Three-dimensional photoelectric synergistic enzymatic gene synthesis chip and method

Abstract

The application discloses a three-dimensional photoelectric collaborative enzymatic gene synthesis chip and method, and belongs to the cross field of semiconductor synthetic biology and integrated circuit design. The chip integrates four types of heterogeneous chiplets, i.e. optical chiplets, biological chiplets and electrical chiplets, through three-dimensional vertical stacking packaging by means of chiplet technology, and is integrated in a closed three-dimensional shell. The chip drives enzymatic gene synthesis through a photoelectric collaborative mechanism, and realizes in-situ detection and real-time error correction of the synthesis process with the aid of on-chip fluorescence sensing, thereby constructing a closed-loop control of the on-chip biological gene synthesis process. The application has the advantages of high synthesis precision, large flux, low cost and environmental friendliness, and can be widely applied to the fields of gene synthesis, DNA storage and synthetic biology research.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of semiconductor synthetic biology and integrated circuit design, specifically relating to a three-dimensional photoelectric synergistic enzymatic gene synthesis chip and method. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Enzymatic DNA Synthesis (EDS) uses engineered enzymes such as terminal deoxynucleotidyl transferase (TdT) as core catalysts to achieve stepwise extension of deoxyribonucleotide (DNA) chains under mild aqueous conditions. It has advantages such as mild reaction conditions, low error rate and environmental friendliness, and is an important direction to overcome the limitations of chemical synthesis.

[0004] Existing enzymatic synthesis schemes are mostly solution-phase batch reactions or microfluidic channel architectures, with reaction volumes still in the micro-level. It is difficult to achieve a balance between spatial resolution and parallel throughput, and cannot meet the needs of large-scale, high-precision, and low-cost gene synthesis.

[0005] Some existing literature has proposed immobilizing terminal deoxynucleotidyl transferases and DNA primers / probes on a chip to achieve base extension synthesis via enzymatic methods, or incorporating 3'-O-reversibly blocked deoxynucleoside triphosphates into the growth chain using different TdT variants to achieve controlled stepwise synthesis. However, in the above schemes, the photocontrolled deprotection unit and the microfluidic reaction unit are spatially independent, and the temporal coordination between the functional units depends on an external control system. This results in the inability to achieve spatiotemporal synchronization and coordination of the three functional units—photocontrolled deprotection, enzymatic base extension, and fluorescence detection—within a single chip package.

[0006] Meanwhile, in existing solutions, the synthesis process and quality inspection are separated in physical space and time. Synthesis errors can only be detected at the end point through off-chip offline detection methods, making it impossible to build a real-time closed-loop control loop. In addition, optoelectronic devices, microfluidic reaction units, single-photon detectors and semiconductor logic controllers belong to different material systems and process platforms, making it difficult to achieve heterogeneous vertical integration and precise optical axis alignment within a single package. Summary of the Invention

[0007] To address the aforementioned problems, this invention proposes a three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip and method. This invention integrates semiconductor integrated circuit design technology, microfluidic technology, micro-electro-mechanical system (MEMS) technology, and biosynthesis technology to construct a heterogeneous 3D stacked photoelectric co-catalyzed enzymatic gene synthesis chip, which helps to achieve high-throughput, high-quality, low-cost, and environmentally friendly gene synthesis.

[0008] According to some embodiments, the present invention adopts the following technical solution: In a first aspect, the present invention provides a three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip, comprising: The light-controlled chip contains multiple light-emitting sources arranged in an array. The light-emitting sources are used to emit excitation light to the target microreactor according to a preset addressing signal, triggering the removal of the protecting groups in the target microreactor. The gene synthesis core contains multiple microreactors arranged in an array, each microreactor being used to immobilize terminal deoxynucleotidyl transferase or its engineered variant to catalyze a base elongation reaction after the protection group is removed; The fluorescence sensing control chip includes multiple photodetectors arranged in an array, used to detect the fluorescence signal of the base extension reaction in the target microreactor after the triggering of the control chip is turned off, and to obtain the fluorescence detection result; The servo chip is communicatively connected to the optical control chip, the gene synthesis chip, and the fluorescence sensing control chip, respectively, and is used to control the timing of the actions of the optical control chip, the gene synthesis chip, and the fluorescence sensing control chip according to the closed-loop logic of gene synthesis, detection, and error correction. The light control chip, the gene synthesis chip, the fluorescence sensing control chip, and the servo chip are vertically stacked and packaged together, and the array formed by the light source, the microreactor, and the photodetector is aligned with each other in the vertical direction to form multiple independent units.

[0009] As an alternative implementation, the optical control chip includes addressing chips, light source driving chips, light source array chips, and focusing microlens array chips arranged vertically from top to bottom, wherein: The addressing core is communicatively connected to the servo core and is used to receive the addressing command sent by the servo core, convert the two-dimensional coordinates of the target microreactor into a gating signal, and perform pixel-level address decoding on the light source corresponding to the target microreactor. The light source driving chip is signal-connected to the addressing chip and is used to drive the corresponding light source to output excitation light according to the received light pulse control command and the gating signal. The light source array core is electrically connected to the light source driving core, and is used to emit excitation light with a predetermined peak wavelength; The focusing microlens array core is located below the light source array core, and is used to focus the excitation light onto the central region of the target microreactor and suppress optical crosstalk between adjacent microreactors.

[0010] As a further defined implementation, the addressing core supports four addressing modes: global scan, row parallel, column parallel, and independent decoding of any unit.

[0011] As a further defined embodiment, the light source driving chip is a μLED driving chip, which integrates multiple independent constant current driving units. Each constant current driving unit includes a digital-to-analog converter, a constant current source, and a gating switch. The digital-to-analog converter is data-connected to the addressing chip and is used to output a reference voltage according to the light intensity level information. The constant current source is electrically connected to the digital-to-analog converter and is used to convert the reference voltage into a driving current. The gating switch is controlled to be turned on or off by the gating signal.

[0012] As a further defined implementation, the adjustment range of the drive current is 0.1mA to 10mA, and the current rise time and fall time do not exceed predetermined values.

