 105 is located at Figure 1A In the three-layer structure shown in Figure 1B As further discussed in the discussion of Figure 1A Additional layers are added above and below the layers 106A, 106B, 107 shown for various purposes, such as, for example, interface smoothing, texturing, and avoiding exposure to processing for patterning the entire detection device (e.g., as described below in, for example, FIG. 4A to FIG. 4C , FIG. 5A to FIG. 5D ), and passivation / protection of sensor 105. Thus, a component in contact with magnetic sensor 105 may be in contact with one of the three illustrated layers 106A, 106B, or 107, or it may be in contact with the sensor 105. Figure 1A Another part not shown in the figure.

[0077] As described further below, by applying an electric current to the device, the magnetic moment of one or both of the FM layers 106A, 106B of the sensor 105 can be excited into a precessing orbit through an effect known as spin transfer. Spin transfer (sometimes referred to as spin torque transfer) involves the interaction of a spin-polarized current (i.e., a current in which a greater portion of electrons have spins oriented in the same direction) with an FM layer (e.g., layer 106A or 106B).

[0078] Figure 1B is a diagram of another exemplary sensor 105 that can utilize spin torque oscillations to sense local magnetic fields caused by magnetic particles (eg, MNPs). Figure 1BA cross-sectional view of the sensor 105 is shown, wherein the sensed MNP is shown to be located on the right side of the sensor 105 .

[0079] Figure 1B The exemplary sensor 105 of includes a sensor stack 304 sandwiched between an optional first magnetic shield 306A and a second magnetic shield 306B. If magnetic shields 306A, 306B are present, they can be made of a conductive magnetic material such as NiFe so that they can serve as electrical leads as well as magnetic shields. The sensor stack 304 includes a pinned layer structure 311, a free layer 310, and a non-magnetic spacer 107 sandwiched between the free layer 310 and the pinned layer structure 311. As described above in Figure 1A As explained in the context of , the nonmagnetic spacer 107 may be a nonmagnetic conductive spacer, or it may be a thin nonmagnetic electrically insulating barrier layer. A capping layer 328 (eg, comprising tantalum) may be positioned adjacent to the free layer 310, such as Figure 1B It should be understood that Figure 1B The sensor 105 is shown with an exemplary orientation of layers (e.g., the pinned layer structure 311 is above the free layer 310), but other orientations are possible (e.g., the pinned layer structure 311 can be below the free layer 310, the sensor 105 can be relative to Figure 1B Rotate in the manner shown and can be omitted Figure 1B Some of the components shown (e.g., shields 306A, 306B), etc.).

[0080] The pinned layer structure 311 may include a magnetic pinned layer 314, a reference layer 319, and a non-magnetic antiparallel coupling layer 321 sandwiched between the pinned layer 314 and the reference layer 319. The pinned layer 314 and the reference layer 319 may include a material such as, for example, CoFe, and the antiparallel coupling layer 321 may include a material such as, for example, Ru having a thickness of, for example, about 10 angstroms. The pinned layer 314 may be exchange coupled with an antiferromagnetic material layer (AFM layer 312), which may include a material such as, for example, IrMn, PtMn, or some other suitable antiferromagnetic material as known in the art. The exchange coupling between the AFM layer 312 and the pinned layer 314 strongly pins the magnetization 324 of the pinned layer 314 in a first direction as shown. The strong antiparallel coupling between the pinned layer 314 and the reference layer 319 pins the magnetization 326 of the reference layer 319 in a second (antiparallel) direction as shown.

[0081] exist Figure 1B In the exemplary embodiment shown, the free layer 310 has its magnetization 330 biased in a direction substantially antiparallel to the magnetization 326 of the reference layer 319. In some embodiments, in a static state of magnetization (e.g., when the STO is not oscillating), the magnetization 330 of the free layer 310 is at a moderate angle relative to the magnetization 326 of the reference layer 319. This can be referred to as Figure 1C As can be seen, the figure shows an exploded schematic diagram of a reference layer 319 and a free layer 310 shown offset. As shown, the reference layer 319 has a magnetization 326 that is pinned in a direction parallel to (but alternatively antiparallel to) an applied magnetic field 327, but the free layer 310 has a magnetization 330 that is biased in a direction that is nearly antiparallel to the direction of the reference layer magnetization 326, but offset by an angle 329. Angle 329, if present, is typically about 20-60 degrees, but can be as large as almost 90 degrees. The biasing of the free layer 310 can be determined by Figure 1B Not shown but will enter and leave Figure 1B When the free layer 310 is magnetically biased, the magnetization 330 of the free layer 310 moves freely with the precession spin torque oscillation 337, as shown in FIG. Figure 1B As shown and discussed previously.

[0082] Reference again Figure 1C , the tilt of the direction of magnetization 330 of the free layer 310 relative to the direction of magnetization 326 of the reference layer 319 can be provided by a magnetic anisotropy having a component oriented perpendicular to the direction of magnetization 326 of the reference layer 319 and / or perpendicular to the direction of the applied magnetic field 327. The magnetic anisotropy can be generated by a layer of antiferromagnetic material that is weakly exchange coupled to the free layer 310, or by shape anisotropy, or by texture-induced magnetic anisotropy. The tilt of the free layer 310 can also be achieved by placing a high-coercivity magnetic material near the free layer 310 and utilizing a magnetization having a significant component perpendicular to the reference layer 319, similar to the hard bias structure that can be used in a recording head to stabilize the free layer of GMR and TMR readback sensors. However, these are given by way of example only; other mechanisms can also be used.

[0083] As described in further detail below, spin torque effects are observed when a high current density of spin-polarized electrons generated by one magnetized layer impinges on a second magnetized layer, and these spin torque effects dynamically excite the magnetization of the second layer through a mechanism known as spin transfer. Here, electrons traveling through a ferromagnet tend to align their spins parallel to the magnetization of the ferromagnet, thereby losing any component of spin angular momentum that is transverse to the magnetization. In order to conserve angular momentum, the polarizing current must then exert a torque on the magnetization.

[0084] FIG. 2A to FIG. 2C This shows in more detail how the electrons in the current interact with the thin film FM layer. Quantum mechanics shows that the electrons interacting with the FM layer will cause the electron spins to be oriented preferentially parallel or antiparallel to the moment of the FM layer to obtain transmitted and reflected electrons, respectively. Figure 2AAs shown, electrons having spin 210 parallel to the moment 206 of the FM layer 204 preferentially pass through the FM layer 204, while electrons having spin 208 antiparallel to the moment 206 of the FM layer 204 are preferentially reflected back. Due to this phenomenon, the interface between the nonmagnetic (NM) layer 202 (for the purpose of this explanation, it is assumed to be a metal layer, but as described above, the NM layer 202 may alternatively be an insulator) and the FM layer 204 acts as a spin filter that can be used to spin-polarize (i.e., make one spin direction more favorable) the incoming current.

[0085] For a device having two FM layers 224 and 228 separated by a nonmagnetic metal layer 226 (spacer), such as Figure 2B and 2C As shown, the incoming current spins polarized by the first FM layer (FM1) 224 interact differently with the second FM layer (FM2) 228, depending on the orientation of the magnetic moment of the second FM layer 228. If the moments of both FM layers 224 and 228 are parallel to each other ( Figure 2B ), many electrons will pass through the device because many of the electrons in the current will be spin-oriented with the moment of the second FM layer 228 (spin 234). Few electrons will be reflected back (spin 232).

[0086] On the other hand, when the moments of the two FM layers 224 and 228 are oriented in an antiparallel manner ( Figure 2C ), many electrons will be blocked from passing through the second FM layer 228 (spin 236), and far fewer electrons will traverse the device (spin 238). This means that the amount of current passing through the device depends on the orientation of the moments of the two FM layers 224 and 228 relative to each other. Since the resistance of a device including FM layers 224 and 228 and NM layer 226 is inversely proportional to the current, the resistance of the device depends on the orientation of the moments of the two FM layers 224 and 228 (i.e., the resistance when the moments are parallel is less than the resistance when they are antiparallel).

[0087] Although the above description assumes the use of a nonmagnetic metal layer 226 separating the two FM layers 224 and 228 (also referred to as a spin valve (SV) or giant magnetoresistance (GMR) device configuration), an insulating layer known as a tunneling barrier layer may alternatively be used as a spacer separating the FM layers 224, 228 (e.g., in place of the NM layer 226). In such embodiments, the spacer layer may be made of an oxide-based material. These types of devices are known as magnetic tunnel junctions (MTJs), and they exhibit a similar resistance response (known as tunnel magnetoresistance or TMR) due to spin-polarized tunneling rather than spin filtering.

[0088] FIG. 3A to FIG. 3C The basic working principle of the STO based sensor 105 is further illustrated. Figure 3A906 or is reflected by the FM layer 906. As shown, those incident electrons 904 whose spins are parallel to the magnetic moment of the FM layer 906 are transmitted electrons 908, while those incident electrons 904 whose spins are antiparallel to the magnetic moment of the FM layer 906 are reflected electrons 902. Any loss of spin angular momentum becomes a torque acting on the FM layer 906. The torque from the single electron interaction is small, but for spin-polarized currents of about a few milliamps, there is about 10 15 electrons interact with the FM layer 906. Therefore, the net torque on the FM layer 906 may be sufficient to induce the moment to enter dynamic modes. These dynamics are determined by the Landau-Lifshitz-Gilbert-Slonczewski (LLGS) formula:

[0089]

[0090] where γ is the magnetogyro ratio, is the normalized moment vector, is the effective magnetic field acting on the FM layer 906, α is the phenomenal Gilbert damping parameter, η is the spin polarization of the current I, and is the direction of the spin polarization of the current.

[0091] The first term in this equation, called the Larmor precession term, indicates that in the absence of any damping, the moments of the FM layer 906 will precess about the effective magnetic field acting on the FM layer 906. However, the second term (Gilbert damping) comes from the intrinsic damping that occurs in every ferromagnet, which serves to dampen any dynamics of the moments. The last term is the Slonczewski spin torque term, which acts like a damping term or an anti-damping term, depending on the polarity of the applied current. In the anti-damped case, the spin torque will completely cancel the Gilbert damping at sufficient current amplitude and will cause the magnetization to oscillate, as shown in Figure 3C As the current amplitude increases further, the oscillation amplitude also increases, eventually leading to a crossover point where the moment is 90 degrees from equilibrium. In this region, the cross product in the Slonczewski term changes sign and acts to dampen the motion, causing the moment to rotate 180 degrees from the original position, as shown in Figure 3B shown.