[0013] As a further defined implementation, the μLED driver chip also integrates an overcurrent protection circuit, a thermal shutdown circuit, and an electrostatic discharge (ESD) protection circuit. The overcurrent protection circuit provides overcurrent protection, the thermal shutdown circuit executes a thermal shutdown mechanism, and the ESD protection circuit clamps abnormal voltages.

[0014] As a further defined embodiment, the substrate material of the focusing microlens array core is quartz glass or sapphire.

[0015] As an alternative implementation, the gene synthesis core comprises a microfluidic channel layer, an upper sealing layer, a microreactor array layer, a lower sealing layer, and a product collection-waste treatment-temperature control layer, arranged vertically from top to bottom, wherein: The microfluidic channel layer includes several reagent supply channels, enzyme channels, rinsing solution channels, and bubble discharge channels; The upper sealing layer is provided with a reagent injection port, an enzyme injection port, a rinsing solution injection port and a bubble discharge port. The reagent injection port is connected to the reagent supply channel, the enzyme injection port is connected to the enzyme channel, the rinsing solution injection port is connected to the rinsing solution channel, and the bubble discharge port is connected to the bubble discharge channel. The lower sealing layer is provided with a waste liquid discharge port and a product collection port; The product collection-waste liquid treatment-temperature control layer includes a waste liquid collection channel, a product collection pool, and a temperature control module. The waste liquid collection channel is connected to the waste liquid outlet, the product collection pool is connected to the product collection port, and the temperature control module is used to maintain the reaction temperature in the micro-reaction pool within a predetermined range.

[0016] As a further defined embodiment, the spacing between adjacent micro-reaction cells in the micro-reaction cell array layer is 20 μm to 50 μm, and the depth of the micro-reaction cell is 5 μm to 10 μm; a high-reflectivity metal thin film is deposited on the inner wall surface of the micro-reaction cell and covered with a protective layer, so that the reflectivity of the wall surface to the excitation light is not less than a set percentage, and the optical crosstalk between adjacent micro-reaction cells does not exceed a set value. The water contact angle on the inner wall surface of the microreactor does not exceed 30°; The microreactor has a primer anchoring point at the bottom center for covalently fixing or affinity capturing deoxyribonucleic acid primers.

[0017] As a further defined embodiment, the temperature control module includes a thin-film thermal sensor array and a microfluidic coolant channel. The thin-film thermal sensor array is distributed on the back of the product collection-waste liquid treatment-temperature control layer and is used to collect reaction temperature information. The microfluidic coolant channel contains circulating temperature control liquid. The microfluidic coolant channel cooperates with the thin-film thermal sensor array to regulate the reaction temperature by using PID closed-loop control of the circulating temperature control liquid.

[0018] As an alternative implementation, the fluorescence sensing and control chip includes a fluorescence filter lens layer, a photodetector array layer, an addressing control layer, and a data processing layer stacked vertically from top to bottom, wherein: The fluorescent filter lens layer is located above the photodetector array layer and is used to focus the fluorescence emitted by the microreactor onto the photodetector photosensitive area and filter out excitation light scattering and ambient stray light. The photoelectric detection array layer is optically coupled to the fluorescent filter lens layer, and includes multiple detection modules, with one detection module corresponding to the area directly below each microreactor. The addressing control layer is located below the photoelectric detection array layer and integrates a photoelectric detection address decoding module and a Geiger-quenching circuit module. The photoelectric detection address decoding module is spatially mapped to the addressing chip of the optical control chip. During the illumination of the light source, the Geiger-quenching circuit module controls the corresponding detection module to be in a quenched state. After the light source is turned off, the Geiger-quenching circuit module controls the corresponding detection module to switch to Geiger mode for single-photon fluorescence detection. The data processing layer, located below the addressing control layer, is used to perform quality assessment on the single-photon fluorescence detection results and send the assessment results to the servo chip.

[0019] As a further defined embodiment, the fluorescent filter lens layer employs a four-color filter array, with the four quadrants corresponding to the characteristic wavelengths of four basic fluorescent dyes: A, T, C, and G. The photodetector array layer includes multiple detector modules arranged in an array. Each detector module includes 2×2 single-photon avalanche diodes (SPADs). Each SPAD corresponds to one of the four quadrants of the fluorescence filter lens layer. By analyzing the relative response intensity of the four SPADs, the type of base currently synthesized and extended is identified, thereby achieving in-situ base type verification.

[0020] As a further defined implementation, the data processing layer includes a Time to Digital Converter (TDC) module, a histogram processing module, and an inference analysis module, wherein: The time-to-digital conversion module is used to record the arrival timestamp of each photon event; The histogram processing module is used to perform statistical histogram analysis on the photon events detected in each microreactor during the synthesis cycle, and extract fluorescence intensity, photon count rate and fluorescence lifetime; The inference analysis module is used to evaluate the synthesis quality of each microreactor in real time based on the extracted data, using a preset threshold or a lightweight machine learning model. The evaluation results are sent to the servo chip via a communication interface.

[0021] As an alternative implementation, the servo chip includes a light-controlled chip servo module, a gene synthesis chip servo module, a fluorescence sensing and control chip servo module, a power management module, and an electrical signal interface module, wherein: The photocontrol particle servo module is communicatively connected to the photocontrol particle and is used to send corresponding addressing instructions and light pulse control instructions to the photocontrol particle according to the base type and target sequence information of the current synthesis cycle. After receiving the synthesis error information fed back by the fluorescence sensing photocontrol particle, it dynamically adjusts the photocontrol parameters at the error point. The gene synthesis core servo module communicates with an external pump valve through an external electrical signal interface, and is used to send control commands for reagent injection, rinsing solution injection, waste liquid discharge and product collection in a time sequence. The fluorescence sensing and control chip servo module is communicatively connected to the fluorescence sensing and control chip and is equipped with a quenching synchronization circuit, which is used to complete the global quenching of the photodetector within a predetermined time after the rising edge of the light source driving signal is triggered, and to configure the readout mode of the photodetector during the fluorescence detection stage. The power management module includes a multi-channel voltage regulator and a DC-DC converter, which are used to provide stable power supplies of multiple voltage levels to each chip and control the power-on and power-off of each chip according to a safe timing sequence. The electrical signal interface module is used to communicate with an external host computer, report the synthesis status in real time, and receive gene sequence data and timing control commands.

[0022] As an alternative implementation, the light control chip, gene synthesis chip, fluorescence sensing control chip and servo chip are three-dimensionally vertically stacked and packaged using through-silicon via (TSV) technology and microbump interconnect structure. The TSV aperture is 5μm to 10μm and the microbump diameter is 10μm to 20μm. The light source array of the light-control chip is manufactured using gallium nitride (GaN)-based technology; the photodetector array of the fluorescence sensing control chip is manufactured using silicon-based or gallium arsenide (GaAs)-based technology; the servo chip is manufactured using complementary metal-oxide-semiconductor (CMOS) logic technology; and the microreactor array layer of the gene synthesis chip is manufactured using semiconductor micromachining (MEMS) technology. The substrate material is selected from one of ceramic substrate, glass substrate or printed circuit board (PCB) substrate, and the encapsulation shell is hermetically sealed.

[0023] As an alternative implementation, the excitation light emitted by the light source is focused by a focusing microlens and then vertically downwards to illuminate the central region of the micro-reaction cell. The fluorescence signal generated by the base extension reaction in the micro-reaction cell passes vertically downwards through the transparent window at the bottom of the micro-reaction cell and the fluorescence filter lens layer and is then focused onto the photosensitive area of ​​the corresponding photodetector. The optical axis alignment error of the corresponding light source, micro-reaction cell and photodetector is within a set range. Each microfluidic channel is arranged in the horizontal plane and does not intersect with the vertical optical path; The liquid inlet is located on one side of the chip, and the liquid outlet is located on the other side of the chip; The communication interfaces are centrally located on the top side of the chip, while the power and ground interfaces are centrally located on the bottom side of the chip.

[0024] Secondly, the present invention provides a photoelectric co-catalyzed enzymatic gene synthesis method based on the above-mentioned chip, comprising the following steps: S1. Synchronization of optically controlled deprotection and quenching: The servo chip sends an addressing command to the optically controlled chip to control the light source array to emit excitation light to the target microreactor, triggering the removal of the protective groups in the target microreactor; the servo chip sends a quenching command to the fluorescence sensing and control chip, and the fluorescence sensing and control chip controls the corresponding photodetector to enter the quenching state, with the quenching duration covering the light pulse width and the subsequent afterglow decay period. S2, Enzymatic Elongation: After the protecting group is removed, the servo core controls the external pump valve to inject a reaction solution containing fluorescently labeled deoxynucleotides into the target microreactor, and the terminal deoxynucleotide transferase or its engineered variant catalyzes base elongation; S3. Rinsing Residue and Fluorescence Detection: After base extension is completed, the servo core controls the external pump valve to inject rinsing solution into the target microreactor to remove residual reagents; after rinsing, the servo core releases the quenching state of the photodetector, the photodetector switches to Geiger mode, and performs single-photon detection on the fluorescence signal in the target microreactor. The fluorescence sensing control core performs real-time quality assessment of the detection results. S4. Feedback Error Correction and Fluorescent Group Removal: The fluorescence sensing and control chip sends the quality assessment result to the servo chip. The servo chip triggers an error correction operation on the erroneous microreactor based on the quality assessment result. After the error correction is completed, the fluorescence background is eliminated, and the next base synthesis cycle begins.

[0025] As an alternative implementation, in step S1, the light pulse width of the light source is 1ms to 100ms, and the driving current is 0.1mA to 10mA; The addressing command includes the two-dimensional coordinates of the target microreactor, the optical pulse width, the optical pulse sequence mode, and the optical intensity level information; The addressing mode includes one of the following: global scan mode, row parallel mode, column parallel mode, and arbitrary unit independent decoding mode.

[0026] As an alternative implementation, in step S3, the photodetector array operates in Geiger mode, the time-to-digital conversion module of the data processing layer records the photon arrival timestamps at a time resolution of 10 ps to 50 ps, ​​the histogram processing module extracts fluorescence intensity, photon count rate and fluorescence lifetime by fitting a time-correlated single-photon count histogram, and the inference analysis module evaluates the synthesis quality of each microreactor in real time based on the extracted data by using fluorescence intensity threshold determination, base type identification and abnormal signal detection.

[0027] As an alternative implementation, in step S4, the error correction operation includes at least one of increasing the optical pulse width or driving current to enhance deprotection, repeating the base extension step, and marking the discarded microreactor. After all synthesis cycles are completed, the servo chip controls the gene synthesis chip to collect the product and export the target DNA product to the outside of the chip.

[0028] As an alternative implementation, the molecular structure of the fluorescently labeled deoxynucleotide (dNTP-NPPOC) includes: The 3'-O-photosensitive protecting group is used to reversibly block the 3'-hydroxyl group of deoxynucleoside triphosphate, which is selectively removed in the photocontrolled deprotection step to expose the free 3'-hydroxyl group; A base is covalently coupled to a characteristic fluorescent dye via an enzymatically cleavable linker arm, which remains stable under base extension reaction conditions and releases the fluorescent dye through photolytic cleavage during the fluorescent group excision step, while simultaneously restoring the free 3'-hydroxyl group.

[0029] Beneficial effects Compared with the prior art, the beneficial effects of the present invention are as follows: This invention implements on-chip unified scheduling of four stages—photo-controlled deprotection and quenching synchronization, enzymatic extension, residual washing and fluorescence detection, and feedback error correction and fluorescent group removal—through a time-series control state machine solidified within the servo chip. This eliminates the fundamental defects of existing technologies where the temporal coordination of each functional unit depends on external mechanical switching and cannot achieve spatiotemporal synchronization within a single package. It shortens the synthesis-detection-error correction closed-loop cycle from several hours in traditional offline sequencing to milliseconds, significantly improving the fidelity and effective throughput of long-chain gene synthesis.