[0092] Thus, considering a complete STO device similar to that described above, where one FM layer 906 is excited by the spin transfer effect, and the second FM layer 906 has a fixed moment in some chosen direction (they are collinear), as Figure 3CThe excited STO shown will generate a radio frequency (RF) voltage signal from the applied DC current due to resistance fluctuations (and hence voltage and current fluctuations) caused by the magnetoresistance effect. The frequency of the generated RF signal may be on the order of GHz.

[0093] Reference again Figure 1B , when electrons flow from the reference layer 319 through the nonmagnetic spacer layer 107 to the free layer 310, the spins of the electrons flowing through the reference layer 319 are polarized by the magnetization 326 of the reference layer 319. These polarized electrons can then apply torque to the free layer magnetization 330, thereby generating spin waves that lead to disordered magnetization dynamics (noise) or collective excitations (oscillations), depending on various parameters of the system, such as the sensor 105 shape, anisotropy, layer materials and thicknesses, and the applied current and magnetic field.

[0094] As mentioned above, spin torque oscillations involve the precession of the magnetization along the equilibrium axis of the ferromagnet by the spin torque excitation. Figure 1B , the precession or oscillation of the magnetization 330 is indicated by oscillation 337. Note that although the magnetization 324 of the pinned layer 314 is constrained by the exchange anisotropy with the antiferromagnetic layer 312, the magnetization of the pinned layer 314 can also oscillate and contribute to the sensor 105 signal when the applied current density is high enough to produce a spin torque excitation in the pinned layer 314.

[0095] The frequency (oscillation frequency) of this precession shifts with the application of a magnetic field. In the case of a suitable choice of sensor materials and geometry, this shift may be very large. Frequency shifts up to 180 GHz / T have been demonstrated, and higher values ​​are possible. Some embodiments described herein utilize these frequency shifts to detect changes in the magnetic field at the free layer 310 caused by magnetic nanoparticles near the sensor 105.

[0096] See also Figure 1B, the sensor 105 is connected to the processing circuit 344 via the leads 341A, 341B. The leads 341A, 341B, which may be magnetic or non-magnetic, may be connected to the optional shield / lead layer 306A, 306B (if present) so that one lead 341A is connected to one lead / shield layer 306A and the other lead 341B is connected to the other lead / shield layer 306B. The processing circuit 344 sends a sense (bias) current through the sensor stack 304 and also measures the resistance across the sensor stack 304. As will be appreciated by those skilled in the art, the resistance across the non-magnetic spacer 107 changes as the orientation of the magnetization 330 of the free layer 310 changes relative to the magnetization 326 of the reference layer 319. As described above, the closer these magnetizations 330, 326 are to parallel, the lower the resistance will be. Conversely, the closer these magnetizations 330, 326 are to anti-parallel, the higher the resistance will be. The resistance of the device effectively acts as a magnetic field-to-voltage transducer.

[0097] The presence of MNPs near the sensor 105 causes the above-mentioned change in the frequency of the oscillation 337 of the magnetization 330. When the magnetization 330 oscillates, the frequency of the oscillation 337 can be measured by the processing circuit 344 by measuring the change in resistance across the sensor stack 304. In addition or alternatively, the change in the frequency of the oscillation 337 can be detected to determine whether the MNP is near the sensor 105. In addition or alternatively, the noisiness of the oscillation frequency can be detected and / or monitored. It is expected that in the presence of one or more SNPs, the frequency noise is greater than in the absence of any SNPs. Therefore, according to some embodiments disclosed herein, spin torque oscillations are used to detect the presence or absence of magnetic nanoparticles near the sensor 105.

[0098] Using the working principle of STO for detection, some embodiments disclosed herein relate to an array of sensors 105 including STO devices, such as Figure 1A , Figure 1B and Figure 1C The sensor 105 shown and e.g. FIG. 4A to FIG. 4C , FIG. 5A to FIG. 5D The array shown, etc. Each sensor 105 in the sensor array can be used to detect magnetic particles (e.g., MNPs) in a fluid channel of a detection device. Each sensor 105 can have a size of less than about 30 nm to detect magnetic fields of the order of tens of Oersteds (Oe). In some embodiments, each sensor 105 in the sensor array is configured to generate an RF signal only within a narrow band magnetic field (e.g., about zero applied field, although this is not required), such as between 50 and -50 Oe.

[0099] In DNA sequencing applications, nucleotide precursors (or more generally, nucleic acids) labeled by MNPs and incorporated by a polymerase in the vicinity of the sensor 105 can be detected by evaluating the phase noise of the sensor 105. For example, in some embodiments, the sensor 105 generates an RF signal at or near a specific frequency in the absence of MNPs, and continues to generate an RF signal at or near a specific frequency in the presence of MNPs, but with a higher phase noise due to the presence of the MNPs. In some embodiments, the MNPs are superparamagnetic and have an effect on the phase noise of the RF signal generated by the sensor 105 in the absence of an applied (e.g., external) magnetic field. In other embodiments, the MNPs are not superparamagnetic, and an applied (e.g., external) magnetic field is applied to detect whether the MNPs are affecting the phase noise of the RF signal generated by the sensor. By monitoring the phase noise of the RF signal generated by the MNPs, the presence or absence of the MNPs can be detected (e.g., when the phase noise is below a threshold, it can be inferred that the MNPs are absent, and when the phase noise is above a threshold, it can be inferred that one or more MNPs are near the sensor and are the cause of the phase noise).

[0100] As discussed in further detail below, an advantage of performing detection using a sensor 105 that includes an STO device is that the MNPs used as labels can be superparamagnetic (e.g., thermally unstable, such that the generated magnetic field fluctuates over time) or ferromagnetic. In addition, the use of STOs does not require that the moments of the individual MNPs be aligned in the same direction (e.g., detection can be accomplished with or without the use of an external magnetic field). One benefit of superparamagnetic particles is that they are not ferromagnetic and, when introduced into a flow cell of a detection device (e.g., as described below in, e.g., FIG. 4A to FIG. 4C and FIG. 5A to FIG. 5D In addition, the use of superparamagnetic particles is attractive because the effect of superparamagnetic particles on the STO oscillation noise can be detected without the use of an applied external magnetic field.

[0101] Testing equipment

[0102] The above-described STO-based sensor 105 can be incorporated into a device for detecting molecules coupled to corresponding magnetic nanoparticles (eg, for nucleic acid sequencing). Figure 4A , Figure 4B and Figure 4C A detection device 100 is shown that can be used, for example, for nucleic acid sequencing, according to some embodiments. Figure 4A is a top view of the device, and Figure 4B yes Figure 4A A cross-sectional view at the position marked “4B” in FIG. Figure 4Cis a block diagram showing components of the detection device 100. Figure 4A As shown, the exemplary detection device 100 includes a sensor array 110, which includes a plurality of sensors 105, of which four sensors 105A, 105B, 105C, and 105D are shown. (For simplicity, this document generally refers to the sensor by the reference numeral 105. Each sensor is given by the reference numeral 105 followed by a letter.) Figure 4A The sensor array 110 shown in the exemplary embodiment of FIG. 1 is a linear array.

[0103] In some embodiments, each sensor in the plurality of sensors 105 is coupled to at least one flow channel 120 for reading characteristics of or output from one or more of the sensors 105 (e.g., determining whether the sensor 105 is oscillating at a particular frequency, detecting phase noise of an RF signal generated by the sensor 105, etc.). (For simplicity, this document generally refers to a flow channel by reference numeral 120. Individual flow channels are given reference numeral 120 followed by a letter.) Figure 4A In the exemplary embodiment shown, each sensor 105 of the sensor array 110 is coupled to two flow channels 120. Specifically, sensor 105A is coupled to flow channels 120A and 120E, sensor 105B is coupled to flow channels 120B and 120E, sensor 105C is coupled to flow channels 120C and 120E, and sensor 105D is coupled to flow channels 120D and 120E. Flow channels 120A, 120B, 120C, and 120D reside below sensors 105A, 105B, 105C, and 105D, respectively, and flow channel 120E resides above sensor 105. Figure 4B The relationship of sensor 105D to flow channels 120D and 120E is shown.

[0104] The detection device 100 also includes a fluid channel 115 (which may also be referred to as a nanochannel or a circulation cell) adjacent to the sensor array 110. As the name implies, the fluid channel 115 is configured to hold a fluid (e.g., liquid, gas, plasma) when the detection device 100 is in use. The fluid channel 115 may be open (e.g., if its shape is rectangular, it may have three sides; if its shape is curved, it may have a shape that is a portion of a cylinder; etc.) or closed (e.g., if its shape is rectangular, it may have four sides; if its shape is curved, it may be cylindrical; etc.). The shape of the fluid channel 115 may be regular or irregular along its length. The fluid channel 115 may be coupled to a device (e.g., a pump) that forces a fluid into the fluid channel 115. Alternatively, the fluid channel 115 may not be coupled to a device that injects or removes a fluid.

[0105] like Figure 4B As shown, the fluid channel 115 has a wall 117 adjacent to the sensor array 110. The wall 117 may also be referred to as a proximal wall. The wall 117 may be substantially vertical, such as Figure 4B Alternatively, the wall 117 can be at least partially inclined (e.g., some or all of the interior of the fluid channel 115 can be at an angle other than 90 degrees, or it can be curved (e.g., in the shape of part or all of a cylinder)). In general, the fluid channel 115 and the wall 117 can have any shape that allows the sensor 105 to detect the presence of magnetic particles located within the fluid channel 115 on the other side of the wall 117.

[0106] When the detection device 100 is in use, the sensor 105 is able to detect the presence or absence of MNPs in the fluid channel 115 through the wall 117. Therefore, the wall 117 has the properties and characteristics of protecting the sensor 105 from any fluid in the fluid channel 115 while still allowing the sensor 105 to detect the MNPs within the fluid channel 115. For example, the material of the wall 117 (and the material of the remaining portion of the fluid channel 115 that may be) can be an insulator material or include an insulator material. For example, in some embodiments, the surface of the wall 117 includes polypropylene, gold, glass and / or silicon. In addition, the thickness of the wall 117 can be selected so that the sensor 105 can detect the MNPs within the fluid channel 115. In some embodiments, the thickness of the wall 117 is between about 2 nm and about 20 nm.