[0030] This invention utilizes a built-in quenching synchronization circuit in the fluorescence sensing and control chip servo module to complete the global quenching of SPADs shortly after the rising edge of the μLED driving signal. This achieves precise mutual exclusion and linkage between the μLED array and the SPAD array in terms of timing, effectively preventing SPAD false triggering or detector saturation caused by 365nm excitation light. It ensures that the fluorescence detection signal-to-noise ratio is not lower than the set value, achieving single-photon level sensitivity. This overcomes the shortcomings of existing technologies where synthesis and detection are separated in physical space and time dimensions, and inability to achieve in-situ real-time detection. It allows the deprotection, extension, and quality inspection of each microreactor to be independently controlled on a microsecond-millisecond timescale, completely eliminating the edge effect and incomplete deprotection problems in chemical method chips, and realizing spatiotemporally specific single-base level precise synthesis.

[0031] This invention integrates a TDC module, a histogram processing module, and an inference analysis module into a data processing layer, forming an on-chip edge computing architecture. This enables real-time evaluation of the synthesis quality of each microreactor with a short response time. Furthermore, through the cooperation of a four-color filter array and a 2×2 SPAD detector module, in-situ verification of base types is completed simultaneously during the fluorescence detection stage, achieving sequencing upon synthesis. This eliminates the need for offline NGS sequencing to complete the entire quality monitoring process. Moreover, the data processing layer performs quality evaluation and error identification on-chip, and the servo chip triggers local error correction in real time, allowing for retrying or discarding. This reduces the error detection and correction cycle from several hours in traditional offline sequencing to milliseconds, significantly improving the fidelity and throughput of long-chain gene synthesis.

[0032] The microreactor of this invention has a volume in the picoliter range, and the amount of reagent used in a single base extension cycle is reduced by 3 to 6 orders of magnitude compared with traditional microreactor systems. Combined with the mild aqueous phase and reaction conditions of enzymatic synthesis, the synthesis cost and environmental impact are significantly reduced, achieving high-throughput, low-cost, and low-environmental-impact gene synthesis. At the same time, the reflective and hydrophilic treatments on the four walls of the microreactor can control the reflectivity to be no less than a predetermined value, the light crosstalk to be no more than a set percentage, and the water contact angle to be no more than 30°, further ensuring the independence and consistency of the reactions in each cell.

[0033] This invention independently manufactures GaN-based μLED arrays, MEMS microreactors, GaAs-based SPAD arrays, and CMOS logic controllers at their respective optimal process nodes, and then performs 3D vertical stacking and packaging via TSV through-silicon vias and μBump microbumps. The high bonding and alignment accuracy of each chip enables deep on-chip integration of heterogeneous devices, breaking through the performance bottleneck of homogeneous integration and avoiding the limitation of a single process node on multi-physics performance. At the same time, the vertical coaxial optical path design ensures that the optical path and liquid path do not intersect in three-dimensional space, maximizing the utilization of light energy.

[0034] The chip of this invention has reserved standardized fluid and electrical interfaces, which facilitates integration with external pumps, valves, temperature control and host computer systems, ensuring the long-term stability of the enzyme reaction microenvironment and providing an engineering foundation for the industrialization path from chip to complete synthesizer.

[0035] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0036] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0037] Figure 1A schematic diagram of a three-dimensional photoelectric co-enzymatic gene synthesis chip structure according to one embodiment; Figure 2 This is a schematic diagram of the structure of a light-controlled chip according to one embodiment; Figure 3 This is a functional schematic diagram of a light-controlled chip according to one embodiment; Figure 4 This is a schematic diagram of the structure of a gene synthesis core according to one embodiment; Figure 5 This is a functional schematic diagram of a gene synthesis core according to one embodiment; Figure 6 This is a schematic diagram of the structure of a fluorescence sensing and control chip according to one embodiment; Figure 7 This is a functional schematic diagram of a fluorescence sensing and control chip according to one embodiment. Figure 8 This is a schematic diagram of the structure of a servo chip according to one embodiment; Figure 9 This is a functional schematic diagram of a servo chip according to one embodiment; Figure 10 This is a schematic diagram of the optical path in a cross-section of a chip according to one embodiment; Figure 11 This is a top view of the optical path alignment of a μLED array-microreactor array-SPAD array according to one embodiment.

[0038] Among them, 1. Light control chip, 2. Gene synthesis chip, 3. Fluorescent sensing control chip, 4. Servo chip, 5. Substrate, 6. Addressing chip, 7. μLED driving chip, 8. μLED array chip, 9. Focusing microlens array chip, 10. Microfluidic channel layer, 11. Upper sealing layer of microreactor, 12. Microreactor array layer, 13. Lower sealing layer of microreactor, 14. Product collection-waste liquid treatment-temperature control layer, 15. Fluorescent filter lens layer, 16. SPAD array layer, 17. Addressing control layer, 18. Data processing layer; 21. Address decoding layer; 22. Gated power switch layer; 23. μLED array layer; 24. Lens layer; 211. Decoding module; 221. Gating switch; 231. μLED; 241. Microlens unit; 311A, dATP mixture secondary channel; 311B, dATP mixture tertiary channel; 311C, dATP mixture tertiary channel inlet; 312A, dTTP mixture secondary channel; 312B, dTTP mixture tertiary channel; 312C, dTTP mixture tertiary channel inlet; 313A, rinsing solution secondary channel; 313B, rinsing solution tertiary channel; 313C, rinsing solution tertiary channel inlet; 314A, dCTP mixture secondary channel; 314B, dC... TP mixture tertiary channel, 314C, dCTP mixture tertiary channel inlet, 315A, dGTP mixture secondary channel, 315B, dGTP mixture tertiary channel, 315C, dGTP mixture tertiary channel inlet, 316A, exhaust secondary channel, 316B, exhaust tertiary channel, 316C, exhaust channel tertiary channel inlet, 317A, TdT enzyme channel secondary channel, 317B, TdT enzyme channel tertiary channel, 317C, TdT enzyme channel tertiary channel inlet; 331. Microreactor cube; 332. Microreactor cavity; 341. Waste liquid outlet; 342. Product outlet; 343. Primer anchoring point. 351. Product collection port; 352. Coolant channel; 353. Product collection tank; 354. Product collection channel; 355. Waste liquid collection channel; 356. Waste liquid outlet. 411. Base A fluorescent lens, 412. Base T fluorescent lens, 413. Base C fluorescent lens, 414. Base G fluorescent lens, 421. SPAD array, 431. Geiger-quenching circuit module, 441. Inference analysis module. 521. Photonic control chip servo module; 522. Fluorescence sensing control chip servo module; 523. Microprocessor; 524. Temperature control unit; 525. Reagent dispensing station; 526. Product collection valve; 527. Host computer; 528. Electrical signal interface module; 529. Power management module; 530. Gene synthesis chip servo module. 611. μLED light source; 612. Condenser lens; 613. Microreactor; 614. Fluorescent filter; 615. SPAD fluorescence detection area. 621. μLED array - microreactor array - SPAD array; 622. μLED unit; 623. microreactor unit; 624. SPAD unit. Detailed Implementation

[0039] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0040] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0041] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0042] Where there is no conflict, the embodiments and features described in this application may be combined with each other.

[0043] Example 1 To enable those skilled in the art to better understand the solutions provided by this invention, the following definitions are provided first.

[0044] Chiplet: Refers to a silicon die with a specific set of functions and a standardized die-to-die interface.

[0045] Chip technology: This is a technology approach that uses advanced packaging technology to interconnect and integrate chips into a complete system.

[0046] Optoelectronic interconnect chip: Uses optoelectronics to transmit information and data between chips.

[0047] 3D stacking: refers to the state in which multiple layers of structure are vertically stacked and mixed and bonded in a three-dimensional space.

[0048] For ease of description, this embodiment uses a 100×100 microreactor array as an example, but it does not mean that the solution provided by this invention can only be used for microreactor arrays of this size, or that the microreactor array has only one size as described above. In other embodiments, the microreactor array can be changed to other sizes according to the gene synthesis requirements, such as 800×600, 3200×3200, etc., which will not be exhaustively listed here.

[0049] It should also be noted that all the accompanying drawings in this embodiment are schematic diagrams, which are only used to illustrate the position and structural relationship of each part, and are not intended to limit the actual size, proportion or appearance of the product. In order to ensure clarity, each drawing only shows a partial quantity (such as one unit) as an example, and not all parts are shown in the drawing.

[0050] The following is a detailed introduction.

[0051] A three-dimensional optoelectronic co-catalyzed enzymatic gene synthesis chip, through a time-sequential control state machine solidified within a servo chip, achieves spatiotemporal synchronization and coordination of the photocontrol chip, gene synthesis chip, and fluorescence sensing control chip within a single 3D stacked package. It constructs a multi-stage closed-loop control circuit of "photocontrol deprotection → quenching synchronization → enzymatic extension → rinsing residues → fluorescence detection → feedback error correction → fluorescent group removal", which solves the fundamental defects of existing enzymatic synthesis systems, such as the lack of temporal coordination among functional units and the separation of synthesis and detection in time and space.

[0052] In one implementation, reference Figure 1 The three-dimensional photoelectric synergistic enzymatic gene synthesis chip includes: The light-controlled chip 1 includes multiple light-emitting sources arranged in an array. The light-emitting sources are used to emit excitation light to the target microreactor 613 according to a preset addressing signal, thereby triggering the removal of the protecting groups in the target microreactor 613. Gene synthesis core 2 contains multiple microreaction pools 613 arranged in an array, each microreaction pool 613 being used to immobilize terminal deoxynucleotidyl transferase or its engineered variant to catalyze a base elongation reaction after the protection group is removed; The fluorescence sensing control chip 3 includes multiple photodetectors arranged in an array, which are used to detect the fluorescence signal of the base extension reaction in the target microreactor 613 after the triggering and closing of the light control chip 1, and obtain the fluorescence detection result. Servo chip 4 is communicatively connected to the light control chip 1, the gene synthesis chip 2 and the fluorescence sensing control chip 3 respectively, and is used to control the timing of the operation of the light control chip 1, the gene synthesis chip 2 and the fluorescence sensing control chip 3 according to the closed-loop logic of gene synthesis, detection and error correction. The light control chip 1, the gene synthesis chip 2, the fluorescence sensing control chip 3, and the servo chip 4 are vertically stacked and packaged together, and the array formed by the light source, the microreactor 613, and the photodetector is aligned with each other in the vertical direction to form multiple independent units.

[0053] In this embodiment, the light-controlled core 1 is located on the top layer of the stacked structure and is used to emit 365nm ultraviolet light to the target micro-reaction pool 613 in the gene synthesis core 2 to trigger the removal of the protecting group; the gene synthesis core 2 is located in the middle layer and includes a micro-reaction pool array and a microfluidic system. Each micro-reaction pool 613 is used to carry terminal deoxynucleotidyl transferase (TdT) or its engineered variant for enzymatic base extension; the fluorescence sensing and control core 3 is located below the gene synthesis core 2 and performs single-photon detection of the fluorescence signal in the micro-reaction pool 613 through a photodetector array; the servo core 4 is located at the bottom layer and has an internally solidified timing control state machine for unified scheduling of the above three cores.

[0054] Of course, in other embodiments, the excitation light can also be ultraviolet light of other wavelengths, such as 380nm ultraviolet light, or blue light, such as 450nm blue light, etc., which will not be listed here.

[0055] The chip in this embodiment is a heterogeneous chip architecture, which is integrated into a closed three-dimensional shell through 3D stacking packaging technology. The closed three-dimensional shell in this embodiment is made of ceramic, glass or PCB substrate. The shell is hermetically sealed to prevent external contaminants from entering and to ensure the long-term stability of the enzyme reaction microenvironment.

[0056] The enclosed three-dimensional housing has reserved standardized fluid interfaces (such as liquid input on the right side, liquid output and product output on the left side) and electrical interfaces (such as IO communication on the upper side, power supply and ground on the lower side) to facilitate integration with external precision pumps and valves, temperature control equipment and host computer control systems.