[0107] In some embodiments, wall 117 has a structure (or structures) configured to anchor or bind a molecule to be sensed (e.g., a molecule of a nucleic acid or a nucleic acid polymerase) to wall 117. For example, the structure (or structures) of wall 117 may include one or more cavities / ridges that provide a binding site associated with sensor 105.

[0108] To simplify the explanation, Figure 4A and Figure 4B An exemplary detection device 100 is shown having a single fluid channel 115 and only four sensors, 105A, 105B, 105C, 105D, in the sensor array 110. It should be understood that the detection device 100 may have more sensors 105 in the sensor array 110, and it may have additional fluid channels 115 or a more complex single fluid channel 115 (e.g., having a different shape or having interconnected channels). In general, any configuration of sensors 105 and fluid channels 115 that allows the sensors 105 to detect MNPs in the fluid channels 115 may be used.

[0109] like Figure 4CAs shown, the detection device 100 includes a detection circuit 130 coupled to the sensor array 110 via a flow channel 120. In some embodiments, in operation, the detection circuit 130 applies a current to the flow channel 120 to detect a characteristic of at least one sensor in the plurality of sensors 105 in the sensor array 110 or outputs from the sensor, wherein the characteristic or output indicates the presence or absence of magnetically labeled molecules in the fluid channel 115. For example, in some embodiments, the characteristic or output is a signal or the absence of a signal. In other embodiments, the characteristic is an oscillation frequency or frequency noise of the signal. The detection circuit 130 may include any suitable components, typically including a suitable detection circuit. Such a detection circuit 130 may include hardware and / or software. The detection circuit 130 may include, for example, one or more of the following: a processor capable of executing machine executable instructions, an application specific integrated circuit (ASIC), a controller, a programmable circuit (e.g., FPGA), etc.

[0110] As an example of a detection device 100 having a larger number of sensors 105 in the sensor array 110, Figure 5A , Figure 5B , Figure 5C and Figure 5D A portion of an exemplary detection device 100 is shown that includes several channels, one or more of which may be separate fluid channels 115 according to some embodiments, or a collection of which may be considered a single fluid channel 115. Figure 5A , Figure 5B , Figure 5C and Figure 5D In the embodiment of the detection device 100 shown, the plurality of sensors 105 of the sensor array 110 are arranged in a rectangular grid pattern. Each flow channel 120 identifies a row or column of the sensor array 110. It should be understood that Figure 5A , Figure 5B , Figure 5C and Figure 5D Only a portion of the detection device 100 is shown to avoid obscuring the components of the detection device 100 being discussed. It should be understood that various illustrated components (e.g., flow channel 120, sensor 105, fluid channel 115, etc.) may not be visible in a physical instantiation of the detection device 100 (e.g., some or all of which may be covered by a protective material such as an insulator material).

[0111] Figure 5A1 is a perspective view of an exemplary detection device 100 according to some embodiments. The detection device 100 includes nine flow channels 120, respectively labeled 120A, 120B, 120C, 120D, 120E, 120F, 120G, 120H and 120I. It also includes five fluid channels, respectively labeled 115A, 115B, 115C, 115D and 115E. As described above, fluid channels 115A, 115B, 115C, 115D and 115E can be considered as separate fluid channels 115 or a single fluid channel 115. The detection device 100 also has a bottom surface 119.

[0112] Figure 5B yes Figure 5A 1 is a top view of an exemplary detection device 100. Flow channels 120G, 120H, and 120I that are not visible in this top view are shown with dashed lines to indicate their locations. Flow channels 120A-120F are shown with solid lines, but as described above, flow channels 120A-120F may also not be visible in this top view (e.g., they may be covered by a protective material such as an insulator material).

[0113] Figure 5C Is the detection device 100 along Figure 5A 5C". As shown, each of flow channels 120A, 120B, 120C, 120D, 120E, and 120F is in contact with the top of one of sensors 105 along the cross section (i.e., flow channel 120A is in contact with sensor 105A, flow channel 120B is in contact with sensor 105B, flow channel 120C is in contact with sensor 105C, flow channel 120D is in contact with sensor 105D, flow channel 120E is in contact with sensor 105E, and flow channel 120F is in contact with sensor 105F). Flow channel 120H is in contact with the bottom of each of sensors 105A, 105B, 105C, 105D, 105E, and 105F. It should be understood that although FIG. 5A to FIG. 5D The flow channel 120 is shown in contact with the sensor 105 , but the flow channel 120 may be coupled to the sensor 105 in general (ie, they may be directly connected, or there may be intervening components disposed between the flow channel 120 and the sensor 105 ).

[0114] Sensors 105A and 105B are connected by fluid channel 115A (at Figure 5C Not marked in Figure 5ASimilarly, sensors 105B and 105C are separated by fluid channel 115B, sensors 105C and 105D are separated by fluid channel 115C, sensors 105D and 105E are separated by fluid channel 115D, and sensors 105E and 105F are separated by fluid channel 115E. As discussed further below, either or both of the vertical walls of each fluid channel 115 may be wall 117.

[0115] In some embodiments, each sensor 105 is assigned to a single fluid channel 115. For example, in FIG. 5A to FIG. 5D In the exemplary device shown, the sensor 105 coupled to the flow channel 120A may be configured to sense the MNPs in the fluid channel 115A, the sensor 105 coupled to the flow channel 120B may be configured to sense the MNPs in the fluid channel 115B, the sensor 105 coupled to the flow channel 120C may be configured to sense the MNPs in the fluid channel 115C, the sensor 105 coupled to the flow channel 120D may be configured to sense the MNPs in the fluid channel 115D, and the sensor 105 coupled to the flow channel 120E may be configured to sense the MNPs in the fluid channel 115E.

[0116] exist FIG. 5A to FIG. 5C In the exemplary embodiment shown, there are more columns of sensors 105 than fluid channels 115 (i.e., in the exemplary embodiment shown, there are six columns corresponding to flow channels 120A-120F and only five fluid channels 115A-115E). In such an embodiment, each vertical wall of a fluid channel 115 can be a wall 117. In other words, a single fluid channel 115 can be sensed by twice as many sensors 105 as each other fluid channel 115. For example, in FIG. 5A to FIG. 5D In an exemplary embodiment of , any one of the flow channels 115 may be sensed by two columns of sensors 105. For example, flow channel 115B may be sensed by sensors 105 coupled to both flow channels 120B and 120C. In this example, the sensor 105 coupled to flow channel 120A would be assigned to sense the contents of flow channel 120A, the sensor 105 coupled to flow channel 120D would be assigned to sense the contents of flow channel 120C, the sensor 105 coupled to flow channel 120E would be assigned to sense the contents of flow channel 120D, and the sensor 105 coupled to flow channel 120F would be assigned to sense the contents of flow channel 120E.

[0117] Figure 5C Is the detection device 100 along Figure 5A5D". As shown, flow channel 120E contacts the top of each of sensors 105G, 105E, and 105H along the cross section. Each of flow channels 120G, 120H, and 120I contacts the bottom of one of sensors 105 along the cross section (i.e., flow channel 120G contacts sensor 105G, flow channel 120H contacts sensor 105E, and flow channel 120I contacts sensor 105H). As described above, Figure 5D The flow channel 120 shown need not be in direct contact with the sensor 105; rather, they may be connected through intervening components.

[0118] In some embodiments (see e.g. Figure 5E , Fig. 5F ), the detection device 100 includes a plurality of selector elements 111, each of which is coupled to a corresponding sensor in the sensor 105, wherein each of the selector elements 111 exhibits a threshold behavior such that for a signal above a given value (i.e., V th ), the selector element 111 has a high conductivity, and below this voltage, the conductivity of the selector element 111 is substantially zero. The selector element 111 may include, for example, a transistor, a diode, etc. One of ordinary skill in the art will appreciate that different schemes for addressing (selecting) the sensors 105 (individually or in groups) may be used to ensure that only the voltage drop across the intended sensors 105 is above V th Thus, the selector element 111 may be used to reduce the likelihood that a “sneak” current may be transmitted through adjacent elements and degrade the performance of the detection device 100 .

[0119] Figure 5E An exemplary sensor 105 selection method according to some embodiments is shown. Figure 5E In the exemplary embodiment shown, a corresponding selector element 111 (e.g., shown as a CMOS transistor) is coupled in series with the sensor 105. In this exemplary embodiment, three flow channels 120A, 120B, and 120C allow for sensing of characteristics of the sensor 105. Conceptually, flow channel 120A may be considered a readout flow channel, flow channel 120C may be considered a control flow channel, and flow channel 120B may be considered either or both a readout flow channel and a control flow channel. Figure 5E For more details on the configuration of the exemplary configuration shown, see BN Engel, J. B. Butcher, RWDave, M.DeHerrera, M.Durlam, G.Grynkewich, J.Janesky, SVPietambaram, NDRizzo, JMSlaughter, K.Smith, JJSun, and S.Tehrani, “A 4-Mb Toggle MRAM Based on a Novel Bit and SwitchingMethod,” IEEE Transactions on Magnetics, vol. 41, p. 132 (2005).

[0120] Fig. 5F Another exemplary sensor 105 selection method according to some embodiments is shown. Fig. 5F In the exemplary embodiment shown, the selector element 111 (e.g., a diode or similar threshold element known in the art, such as a semiconductor diode, an operational transconductance amplifier (OTA), a vanadium oxide layer, a capacitive threshold logic gate, etc.) is deposited "in-stack" with the magnetic film of the sensor 105 and then placed in a cross-point architecture. Fig. 5F The in-stack selector element 111 is shown below the sensor 105, but it should be understood that the order of the in-stack selector element 111 and the sensor 105 can be reversed. Corresponding selector devices (e.g., CMOS transistors) can be used to switch on each flow channel 120A, 120B to address / access each sensor 105 in the detection device 100. The use of CMOS select transistors may be simple due to the popularity of device factories available for manufacturing the front end (i.e., nano-fabrication for building CMOS transistors and underlying circuits), but the type of current used for operation can use a cross-point design to ultimately achieve the required density. Additional details regarding configurations suitable for selecting sensor 105 (e.g., using a cross-point array) can be found in C. Chappert, A. Fert, and F. N. Van Daul, “The emergence of spin electronics in data storage,” Nature Materials, Vol. 6, p. 813 (2007) and J. Woo et al., “Selector-less RRAM with non-linearity of device for cross-point array applications,” Microelectronic Engineering 109 (2013) pp. 360-363.