[0057] In some implementations, refer to Figure 2 and Figure 3 The light control chip 1 consists of four vertically stacked layers from top to bottom: addressing chip 6, μLED driver chip 7, μLED array chip 8, and focusing microlens array chip 9, which correspond to address decoding layer 21, gated power switch layer 22, μLED array layer 23, and lens layer 24, respectively.

[0058] Addressing core 6 includes a decoding module 211, which integrates a row and column decoder, an address latch, and an instruction parsing unit. It communicates with servo core 4 through a high-speed serial communication interface (such as SPI or I²C serial interface) and receives addressing instruction frames sent by servo core 4.

[0059] In this embodiment, the data structure of the addressing instruction frame includes the two-dimensional coordinates of the target microreactor 613 (8 bits for X and Y, which are 8-bit binary addresses, covering the addressing space of 0–99, covering the corresponding 100×100 microreactor array), the optical pulse width (resolution 1ms, maximum 100ms), the optical pulse sequence mode (defining single pulse, double pulse, continuous pulse and custom sequence), and the light intensity level information (maximum 100 levels).

[0060] The row and column decoder inside the addressing chip 6 converts the two-dimensional coordinates into word line (WL) and bit line (BL) gating signals, and realizes pixel-level addressing of the μLED231 directly above the specified microreactor 613 through the intersection of word line / bit line. In this embodiment, the addressing response time does not exceed 10μs.

[0061] The decoding method is switched via the optical pulse sequence mode selection bit in the instruction frame, supporting the following four configurable modes: Global scan mode illuminates all 10,000 μLED231 one by one for chip initialization calibration or global synchronization deprotection; In parallel mode, all 100 μLED231 in a row are lit simultaneously, which is suitable for batch operations of base extension in the same row of microreactors 613. In parallel column mode, all 100 μLED231 in the column are lit simultaneously, which is suitable for batch operations of base extension in the same column of microreactor 613. The arbitrary unit independent decoding mode can precisely illuminate a single μLED231, achieving space-specific photocontrolled deprotection at the single-base level, enabling large-scale parallel gene synthesis of arbitrary sequences, or error correction of individual reaction cells.

[0062] In this embodiment, the addressing space is 100×100, with a total of 10,000 independently addressable pixels, corresponding to the 10,000 microreactors 613 below. Of course, the addressing space can be expanded to 64K, 1024K, etc.

[0063] In this embodiment, the μLED driver chip 7 integrates 10,000 independent constant current drive units, and each drive unit corresponds one-to-one with the word line / bit line strobe signal output by the addressing chip 6.

[0064] Each drive unit includes: an 8-bit digital-to-analog converter (DAC) that receives light intensity level information from the addressed chip 6 and outputs a precisely adjustable reference voltage; A constant current source consisting of an operational amplifier and a power transistor converts a reference voltage into a drive current; And a gating switch 221 controlled by word line / bit line signals, which is turned on only during addressing.

[0065] In this embodiment, the driving current output by the digital-to-analog converter is adjustable from 0.1mA to 10mA, with a full-range accuracy of ±1%. The current rise time and fall time are both less than 50ns, which makes the edge of the μLED231 light pulse steep, which is beneficial for achieving precise timing alignment with the SPAD quenching circuit.

[0066] The gating switch 221 can be implemented using a gated circuit.

[0067] In some embodiments, the μLED driver chip 7 also integrates an overcurrent protection circuit, a thermal shutdown circuit, and an electrostatic discharge (ESD) protection circuit. The overcurrent protection circuit provides overcurrent protection (e.g., automatic shutdown when a single current exceeds 12mA), the thermal shutdown circuit provides a thermal shutdown mechanism or function (e.g., globally shutting down the driver output when the chip junction temperature exceeds 125°C), and the ESD protection circuit provides ESD protection to ensure the long-term reliability of the μLED array.

[0068] The μLED array chip 8 is manufactured using GaN-based epitaxial wafer technology. A miniature ultraviolet light-emitting diode (μLED231) is formed on the epitaxial wafer through photolithography, etching, and metallization processes, with a peak emission wavelength of 365nm±5nm. The pixel arrangement of the μLED array corresponds one-to-one with the microreactor array below, ensuring that each microreactor 613 is illuminated by a beam of ultraviolet light pointing vertically downwards.

[0069] The μLED array comprises multiple μLEDs 231 (or μLED units 622).

[0070] In this embodiment, 10,000 miniature ultraviolet light-emitting diodes (μLEDs) can be formed using photolithography, dry etching (ICP-RIE), and Ti / Al / Ni / Au multilayer metallization processes. Each μLED 231 has an effective light-emitting area of ​​50μm × 50μm, a pixel pitch of 200μm, and is arranged in a 100 × 100 array, with a total area of ​​20mm × 20mm. The peak emission wavelength of the μLED 231 is 365nm ± 5nm (UV-A band), and the single-LED light output power is not less than 0.5mW under a 10mA drive current.

[0071] Of course, in other embodiments, the micro LED 231 can be replaced with other micro light sources depending on the specific circumstances, and these will not be exhaustively listed here. This is something that those skilled in the art would readily conceive of, and should therefore fall within the scope of protection of this invention.

[0072] In some embodiments, the focusing microlens array core 9 uses a quartz glass (JGS1) or sapphire (Al2O3) substrate, and forms 10,000 microlens units 241 through photolithography, reactive ion etching (RIE) or grayscale photolithography. Each microlens unit 241 is located directly below the corresponding μLED 231 and is coaxially aligned with the light-emitting area of ​​the μLED 231 to ensure a predetermined alignment accuracy (such as a deviation of no more than 1 μm).

[0073] In this embodiment, the focal length of the microlens is designed to be 50μm to 60μm, the focal plane is located near the top surface of the reaction liquid layer of the microreaction cell 613, and the light spot covers the central main area of ​​the 150μm×150μm microreaction cell 613. The optical crosstalk between adjacent microreaction cells 613 does not exceed 5%, preventing unintended photochemical deprotection reactions from occurring in non-target microreaction cells 613.