[0121] FIG. 6A to FIG. 6C1 shows an embodiment of a cross-point array architecture 300 that may be included in the detection device 100 according to some embodiments. For example, FIG. 6A to FIG. 6C The illustrated sensors 105 include MTJ elements 308, but it should be understood that some or all of the sensors 105 may be spin valves, as described above.

[0122] See also Fig. 6A , the cross-point array architecture 300 includes a top conductor 318 and a bottom conductor 320. Fig. 6A As shown in the exemplary embodiment of , the top wire 318 is oriented substantially perpendicular to the bottom wire 320 (at an angle of approximately 90 degrees to the bottom wire). An exemplary MTJ element 308 (e.g., sensor 105) is located between the intersections (dashed circles) of the array. The exemplary MTJ element 308 includes two or more FM layers 310, 314 separated by one or more non-magnetic layers 107 (e.g., comprising MgO). As shown, one of the FM layers is a free layer 310 that will rotate in the presence of a magnetic field, and the other of the FM layers is a pinned (or fixed) layer 314, which can be a single FM layer coupled to an AFM layer 312. Alternatively, a composite structure known as a synthetic antiferromagnet (SAF) can be used. The SAF includes two FM layers separated by a magnetic coupling layer (e.g., ruthenium), wherein one of the two FM layers is coupled to an AFM layer. It should be understood that while the exemplary layer arrangement of the MTJ element 308 shows a general structure with layers above or below other layers, intervening layers not shown may be inserted. Furthermore, as described above, additional layers may be disposed above and / or below the illustrated structures.

[0123] To illustrate some features of the cross-point array architecture 300, Figure 6B The cross point array architecture 300 is shown along the top conductor 318 direction (eg Fig. 6A denoted by the dashed line labeled "6B"), and Figure 6C The cross-point array architecture 300 is shown along the bottom conductor 320 (in Fig. 6A As shown, the side of the MTJ element 308 (sensor 105) is encapsulated by material 336, which can be an insulator. Optionally, as Figure 6B As shown, hard bias magnetic material 338 may also be deposited between MTJ elements 308. In embodiments including hard bias magnetic material 338, a thin layer of insulator 340 is also deposited on top of the hard bias magnetic material 338 to electrically insulate it from the top conductive line 318.

[0124] See also Figure 6C, the cross section shows a fluid channel 115 (e.g., a nanofluidic channel or a microfluidic channel), which can be, for example, a groove etched in an insulator. As shown, a small amount of insulator 322 is left on the sidewalls of the sensor 105 (shown as MTJ element 308) so that the MNPs do not electrically interact with the sensor 105. The portion of the insulator exposed to (and forming) the fluid channel 115 can form a wall 117, to which a polymerase molecule or a molecule to be detected (e.g., a nucleic acid sample) can be attached for detection.

[0125] In some embodiments, one or more sensors 105 are designed to generate radio frequency (RF) voltage when excited under the magnetic field and DC current bias of appropriate application. In a specific implementation, substantially all sensors 105 should oscillate with the same nominal frequency, wherein the frequency can be any value in the range of about 100MHz-10GHz. Here, the MNP for marking nucleotide precursors or other molecules can be ferromagnetic or superparamagnetic. The external magnetic field can be used to make the magnetic moment of all SNPs deposited in the array of the sensor 105 based on STO be oriented in the same direction (nominally parallel to or antiparallel to the FM layer of STO). The size of the field should also be within the range that STO can oscillate. As explained herein, when SNP is superparamagnetic, their presence can be detected by the sensor 105 based on STO without using the applied magnetic field.

[0126] For any individual STO sensor 105 in the array, the presence of MNPs will generate an additional magnetic field local to that STO. Since the frequency and noise of the RF signal generated by the STO depends on the magnitude of the magnetic field acting on the STO, its frequency may be shifted up or down from the case where there are no MNPs, and / or the phase noise of the RF signal may increase in the presence of MNPs. Therefore, measurement of the STO signal frequency or detection / monitoring of the phase noise of the RF signal can be used to detect the presence of MNPs.

[0127] Detection circuit

[0128] Determining the state of the sensor 105 (eg, determining at which frequency the STO is oscillating, detecting phase noise of an RF signal generated by the STO, etc.) may be accomplished using various types of detection circuits.

[0129] Some embodiments use delay line phase detectors similar to those described in U.S. Patents 8,654,465, 8,555,346, and 8,570,677, which describe detection circuits for reading data from magnetic media in hard drives using STOs. U.S. Patents 8,654,465, 8,555,346, and 8,570,677 are hereby incorporated by reference in their entirety for all purposes. Similar detectors can be used for molecular detection, i.e., detecting whether the frequency of the STO has shifted (e.g., due to the presence or absence of MNPs).

[0130] Figure 7 FIG. 6 shows a delay line detection circuit 600 that can be used to measure the frequency of an RF signal according to some embodiments. Figure 7 As shown, the RF signal from STO 604 is amplified by RF amplifier 606 and split into two paths (e.g., two different flow channels) by power divider 608. One path goes directly to the input of mixer 612 (and thereby suffers a small, negligible delay), while the other path adds a delay (T ) before sending the signal to the other input of mixer 612. d ) 614 is added to the signal. The time delay 614 need not be very large and can be introduced in any of several ways that will be understood by those of ordinary skill in the art. For example, the time delay 614 can be the result of directing one output of the power divider 608 through a longer wire trace than another output of the power divider 608. Alternatively or in addition, more complex methods can be used.

[0131] like Figure 7 As shown, the output of mixer 612 provides an input to a low pass or band pass filter 616. The output of low pass or band pass filter 616 may be amplified by an optional amplifier 618 which provides a DC output 620 suitable for processing. DC output 620 is proportional to the frequency of the RF signal generated by STO 604.

[0132] To understand the operation of the delay line detection circuit 600, consider the signal output from the STO 604 Where V, f and are the peak amplitude, frequency and phase of the STO signal respectively. Once the signal is split by the power divider 608, the delayed signal has a new phase at the mixer 612:

[0133]

[0134] When the delayed signal is mixed with the original signal, the resulting mixer 612 output is:

[0135]

[0136]

[0137] The low pass or band pass filter 616 has a cutoff frequency selected so that it removes the second cosine term. The output of mixer 612 thus simplifies to

[0138]

[0139] From this expression, it can be seen that the DC output 620 depends on the frequency of the signal from the STO 604 and the introduced time delay 614. If the time delay 614 remains constant, the circuit 600 can be used to detect the STO 604 frequency by measuring or detecting the output voltage of the detection circuit 600 (DC output 620).

[0140] Unlike magnetic recording, where the characteristic of interest is the high frequency modulation of the STO frequency due to the fluctuating magnetic field from the bits on the magnetic medium, the molecular detection (e.g., nucleic acid sequencing) applications contemplated by the present disclosure will be effectively steady state, since the chemical cycles used to introduce sequencing reagents are typically on a time scale of several minutes and real-time detection is not required. Therefore, the above mathematical model is generally sufficient to describe the output of a detection circuit such as detection circuit 600.

[0141] like Figure 7 As shown, an optional phase shifter 610 can be used to tune the phase difference between the signals on the two paths leaving the power divider 608 to a value that results in an output signal (DC output 620) of approximately zero when the STO 604 is operating in its non-perturbed (i.e., in the absence of MNPs) state so that the detection circuit 600 operates in the linear region of phase / frequency. This approach allows for very rapid assessment of the presence / absence of MNPs in a large area array of STOs 604, which can increase the bandwidth of the sequencing system and improve the speed of data collection.

[0142] One advantage of this detection method is that if MNPs made of different ferromagnetic materials (and therefore having different saturation magnetizations M) are added simultaneously to the fluidic channel 115 of the detection device 100 (e.g., a nanofluidic or microfluidic channel disposed near the array 110 of sensors 105), these MNPs can be distinguished from each other because each type of MNP exerts a different, distinguishable magnetic field on the STO 604. Therefore, each MNP type causes the STO 604 to produce a signal having a frequency that results in a DC output 620, thereby allowing it to be distinguished from other MNP types in use. Therefore, if the STO / circuit signal-to-noise ratio (SNR) is large enough, identification of the type of MNP detected by the STO 604 can be achieved by reading the voltage at the output of the detection circuit 600 (DC output 620) and determining which MNP type is associated with the detected voltage level.

[0143] For example, in a DNA sequencing application, a different MNP type can label each of four different nucleotide precursors, thereby reducing the number of cycles per base read from two or four (which is currently performed depending on the sequencing technology) to one. Since cycles take on the order of several minutes to complete, reducing the number of cycles required for each base read will significantly improve the data collection throughput of the sequencing system.

[0144] For example, assume that the DC output 620 of the detection circuit 600 varies based at least in part on the type of MNPs in the vicinity of the STO 604 according to the following table as described above:

[0145] Magnetic nanoparticle identity Labeled nucleotide precursors Expected DC output Base identity MNP1 A <![CDATA[V 1 <v 输出 <V 2 ]]> T MNP2 C <![CDATA[V 2 <v 输出 <V 3 ]]> G MNP3 T <![CDATA[V 3 <v 输出 <V 4 ]]> A MNP4 G <![CDATA[V 4 <v 输出 <V 5 ]]> C

[0146] If the DC output 620 is detected at V 3 and V 4 In the range between , it can be determined that MNP3 labeled thymine has been detected, which means that the identity of the last paired base is adenine.

[0147] Note that although Figure 7 A single STO 604 is shown at the input of the delay line detection circuit 600, but a collection of STOs 604 may share a single detection circuit 600. In such embodiments, a multiplexer, switch, or similar device may be used to couple the STO 604 to the detection circuit 600. Each STO 604 may then be read by cycling through the multiplexer, switch, etc. Using a multiplexer or similar device to allow sharing of the detection circuit 600 may reduce the footprint of the detection circuit 130 included in the detection device 100. It is also contemplated to use a separate circuit board or chip to implement the detection circuit 600 of the sequencing system; however, this approach may increase the latency of the system.

[0148] Some embodiments use superparamagnetic MNPs (thermally unstable MNPs, such that the magnetic field they generate fluctuates over time) as magnetic labels. In such embodiments, the fluctuating magnetic field from one or more thermally unstable MNPs near the STO 604 can increase the phase / frequency noise of the STO 604 in the absence of an applied external magnetic field. Figure 8 A detection circuit 700 is shown that may be used according to some embodiments using superparamagnetic MNPs. Figure 8 It shows that Figure 7 Many of the same components discussed in the context of . Figure 7 The explanation of those components in the context of Figure 8 , and will not be repeated here.

[0149] exist Figure 8 In the example, the output signal is directed to the spectrum analyzer 622, rather than directly analyzing the output of the detection circuit ( Figure 7 The voltage of the DC output 620 in the STO 604 is used to determine the frequency of the RF signal generated by the STO 604. Since the MNPs are superparamagnetic, the trace measured on the spectrum analyzer 622 is proportional to the spectral density of the combination of the intrinsic (thermal / magnetic) noise from the STO 604 and the induced noise from the superparamagnetic MNPs. By integrating the measured trace (e.g., measuring the area under the trace), an approximation of the integrated phase / frequency noise of the STO 604 signal can be obtained. The presence of MNPs can be detected by comparing the measured integrated noise with the integrated noise measured in a reference cell (where the STO 604 is not exposed to MNPs).

[0150] Figure 8 The use of detection circuit 700 may be advantageous because it does not require any equipment to generate a global magnetic field, but it may be slower than other methods, such as using Figure 7 Detection circuit 600. Figure 8 The detection circuit 700 may use external spectrum analyzer electronics, which may increase cost and may be less convenient than other approaches. However, if an external spectrum analyzer 622 is used, the external spectrum analyzer may be shared by multiple instances of the detection circuit 700 (e.g., coupled to the spectrum analyzer 622 via a switch matrix, multiplexer, filter bank, etc.) to reduce incremental cost. Alternatively, the functionality of the spectrum analyzer 622 may be performed using a processor such as a digital signal processor (DSP), which may be included in the detection circuit 130 or used in addition to the detection circuit.

[0151] and Figure 7 Similar to the detection circuit 600, Figure 8The detection circuit 700 may be shared by multiple STOs 604. For example, a set of STOs 604 may be coupled to the detection circuit 700 via one or more multiplexers, switches, etc., which may reduce the footprint of the detection circuit 130 on the detection device 100.

[0152] Fig. 9 Another detection circuit 800 is shown according to some embodiments. The detection circuit 800 allows the phase noise of the STO 604 to be measured directly (e.g., without using a spectrum analyzer 622). The frequency of the STO 604 is tracked by a phase locked loop (PLL) 820 having a low pass filter 810. The PLL 820 attempts to match the frequency of the RF signal generated by the STO 604, in which case the output signal from the PLL 820 is approximately zero. When the frequency of the RF signal is offset, the output of the PLL 820 becomes non-zero, which produces a non-zero error signal 818. Since the RF signal generated by the STO 604 will be noisier when there are MNPs (ferromagnetic or superparamagnetic) in its vicinity than when there are no MNPs in its vicinity, the detection circuit 800 detects the phase / frequency noise in the RF signal generated by the STO 604.

[0153] The detection circuit 800 includes many Figure 7 and Figure 8 The detection circuits 600 and 700 shown have the same elements, and the description of those elements will not be repeated here. Fig. 9 In an exemplary embodiment of the present invention, the PLL 820 includes a signal generator 808 that generates a clean RF signal based on a tuning input 824. The tuning input 824 is determined by a loop filter including components 812 and 814 (which may be resistors) and an error amplifier 816. The mixer 802 mixes the clean RF signal from the signal generator 808 with the signal output from the amplifier 806. The low-pass filter 810 of the PLL 820 then filters the output from the mixer 802. The resulting error signal 818 output from the PLL 820 is the phase noise of the STO 604, and its characteristics depend on the presence or absence of the MNP (e.g., affected or changed by the presence or absence of the MNP). Therefore, the error signal 818 can be used to detect the presence or absence of one or more SNPs near the STO 604.

[0154] Detection Methods

[0155] The above-described sensors 105 and / or detection devices 100 can be used to detect molecules labeled by MNPs, as further described below. Suitable detection methods include those in which a binary decision (e.g., yes / no, 1 / 0, etc.) is made as to whether the MNPs, and therefore the molecules to which the MNPs are coupled, are present in the vicinity of the sensor 105. For simplicity, the following explanation is given in the context of DNA sequencing, but as previously mentioned, it should be understood that the described methods can also be used in other applications and for detecting types of molecules other than nucleic acids.

[0156] In some embodiments, a target molecule to be detected (e.g., a nucleic acid strand to be sequenced) is attached to a wall 117 of a fluid channel 115 of the detection device 100. A polymerase can be introduced at this point. For example, the polymerase can be bound (attached or coupled) to the wall 117 along with the target ssDNA to be sequenced. Nucleotide precursors labeled by MNPs can then be introduced into the fluid channel 115. The polymerase is used to incorporate complementary nucleotide precursors labeled by the MNPs into the target DNA strand. Only appropriate (complementary) bases (i.e., for DNA sequencing, cytosine (C) is complementary to guanine (G), or adenine (A) is complementary to thymine (T)) are incorporated, and their presence can be detected by the sensor 105. Assuming that the process processes one base pair at a time, the presence or absence of the MNPs labeling the complementary nucleotide precursors, and therefore the nucleotide precursors having that base identity in the target DNA strand, can be detected using, for example FIG. 4A to FIG. 5F to be determined by various device implementation schemes.

[0157] The presence or absence of MNPs near a particular sensor 105 can be detected, for example, by applying a magnetic field across the sensor 105 and applying a bias current to read the sensor 105. Applying a magnetic field across the sensor 105 is optional (e.g., as described above, Figure 8The detection circuit 700 shown can be used without applying a magnetic field), but may be beneficial in applications where multiple types of MNPs are present (e.g., in DNA sequencing applications where different nucleotide precursors are labeled with different MNP types and multiple nucleotide precursors are added to the fluid channel 115 substantially simultaneously). If applied, a magnetic field can be applied using an electromagnet (e.g., by placing pole pieces on either side of the detection device), a distributed coil, a solenoid oriented perpendicular to the fluid channel 115, etc. to generate a magnetic field in the direction of the moment of the pinned layer 314. The device for generating a magnetic field can be mounted on, for example, the bottom surface 119 of the detection device 100. As another example, the device for generating a magnetic field can be included in a system including the detection device 100. It should be understood that other suitable devices for generating a magnetic field, such as, for example, by using a permanent magnet or a superconducting magnet, are possible, which are specifically contemplated herein and are not excluded. If used, the applied magnetic field aligns the moments of all MNPs in a common direction so that the signals measured due to the presence of the MNPs are similar.

[0158] In the case where the free layer is excited by the spin transfer effect and the moment of the fixed layer is fixed, the STO excited as described above (e.g., Figure 3C In the context of ) an RF voltage signal will be generated from the applied DC current due to resistance fluctuations caused by the magnetoresistance effect. Thus, by connecting the sensor 105 to detection electronics / circuitry as described above, the presence and / or absence of MNPs in the vicinity of the sensor 105 can be detected. In, for example, DNA sequencing applications, nucleotide precursors (or more generally, nucleic acids) labeled by MNPs and incorporated into the target DNA strand by a polymerase can be detected by determining the frequency or frequency noise of the RF signal generated by the STO, because only in the presence of MNPs that label the nucleotide precursors incorporated into the target DNA strand being sequenced, the local magnetic field is sufficient to shift the oscillation frequency of the RF signal generated by the STO or increase its noise.

[0159] Molecular detection methods can use a single MNP type or multiple MNP types. Fig.10 An exemplary sequential binary method 500 suitable for DNA sequencing is shown, in accordance with some embodiments, where a single MNP type is used to label all four nucleotide precursors. It should be understood that Fig.10 The process is shown for a single sensor 105. In embodiments where the detection device 100 includes a plurality of sensors 105, some steps of the method 500 (eg, steps 510, 512, 516) may be performed independently for each of the plurality of sensors 105.

[0160] At 502, method 500 begins. At 504, molecules of each of four nucleotide precursors (A, T, C, and G) are labeled by the same type of MNP. Then, different nucleotide precursors, each labeled with the same MNP type, are incorporated into, for example, the fluid channel 115 of the detection device 100 one at a time. Therefore, at 506, the first nucleotide precursor to be tested is selected. At 508, the selected (magnetically labeled) nucleotide precursor is added to the fluid channel 115 of the detection device 100. After enough time has passed to allow the nucleotide precursor to be incorporated into the target DNA strand being sequenced, at 510, the detection circuit 130 detects whether the MNP is present near the STO 604.

[0161] Step 510 can be implemented in a variety of ways. For example, using Figure 7 In an embodiment of the detection circuit 600 (or similar circuit), when the DC output 620 exceeds the threshold, it can be inferred that MNP exists. Figure 8 In the embodiment of the detection circuit 700 (or similar circuit) and superparamagnetic SNP, when the spectrum analyzer 622 detects that the STO 604 is generating an RF signal with a phase noise exceeding a threshold, it can be inferred that MNP exists. Fig. 9 In embodiments of the detection circuit 800 (or similar circuits), when the error signal 818 exceeds a threshold, the presence of MNPs (which may be, but need not be, superparamagnetic) may be inferred.

[0162] If it is determined at 510 that the MNP is present in the vicinity of the STO, then at 512 it is determined that the tested nucleotide precursor is incorporated into a DNA strand that is coupled to a binding site associated with the sensor 105. The identity of the base (its complementary sequence) paired with the tested nucleotide precursor can then be recorded, and the method ends at 514. However, if it is determined at 510 that the MNP is not present in the vicinity of the STO (interpreted to mean that the previously tested nucleotide precursor is not incorporated at the binding site associated with the sensor 105), then at 516, it is determined whether the previously tested nucleotide precursor is the last of the four nucleotide precursors to be tested. If yes, the method ends at 514. If not, the method returns to 506, where the next nucleotide precursor to be tested is selected, and at least steps 508 and 510 are repeated. Once the method 500 has ended, the magnetic label can be cut before starting the next sequencing cycle.