[0074] Based on the above optical path parameters, the photochemical deprotection energy threshold for the p-nitrophenylpropoxycarbonyl (NPPOC) protecting group is approximately 50 mJ / cm². Under the condition of a target deprotection time of 100 ms, the required optical power density of the reaction solution is 500 mW / cm², corresponding to a minimum required optical power of approximately 0.048 mW for the corresponding spot size. Considering an 80% optical path efficiency, the minimum required optical power at the μLED231 output is approximately 0.060 mW. It is known that the μLED231's light output is not less than 0.5 mW under 10 mA driving, and that the light output is approximately linear with the driving current in the low current range. Therefore, the minimum driving current required to meet the 100 ms deprotection time is approximately 1.4 mA. The recommended operating current range is 2 mA to 5 mA, corresponding to a deprotection time controllable between 50 ms and 100 ms.

[0075] The specific calculation process is as follows: Under the condition that the target deprotection time is 100ms, the required optical power density of the reaction solution is: P_min = 50mJ / cm² ÷ 100ms = 500mW / cm² = 5mW / mm²; The minimum optical power required to produce the corresponding light spot (spot area of ​​0.0095 mm² after focusing by the lens) is: E_min = 5mW / mm² × 0.0095mm² ≈ 0.048mW; Considering optical path efficiency (taken as 80%), the minimum optical power required at the output of μLED231 is: P_LED_min = 0.048mW ÷ 0.80 ≈ 0.060mW; Given that the μLED231 has an output power ≥ 0.5mW under 10mA drive, and assuming that the light output in the low current range is approximately linear with the drive current (with an engineering margin factor of 1.2), the minimum drive current required to satisfy the 100ms off-protection time is: I_min ≈ (0.060mW ÷ 0.5mW) × 10mA × 1.2 ≈ 1.4mA; Similarly, under the condition that the target's unprotection time is 50ms: P_min = 50mJ / cm² ÷ 50ms = 1,000mW / cm² = 10mW / mm²; The corresponding minimum drive current is approximately 2.8mA.

[0076] In summary, the minimum operating current of the μLED driver chip in this embodiment is set to 1.5mA, and the recommended operating current range is 2mA–5mA. Within this range, the deprotection time can be controlled within 50ms–100ms, while allowing sufficient process tolerance and temperature drift margin.

[0077] In some implementations, refer to Figure 4 and Figure 5 The gene synthesis core 2 is vertically stacked from a microfluidic channel layer 10, an upper sealing layer 11 for the microreactor, an array layer 12 for the microreactor, a lower sealing layer 13 for the microreactor, and a product collection-waste liquid treatment-temperature control layer 14.

[0078] The microfluidic channel layer 10 integrates a reagent supply channel, a TdT enzyme channel, a rinsing solution channel, and a bubble discharge channel. In this embodiment, the reagent supply channel has four main supply channels, which are connected to external micro-precision pump valves through fluid interfaces set on the side of the chip. These channels are used to pump four base mixtures (dATP, dTTP, dCTP, and dGTP in this embodiment, each infused with a different fluorescent label) into the microfluidic channel layer 10 in a preset sequence.

[0079] The TdT enzyme channel and the rinsing solution channel also have corresponding main supply channels, which are connected to external micro-precision pump valves through fluid interfaces set on the side of the chip, respectively, for pre-adding and solidifying the TdT enzyme, and for pumping the rinsing solution into the microfluidic channel layer 10 according to a preset time sequence.

[0080] In this embodiment, each channel includes a main liquid supply channel and a branch distribution channel. The main liquid supply channel is connected to an external micro-precision pump valve through a fluid interface on the side of the chip to allow the corresponding solution / substance to flow into the chip. The branch distribution channel is connected to the corresponding main liquid supply channel at a certain angle, such as vertically, to introduce the solution / substance in the main liquid supply channel into each micro-reaction cell 613.

[0081] In some embodiments, the main liquid supply channel has a cross-sectional width of 100 μm to 200 μm, a depth of 50 μm to 100 μm, and a total length of 10 mm to 15 mm. The branch distribution channels adopt a tree-like (H-type) or grid-like structure design to evenly distribute the fluid from the main liquid supply channel to the top of 100×100 micro-reaction cells 613, with the inlet flow deviation of each micro-reaction cell 613 not exceeding 5%.

[0082] In this embodiment, the sealing layer 11 of the microreactor of each unit has 7 openings, including 4 reagent injection ports, namely dATP mixture tertiary channel inlet 311C, dTTP mixture tertiary channel inlet 312C, dCTP mixture tertiary channel inlet 314C and dGTP mixture tertiary channel inlet 315C.

[0083] Among them, the dATP mixture tertiary channel inlet 311C corresponds to the dATP mixture tertiary channel 311B, and the dATP mixture tertiary channel 311B is connected to the dATP mixture secondary channel 311A; The dTTP mixture tertiary channel inlet 312C corresponds to the dTTP mixture tertiary channel 312B, and the dTTP mixture tertiary channel 312B is connected to the dTTP mixture secondary channel 312A; The dCTP mixture tertiary channel inlet 314C corresponds to the dCTP mixture tertiary channel 314B, and the dCTP mixture tertiary channel 314B is connected to the dCTP mixture secondary channel 314A; The dGTP mixture tertiary channel inlet 315C corresponds to the dGTP mixture tertiary channel 315B, and the dGTP mixture tertiary channel 315B is connected to the dGTP mixture secondary channel 315A.

[0084] The opening also includes a TdT enzyme injection port (in this embodiment, the inlet 317C of the tertiary channel of the TdT enzyme channel), a rinsing solution injection port (in this embodiment, the inlet 313C of the tertiary channel of the rinsing solution), and a bubble outlet port (in this embodiment, the inlet 316C of the tertiary channel of the exhaust channel). The TdT enzyme injection port is connected to the tertiary channel 317B of the TdT enzyme channel, the tertiary channel 317B of the TdT enzyme channel is connected to the secondary channel 317A of the TdT enzyme channel, the inlet 313C of the rinsing solution is correspondingly connected to the tertiary channel 313B of the rinsing solution, the tertiary channel 313B of the rinsing solution is connected to the secondary channel 313A of the rinsing solution, the inlet 316C of the tertiary channel of the exhaust channel is correspondingly connected to the tertiary channel 316B of the exhaust channel, and the tertiary channel 316B of the exhaust channel is connected to the secondary channel 316A of the exhaust channel.