[0163] Method 500 can be performed using one or more sensors 105. It should be understood that when more than one sensor 105 is used, the decision at 510 may be different for different sensors 105. For example, in some types of SBS, long chains of DNA (or multiple long chains of DNA from a single donor organism) are divided into smaller random length fragments before sequencing. All these shorter chains from the same donor are randomized subchains of the complete chain to be sequenced. For example, if the complete chain includes the sequence ATGGCTTAG, the shorter chain may include, for example, non-overlapping subchains (e.g., ATGG and TTAG), and if multiple longer chains are divided into subchains, the shorter chain may include subchains (e.g., GGCTTA and TTAG) that overlap partially or completely with other subchains. All shorter randomized subchains can be sequenced at the same time, which may be after amplification. In such applications, it should be understood that because the subchains do not represent the same subsequence, it may be advantageous to detect the RF signal generated by each sensor 105 to detect MNPs, because the sequencing of the subchains will not be coordinated (or synchronized) between the subchains. For example, during a single sequencing cycle, the first daughter strand may incorporate cytosine, the second daughter strand may incorporate thymine, and the third daughter strand may incorporate adenine. In order to sequence multiple random fragments of a larger nucleic acid strand, it is desirable to determine in each sequencing cycle whether each dNTP type is incorporated and at which physical position each dNTP type is incorporated. Therefore, when using Fig.10 In the exemplary method 500 shown, the decision at 510 may be "yes" for one sensor 105 after adding a particular nucleotide precursor and "no" for another sensor. Thus, when sequencing a randomized substrand of a nucleic acid such as DNA, it may be advantageous to test all four nucleotide precursors during each sequencing cycle, even if for some sensors 105 the decision at 510 is "yes" for the first, second, or third tested nucleotide precursor.

[0164] Although Fig.10 It is assumed that each nucleotide precursor is labeled by the same type of MNP, but it is not necessary to use the same type of MNP for each nucleotide precursor. For example, it may be convenient to use the same type of MNP for each nucleotide precursor, but alternatively, different nucleotide precursors can be labeled with different types of MNPs. In other words, two or more nucleotide precursors can be labeled with the same type of MNP, or two or more nucleotide precursors can be labeled with different types of MNPs.

[0165] For example, various other embodiments involve using multiple MNP types (e.g., MNP 1, 2, 3, and 4), each type causing sensor 105 to generate an RF signal having properties that distinguish it from other RF signals caused by the other three MNP types. Focusing on the DNA example to illustrate, each individual base (A, T, C, G) can be labeled with a different type of MNP (e.g., base A is labeled with MNP 1, base C is labeled with MNP 2, base G is labeled with MNP 3, and base T is labeled with MNP 4), specifically by labeling each base separately and then mixing them together, or functionalizing each type of MNP in a different way so that it has an affinity for a specific (e.g., its designated) base. In a single chemistry run, all labeled (magnetically labeled) bases can be introduced into a microfluidic pool (e.g., fluid channel 115 of detection device 100) to which the DNA strands (e.g., fragments) to be sequenced are already attached (e.g., as described above in the discussion regarding detection device 100).

[0166] After the target DNA chain to be sequenced is attached to the detection device 100, all four kinds of magnetically labeled nucleotide precursors can be introduced into the fluid channel simultaneously (or substantially simultaneously). Polymerase is used to incorporate the nucleotide precursors complementary to those nucleotide precursors into the target chain. The RF signal produced (or not produced) by the STO of the detection device 100 can be used to identify which MNP (and therefore nucleotide precursors) (if any) has been incorporated near each sensor 105. After every kind of nucleotide precursor has been introduced into the fluid channel 115 and the sensor 105 has been read, for example enzymatic or chemical cutting can be used to cut the MNP from the introduced magnetically labeled nucleotide precursors, as known in the art. Then the process can be repeated for the next unpaired base in the chain to be sequenced.

[0167] Thus, in some embodiments of DNA sequencing applications, instead of using a binary approach with four chemical steps for each base read (sequencing cycle), four different MNPs (each causing the STO to produce a discernible RF signal or a discernible change in an RF signal) can be used as magnetic labels, and all of them can be detected in a single chemical step. For example, each type of molecule (e.g., in a DNA sequencing application, each dNTP type) can be labeled by a different MNP type, wherein each MNP type causes the STO to produce (or not produce) an RF signal having at least one characteristic (e.g., frequency, frequency noise) that enables the presence or absence of the MNP to be discerned from all other MNPs used as magnetic labels. For example, in a DNA sequencing application, A can be labeled by MNP1, T can be labeled by MNP2, C can be labeled by MNP3, and G can be labeled by MNP4, wherein the RF signals produced by the STOs affected by MNP1, MNP2, MNP3, and MNP4 can be discerned by the detection circuit 130. The detection circuit 130 (e.g., in Figures 7 to 9 The exemplary embodiments shown and described in the context of ) can identify which nucleotide precursors have been incorporated into the DNA chain bound near and associated with each STO.

[0168] For example, as described above (see e.g. Figure 7 ), the DC output 620 of the detection circuit 600 may be at different levels depending on the type of MNPs near the STO 604. Thus, using a table or key as described above, the detected DC output 620 voltage may be mapped to one of four MNP types.

[0169] Fig.11 FIG. 1 is a diagram showing a method for using multiple nucleotide precursors labeled with different MNP types according to some embodiments and comprising Figure 7 The detection circuit 600 (or similar circuit) shown is a flow chart of the method 400 of DNA sequencing using the detection circuit 130 . Fig.11 It is assumed that all four nucleotide precursors are labeled with different MNP types that allow them to be distinguished, and it is assumed that all four magnetically labeled nucleotide precursors are added to the fluid channel 115 substantially simultaneously, but it will be understood that method 400 can be performed with fewer nucleotide precursors labeled with distinguishable MNPs and / or introduced into the fluid channel 115 substantially simultaneously.

[0170] At 402, method 400 begins. At 404, molecules of each of four nucleotide precursors (A, T, C, and G) are labeled with different types of MNPs. At 406, up to all of the different nucleotide precursors, each labeled with a different MNP type, are then introduced substantially simultaneously into, for example, a fluid channel 115 of a detection device 100. After sufficient time has passed to allow the magnetically labeled nucleotide precursors to be incorporated into the target DNA strand being sequenced, at 408, detection circuit 130 detects a DC output from the detection circuit (e.g., DC output 622 from detection circuit 600).

[0171] Next, we determined which of the four MNP types was most likely responsible for the detected DC output. Fig.11 One way to make this determination is shown. Fig.11 In the method 400, at 410, it is determined whether the DC output is within a first range of values, wherein the first range of values ​​corresponds to a first MNP type. If so, at 418, the first MNP type is identified as the detected MNP type, and at 426, the identity of the base complementary to the nucleotide precursor labeled by the first MNP type is recorded, and the method 400 ends at 428 for the sensor 105. If it is determined at 410 that the DC output is not within the first range of values, at 412, it is determined whether the DC output is within a second range of values, wherein the second range of values ​​corresponds to a second MNP type. If so, at 420, the second MNP type is identified as the detected MNP type, and at 426, the identity of the base complementary to the nucleotide precursor labeled by the second MNP type is recorded, and the method 400 ends at 428 for the sensor 105. If it is determined at 412 that the DC output is not within the second range of values, at 414, it is determined whether the DC output is within a third range of values, wherein the third range of values ​​corresponds to a third MNP type. If yes, then at 422, the third MNP type is identified as the detected MNP type, and at 426, the identity of the base complementary to the nucleotide precursor labeled by the third MNP type is recorded, and the method 400 ends at 428 for the sensor 105. If it is determined at 414 that the DC output is not within the third range of values, then at 416, it is determined whether the DC output is within a fourth range of values, where the fourth range of values ​​corresponds to the fourth MNP type. If yes, then at 424, the fourth MNP type is identified as the detected MNP type, and at 426, the identity of the base complementary to the nucleotide precursor labeled by the fourth MNP type is recorded. If no, the method 400 (and sequencing cycle) ends at 428 without identifying any MNP type.

[0172] It should be understood that Fig.11 Only for use as Figure 7The detection circuit shown is used to perform the detection in one way. In addition, the Fig.11 For example, although Fig.11 Steps 410, 412, 414, and 416 are shown as being performed sequentially, but they may be performed in parallel. Furthermore, it should be understood that once a DC output has been detected, a lookup table may be used to determine which MNP type, if any, has been detected. Fig.11 Other ways of, according to the disclosure of this article, these ways will be obvious. Therefore, Fig.11 It is exemplary only and is not intended to be limiting.

[0173] In addition, as mentioned above Fig.10 As explained in the context of , the method 400 may be performed using one or more sensors 105. It should be understood that when more than one sensor 105 is used, the decisions at different decision points (e.g., 410, 412, 414, 416) may be different for different sensors 105.

[0174] After completion of method 400, the MNPs can be cleaved from the incorporated magnetically labeled nucleotide precursors using, for example, enzymatic or chemical cleavage, as is known in the art. Method 400 can then be repeated to identify the next unpaired base in the strand being sequenced.

[0175] In the use of superparamagnetic MNPs and Figure 8 In the embodiment of the detection circuit 700 shown, the frequency noise of the RF signal generated by the STO 604 can be directly measured / detected without applying a magnetic field to detect the molecule. As long as each type of MNP in use causes the RF signal generated by the STO 604 to have a frequency noise that is distinguishable from the frequency noise of the RF signal generated by the STO 604 in the presence of each other type of MNP, the identity of the MNP and therefore the final paired base can be determined based on the detected frequency noise of the RF signal generated by the MNP.

[0176] Fig.12 FIG. 1 is a diagram showing a method for using multiple nucleotide precursors labeled with different (eg, superparamagnetic) MNP types and comprising Figure 8 The detection circuit 700 (or similar circuits) shown is a flow chart of a method 450 for DNA sequencing using the detection circuit 130 . Fig.12 It is assumed that all four nucleotide precursors are labeled with different MNP types that allow them to be distinguished, and it is assumed that all four magnetically labeled nucleotide precursors are added to the fluid channel 115 substantially simultaneously, but it will be understood that method 450 can be performed with fewer nucleotide precursors labeled with distinguishable MNPs and / or introduced into the fluid channel 115 substantially simultaneously.

[0177] At 452, method 450 begins. At 454, molecules of each nucleotide precursor in four nucleotide precursors (A, T, C and G) are labeled by different types of MNPs. As described above, MNPs can be superparamagnetic, which allows detection without applying a magnetic field. At 456, then, the most all different nucleotide precursors each labeled by different MNP types are substantially simultaneously introduced into, for example, the fluid channel 115 of the detection device 100. After enough time has passed to allow the magnetically labeled nucleotide precursors to be incorporated into the target DNA strand being sequenced, at 458, the detection circuit 130 detects the frequency noise of the signal generated by the sensor 105 (e.g., using a spectrum analyzer as shown in the detection circuit 700).

[0178] Next, one of the four MNP types is identified based on the detected frequency noise. Fig.12 In the embodiment of the present invention, at 460, it is determined whether the detected frequency noise is a first frequency noise corresponding to a first MNP type. This determination can be made in any of a variety of ways. For example, as described above in Figure 8As described in the discussion of , the spectral density of the combination of intrinsic (thermal / magnetic) noise from the STO 604 and induced noise from the superparamagnetic MNPs (e.g., without applying a magnetic field) can be determined to determine whether the frequency noise indicates the presence of one or more MNPs. If so, at 468, the first MNP type is identified as the detected MNP type, and at 476, the identity of the base complementary to the nucleotide precursor labeled by the first MNP type is recorded, and the method 450 ends at 478 for the sensor 105. If the detected frequency noise is not the first frequency noise, at 462, it is determined whether the detected frequency noise is a second frequency noise corresponding to a second MNP type. If so, at 470, the second MNP type is identified as the detected MNP type, and at 476, the identity of the base complementary to the nucleotide precursor labeled by the second MNP type is recorded, and the method 450 ends at 478 for the sensor 105. If the detected frequency noise is not the second frequency noise, then at 464, it is determined whether the detected frequency noise is a third frequency noise corresponding to a third MNP type. If yes, then at 472, the third MNP type is identified as the detected MNP type, and at 476, the identity of the base complementary to the nucleotide precursor labeled by the third MNP type is recorded, and the method 450 ends at 478 for the sensor 105. If the detected frequency noise is not the third frequency noise, then at 466, it is determined whether the detected frequency noise is a fourth frequency noise corresponding to a fourth MNP type. If yes, then at 474, the fourth MNP type is identified as the detected MNP type, and at 476, the identity of the base complementary to the nucleotide precursor labeled by the fourth MNP type is recorded, and the method 450 ends at 478 for the sensor 105. If no, the loop ends at 478 for the sensor 105 without identifying any MNP type.

[0179] It should be understood that Fig.12 Only for use as Figure 8 The detection circuit shown is used to perform the detection in one way. In addition, the Fig.12 For example, although Fig.12 Steps 460, 462, 464, and 466 are shown as being performed sequentially, but they may be performed in parallel. Furthermore, it should be understood that once the frequency noise of the RF signal generated by the STO 604 has been detected, a lookup table may be used to determine which MNP type, if any, has been detected. Thus, Fig.12 It is exemplary only and is not intended to be limiting.

[0180] In addition, as mentioned above Fig.10As explained in the context of , the method 450 may be performed using one or more sensors 105. It should be understood that when more than one sensor 105 is used, the decisions at different decision points (e.g., 460, 462, 464, 466) may be different for different sensors 105.

[0181] Fig. 9 The detection circuit 800 shown can also be used in embodiments where different nucleotide precursors are labeled with different types of MNPs and multiple nucleotide precursors are introduced into the fluid channel 115 substantially simultaneously. For example, the detection circuit 800 may include an adjustable reference oscillator (signal generator 808) to generate the signal described above in Fig. 9 Optionally, other portions of the detection circuit 800 may be adaptive or adjustable (e.g., the loop filter 820 may be adaptive or modifiable). During each sequencing cycle, after allowing sufficient time for incorporation, the frequency of the clean RF signal may then be set to the frequency of the RF signal produced by the STO 604 that is expected to be present in the presence of each MNP type in use. By monitoring whether the error signal 818 exceeds a threshold for each test frequency, thereby detecting the "noisiness" of the RF signal at the expected frequency, the detection circuit 130 may determine which MNP type (if any) is near the STO 604.

[0182] Fig.13A method 570 for DNA sequencing using nucleotide precursors labeled with MNPs and an adjustable reference oscillator 808 according to some embodiments is shown. At 572, the method 570 begins. At 574, each nucleotide precursor type (G, A, C, T) is labeled by a different MNP type (e.g., A is labeled by MNP1, T is labeled by MNP2, C is labeled by MNP3, and G is labeled by MNP4). At 576, the nucleic acid chain to be sequenced, the polymerase molecule, and the nucleotide precursors labeled with MNPs are introduced into the fluid channel 115 of the detection device 100. After a time period suitable for allowing the incorporation of nucleotide precursors, at 578, the reference oscillator is set to generate an RF signal at a first frequency, wherein the first frequency is the expected oscillation frequency of the RF signal generated by the ST0604 in the presence of the first MNP type. At 580, the detection circuit 130 (e.g., the detection circuit 800) determines whether the error signal is non-zero (e.g., whether it is greater than a threshold value). If yes, then at 594 it can be inferred that the STO 604 has detected the first MNP type, the identity of the complementary base can be recorded, and the method 570 ends at 596 for the sensor 105. If no, the method 570 proceeds to 582, and the reference oscillator is set to generate an RF signal at a second frequency, wherein the second frequency is the expected oscillation frequency of the RF signal generated by the STO 604 in the presence of the second MNP type. At 584, the detection circuit determines whether the error signal is non-zero (e.g., whether it is greater than a threshold). If yes, then at 594 it can be inferred that the STO 604 has detected the second MNP type, the identity of the complementary base can be recorded, and the method 570 ends at 596 for the sensor 105. If no, the method 570 proceeds to 586, and the reference oscillator is set to generate an RF signal at a third frequency, wherein the third frequency is the expected oscillation frequency of the RF signal generated by the STO 604 in the presence of the third MNP type. At 588, the detection circuit determines whether the error signal is non-zero (e.g., whether it is greater than a threshold value). If so, at 594, it can be inferred that the STO 604 has detected the third MNP type, the identity of the complementary base can be recorded, and the method 570 ends for the sensor 105 at 596. If not, the method 570 proceeds to 590 and the reference oscillator is set to generate an RF signal at a fourth frequency, wherein the fourth frequency is the expected oscillation frequency of the RF signal generated by the STO 604 in the presence of the fourth MNP type. At 592, the detection circuit determines whether the error signal is non-zero (e.g., whether it is greater than a threshold value). If so, at 594, it can be inferred that the STO 604 has detected the fourth MNP type, the identity of the complementary base can be recorded, and the method 570 ends for the sensor 105 at 596.If not, the method ends at 596 for the sensor 105 without detecting any MNPs near the STO 604 .

[0183] It should be understood that the detection circuit 130 may include a plurality of reference oscillators 808, each of which is configured to generate a frequency close to the frequency of the RF signal of the STO in the presence of one of the MNP types used. During each sequencing cycle, after the introduction of the four MNP-labeled nucleotide precursors, the switch may be cycled through each reference oscillator 808 in turn, and the error signal may be detected as described above. Likewise, the error signal should be non-zero only when the switch is connected to a reference oscillator 808 that oscillates at the frequency associated with the approximate oscillation frequency of the STO in the presence of the MNP type. From this information, the identity of the particle can be determined.

[0184] Methods of manufacturing sensors and detection devices

[0185] In some embodiments, the detection device 100 is fabricated using photolithography processes and thin film deposition. Fig.14A A method 850 of manufacturing the detection device 100 is shown, and Fig. 14B Results of steps of a manufacturing method 850 according to some embodiments are shown, with the final panel showing a polymerase bound to a wall 117 adjacent to a sensor 105 (e.g., when the detection device 100 is used for nucleic acid sequencing). At 852, method 850 begins. At 854, at least one flow channel 120 is fabricated on a substrate, for example, by depositing one or more metal layers, patterning an array of flow channels and spaces in a polymer layer applied on top of the metal layer using, for example, photolithography, using the polymer layer as a mask for etching the metal layer into an array of flow channels, depositing an insulating dielectric material, stripping the polymer layer and dielectric material above the flow channels, and performing chemical mechanical polishing to planarize the surface. At 856, a sensor array 110 is fabricated on at least one flow channel 120. Each sensor 105 of the sensor array 110 has a bottom portion 108 and a top portion 109. (See Figure 1A . ) Bottom portion 108 is coupled to at least one flow channel 120 . In some embodiments, bottom portion 108 of each sensor 105 is in contact with at least one flow channel 120 .

[0186] At 858, dielectric material is deposited between sensors 105 of sensor array 110. At 860, additional flow channels 120 are fabricated. Each of these additional flow channels 120 is coupled to a top portion 109 of at least one sensor 105 in sensor array 110. In some embodiments, the top portion 109 of each sensor 105 is in contact with a flow channel 120. In some embodiments, the bottom portion 108 of sensor 105 is in contact with first flow channel 120A, and the top portion 109 of sensor 105 is in contact with second flow channel 120B. At 862, a portion of the dielectric material adjacent to sensor 105 is removed (e.g., by milling, etching, or any other suitable removal process) to form fluid channel 115. At 864, method 850 ends.

[0187] Electrical detection for DNA sequencing described in the present disclosure can provide multiple advantages over currently used technologies involving optical detection methods. For example, electrical detection is not limited in terms of scaling flow cell size in the same way that optical detection is limited due to diffraction-limited optical imaging. Magnetic detection is a form of electrical detection for sequencing that has advantages over commonly proposed tunnel current detection schemes because tunnel current detection methods measure very small currents (which reduces the SNR) and tunnel junction elements are directly exposed to sequencing chemicals, which can cause corrosion or other harmful problems that reduce the accuracy of the sequencing process. In contrast, magnetic detection has a larger signal (and better SNR) and can be performed without direct contact of labeled particles (e.g., MNPs) with sensor 105, thereby allowing sensor 105 to be coated with a protective layer that can mitigate interactions with sequencing reagents.

[0188] According to various embodiments described herein, the STO detection technology can be used in a relatively simple binary method to detect the presence of incorporated DNA nucleotide precursors (e.g., via detecting a finite or approximately zero voltage at the output of an analog detection circuit). Therefore, it can reduce the SNR required to operate the detection system with high accuracy, which makes the STO design easier. It also provides flexibility in the selection of MNPs used as markers for the molecules to be detected, because only a small magnetic field without any specific field direction will turn the STO off or on. Therefore, both superparamagnetic and ferromagnetic particles can be used to align particles at different sites in a circulation pool (e.g., sensor array 110) without using an external magnetic field.

[0189] The embodiments of the present invention using digital processing for detection may also facilitate the use of reliable, accurate hardware components (e.g., filters, amplifiers, PLLs, DSPs, and / or other similar components) and well-known algorithms (e.g., Fourier transforms or any other known frequency analysis techniques for evaluating the frequency content and / or frequency noise of RF signals) to detect changes in the STO oscillation frequency and / or frequency noise.

[0190] The limitation of magnetic detection may be the SNR of the sensor 105. An advantage of some disclosed embodiments is that the STO 604 operates at a higher frequency and therefore will have reduced 1 / f noise, which results in a reduction in the overall noise under equilibrium conditions. Another advantage is that because a single voltage is detected at the output of the detector, the use of the STO 604 should be fast and should allow high data collection throughput, which is desirable in detection systems (e.g., for DNA sequencing).

[0191] In the foregoing description and drawings, specific terminology is set forth to provide a thorough understanding of the embodiments disclosed herein. In some cases, the terminology or drawings may suggest specific details that are not necessary to practice the invention.

[0192] In order to avoid unnecessarily obscuring the present disclosure, well-known components are shown in block diagram form and / or have not been discussed in detail, or in some cases not discussed at all.

[0193] Unless otherwise specifically defined herein, all terms should be given their broadest possible interpretation, including the meaning implied by the specification and drawings and the meaning understood by those skilled in the art and / or as defined in dictionaries, books, etc. As expressly set forth herein, some terms may not be conformed to their ordinary or customary meanings.

[0194] As used herein, the terms "above," "below," "between," "on," and other similar terms refer to the relative position of one layer with respect to other layers. Thus, for example, a layer disposed above or below another layer may be directly in contact with the other layer, or may have one or more intervening layers. Additionally, a layer disposed between layers may be directly in contact with both layers, or may have one or more intervening layers. Conversely, a first layer "on a second layer" is in contact with the second layer. The relative positions of these terms do not define or limit the layers to a certain vector space orientation of the layers.

[0195] The terms "exemplary" and "embodiment" are used to represent examples, rather than preferences or requirements. Unless otherwise specifically stated or understood in other ways within the context of use, conditional language used herein such as "may", "can", "may", "may", "for example", etc. is generally intended to convey that certain embodiments include certain features, elements and / or steps, while other embodiments do not include these certain features, elements and / or steps. Therefore, such conditional language is generally not intended to imply that features, elements and / or steps are required by one or more embodiments in any way, or one or more embodiments must include logic for determining whether these features, elements and / or steps are included in any specific embodiment or will be performed in any specific embodiment with or without other inputs or prompts. The terms "include", "include", "have", etc. are synonymous and are used inclusively in an open manner, and do not exclude additional elements, features, actions, operations, etc. In addition, the term "or" is used with its inclusive meaning (rather than with its exclusive meaning) so that when used for example to connect a list of elements, the term "or" refers to one, some or all elements in the list.

[0196] Unless specifically stated otherwise, disjunctive language such as phrases "at least one of X, Y, and Z," "at least one of X, Y, or Z," "one or more of X, Y, and Z," and "one or more of X, Y, or Z" are to be otherwise understood in conjunction with the context generally used to present that an item, term, etc. may be X, Y, or Z, or any combination thereof (e.g., X, Y, and / or Z). Thus, such disjunctive language is generally not intended to, and should not, imply that certain embodiments require that at least one X, at least one Y, or at least one Z each be present.

[0197] Unless expressly stated otherwise, articles such as "a" or "an" should generally be interpreted as including one or more of the stated items. Thus, phrases such as "a device configured to..." are intended to include one or more of the stated devices. Such one or more of the stated devices may also be collectively configured to perform the stated statements. For example, "a processor configured to perform statements A, B, and C" may include a first processor configured to perform statement A working in conjunction with a second processor configured to perform statements B and C.

[0198] The drawings are not necessarily drawn to scale, and dimensions, shapes, and sizes of features may vary significantly from how they are depicted in the drawings.

[0199] Although the above detailed description has shown, described and pointed out the novel features applied to various embodiments, it is understood that various omissions, substitutions and changes may be made to the form and details of the illustrated devices or algorithms without departing from the essence of the present disclosure. It is recognized that certain embodiments described herein may be embodied in a form that does not provide all the features and benefits described herein, because some features can be used or practiced separately from other features. The scope of the specific embodiments disclosed herein is indicated by the appended claims rather than the foregoing description. All changes within the meaning and scope of the equivalent of the claims are included within their scope. Therefore, this specification and drawings are to be considered in an illustrative sense rather than a restrictive sense.

Claims

1. A device for molecular detection, the device comprising: at least one fluid channel configured to receive molecules to be detected, wherein at least some of the molecules to be detected are labeled with magnetic nanoparticles (MNPs); a sensor comprising a spin torque oscillator (STO) and encapsulated by a material separating the sensor from the at least one fluidic channel, wherein a surface of the material provides binding sites for the molecule to be detected; and a detection circuit coupled to the sensor and configured to detect a frequency or frequency noise of a radio frequency (RF) signal generated by the spin torque oscillator in response to the presence or absence of at least one magnetic nanoparticle coupled to one or more binding sites associated with the sensor, wherein the detection circuit comprises a phase-locked loop (PLL) configured to provide an error signal output corresponding to the frequency noise of the radio frequency signal generated by the spin torque oscillator in response to the presence of the at least one magnetic nanoparticle coupled to the one or more binding sites associated with the sensor.

2. The device according to claim 1, wherein: The detection circuit is configured to detect the frequency or frequency noise of the radio frequency signal generated by the spin torque oscillator in response to the presence or absence of the at least one magnetic nanoparticle coupled to the one or more binding sites associated with the sensor in part by applying a DC current to the spin torque oscillator.

3. The apparatus according to claim 1, wherein: The molecules to be detected include a first type of molecules and a second type of molecules, the first type of molecules are labeled by a first type of magnetic nanoparticles, and the second type of molecules are labeled by a second type of magnetic nanoparticles, The frequency or frequency noise of the radio frequency signal generated by the spin torque oscillator is a first frequency or frequency noise responsive to the presence of the first magnetic nanoparticle type, or a second frequency or frequency noise responsive to the presence of the second magnetic nanoparticle type, and The detection circuit is configured to discriminate between the first frequency or frequency noise and the second frequency or frequency noise to distinguish between the first type of molecules and the second type of molecules.

4. The device according to claim 1, wherein: The phase-locked loop comprises: a loop filter; and Low pass filter.

5. The device according to claim 4, wherein: The loop filter includes an amplifier and at least one resistor.

6. The device according to claim 4, wherein: The detection circuit also includes: a mixer having a first input, a second input, and an output; and a reference oscillator having an input coupled to a tuning input from the phase locked loop and an output coupled to the first input of the mixer, wherein the spin torque oscillator is coupled to the second input of the mixer and the output of the mixer is coupled to an input of the low pass filter.

7. The device according to claim 1, wherein: The spin torque oscillator includes a pinned layer, a free layer, and a spacer disposed between the pinned layer and the free layer.

8. The device according to claim 7, wherein: The pinned layer includes one or more ferromagnetic (FM) layers.

9. The device according to claim 8, wherein: The one or more ferromagnetic layers are first one or more ferromagnetic layers, and wherein the free layer includes a second one or more ferromagnetic layers.

10. The device according to claim 9, wherein: The spacer includes an insulating layer or a metal layer.

11. The device according to claim 7, wherein: In a quiescent state of magnetization, the magnetic moment of the free layer is oriented collinearly with the magnetic moment of the pinned layer.

12. The device according to claim 7, wherein: In the quiescent state of magnetization, the magnetic moment of the free layer is oriented parallel or antiparallel to the magnetic moment of the pinned layer.

13. The apparatus according to claim 7, wherein: In a quiescent state of magnetization, the magnetic moment of the free layer is oriented at an angle to the magnetic moment of the pinned layer, wherein the angle is between 20 degrees and 60 degrees.

14. A method for nucleic acid sequencing using an apparatus comprising a plurality of spin torque oscillators (STOs) and at least one fluid channel, the method comprising: Labeling nucleotide precursors with magnetic nanoparticles (MNPs); adding the labeled nucleotide precursor to the at least one fluidic channel of the device; detecting a frequency or frequency noise of a radio frequency (RF) signal generated by at least one of the plurality of spin torque oscillators; as well as determining whether a labeled nucleotide precursor has been detected based at least in part on a detected frequency or frequency noise of the radio frequency signal generated by the at least one spin torque oscillator of the plurality of spin torque oscillators, Wherein, detecting the frequency or frequency noise of the radio frequency signal generated by at least one spin torque oscillator among the plurality of spin torque oscillators comprises: monitoring an error signal of a detection circuit comprising a phase locked loop.

15. The method according to claim 14, further comprising: Prior to adding the labeled nucleotide precursors to the fluidic channel of the device, at least one nucleic acid strand is bound to a binding site in the fluidic channel, and an extendable primer and a plurality of nucleic acid polymerase molecules are added to the fluidic channel.

16. The method according to claim 14, further comprising: In response to determining that the labeled nucleotide precursor has been detected, the identity of the nucleotide precursor, or the identity of the base that is complementary to the labeled nucleotide precursor, is recorded.

17. A system for performing nucleic acid sequencing, the system comprising: Multiple spin torque oscillators (STOs); at least one fluid channel; a device for labeling nucleotide precursors with magnetic nanoparticles (MNPs); means for adding labeled nucleotide precursors to the at least one fluidic channel; means for detecting a frequency or frequency noise of a radio frequency (RF) signal generated by at least one of the plurality of spin torque oscillators; means for determining whether a labeled nucleotide precursor has been detected based at least in part on a detected frequency or frequency noise of the radio frequency signal generated by the at least one spin torque oscillator of the plurality of spin torque oscillators; as well as a phase locked loop (PLL) configured to provide an error signal output corresponding to the frequency noise of the radio frequency signal generated by the spin torque oscillator in response to the presence of the at least one magnetic nanoparticle coupled to one or more binding sites associated with the spin torque oscillator.

18. The system of claim 17, further comprising: means for binding at least one nucleic acid strand to a binding site in said fluid channel; and Means for adding an extendable primer and a plurality of nucleic acid polymerase molecules to the fluidic channel.

19. The system of claim 18, further comprising: Means for recording the identity of the nucleotide precursor, the identity of the base complementary to the labeled nucleotide precursor, or both the identity of the nucleotide precursor and the identity of the base complementary to the labeled nucleotide precursor.

Images