[0085] The aforementioned secondary channels are branch distribution channels, which are connected to the corresponding main liquid supply channels, and the main liquid supply channels serve as primary channels.

[0086] Each injection port has a diameter of 20μm to 50μm and is coaxially aligned with the top opening of the microreactor 613. The lower sealing layer 13 of the microreactor has two openings, including a waste liquid outlet 341 and a product outlet 342.

[0087] The micro-reaction cell array layer 12 contains 10,000 micro-reaction cells 613 (or micro-reaction cell units 623) of size 100×100. In this embodiment, each micro-reaction cell 613 includes a micro-reaction cell cube 331, within which is a rectangular cavity, namely a micro-reaction cell cavity 332. The cavity dimensions are 150μm×150μm, and the depth is 5μm. The single-cell volume V = 150μm×150μm×5μm = 112500μm³ = 112.5pL, which is at the picoliter level. The center-to-center spacing between the micro-reaction cells 613 is 200μm, which is strictly consistent with the pixel spacing of the μLED array above. The physical spacing between adjacent cells is 50μm, and the total array size is 20mm×20mm. The four corners of the cell are rounded with a radius of not less than 2μm to avoid liquid stagnation and bubble accumulation at right-angle corners.

[0088] The walls of the micro-reaction cell 613 undergo dual surface treatment, specifically including reflective treatment and hydrophilic treatment.

[0089] The reflective treatment involves depositing a thin film of highly reflective Al or Ag metal (50 nm to 100 nm thick) on the four wall surfaces using electron beam evaporation or sputtering processes, and covering it with a protective layer of SiO2 or Al2O3 (20 nm to 50 nm thick) to form a mirror-reflective structure. This ensures that the reflectivity of the four walls to 365 nm ultraviolet light is not less than 90%, which can reduce optical crosstalk between adjacent micro-reaction cells 613 to no more than 3% and prevent photochemical deprotection reactions from occurring at non-target locations.

[0090] The hydrophilic treatment involves treating the reflective layer with O2 plasma (in this embodiment, the treatment power is 100W and the time is 60s) or modifying it with self-assembled monolayer (SAM) (e.g., APTES silanization treatment). This introduces hydrophilic groups such as hydroxyl (–OH) or carboxyl (–COOH) groups onto the surface of the pool wall, ensuring that the water contact angle does not exceed 30°. This guarantees that the reagent liquid spreads rapidly and is evenly distributed in the pool, reduces reagent residue, and improves the consistency of the base extension reaction.

[0091] Of course, in some embodiments, the shape of the micro-reaction cell 613 can be adjusted according to requirements or processing difficulty, such as changing it to a square cavity, a cylindrical cavity or an arc cavity, etc., which are not exhaustive here.

[0092] Each microreactor 613 has a primer anchoring point 343 at its central bottom region. In this embodiment, the anchoring point region is 40μm × 40μm in size and located at the geometric center of the bottom of the reactor. The primer anchoring point 343 is formed by locally depositing gold nanoparticles (particle size 5nm to 20nm, areal density not less than 10) at the bottom of the reactor. 11 / cm 2This can be achieved by modifying carboxyl / amino functionalized silane layers (e.g., APTES-Glutaraldehyde system) to covalently fix or affinity capture DNA primers, ensuring that the primers remain stably anchored during multiple enzymatic extension cycles.

[0093] The product collection-waste liquid treatment-temperature control layer 14 is located at the bottom of the gene synthesis core 2 and is equipped with a waste liquid collection channel 355, a product collection pool 353 and a temperature control channel.

[0094] The waste liquid collection channels 355 are arranged in a tree-like pattern, with a cross-sectional width of 50μm to 100μm and a depth of 30μm to 50μm. They are connected to the waste liquid outlet 341 at the bottom of each micro-reaction cell 613 via a controllable micro-valve and capillary micropores. An external negative pressure suction system (vacuum degree -10kPa to -30kPa) discharges the waste liquid through the waste liquid collection channels 355 to the waste liquid outlet 356.

[0095] After the completion of all synthesis cycles (typically 50–200 bases), the product collection pool 353 collects the target DNA product from each microreactor 613 in nanoliters. It is connected to the product collection port 351 via the product collection channel 354. The product collection port 351 corresponds to the product discharge port 342. The target DNA product in each microreactor 613 is finally released into the product collection pool 353. The product collection pool 353 is connected to the microfluidic channel outside the chip via the product collection channel 354 to realize the automatic extraction of the product.

[0096] The temperature control module consists of a thin-film thermal sensor array and a microfluidic coolant channel 352. The thin-film thermal sensor array is distributed on the back of the product collection-waste liquid treatment-temperature control layer 14, with a temperature measurement accuracy of ±0.05℃. The coolant channel 352 circulates temperature control liquid, and the temperature of the reaction liquid in the micro-reaction tank 613 is precisely maintained at 37℃±0.1℃ through PID closed-loop control.

[0097] After the μLED array of the light-control chip 1 and the micro-reaction cell array of the gene synthesis chip 2 are vertically stacked, the spatial deviation between the light-emitting center of each μLED231 and the center of the micro-reaction cell 613 directly below it does not exceed 1μm, which is guaranteed by the alignment marks in the 3D packaging process.

[0098] In some implementations, refer to Figure 6 and Figure 7 The fluorescence sensing and control chip 3 is composed of a fluorescence filter lens layer 15, a SPAD array layer 16, an addressing control layer 17, and a data processing layer 18 stacked vertically from top to bottom.

[0099] The fluorescence sensing control chip 3 is used for synthesis accuracy detection and error correction. The four-layer structure is interconnected through copper-copper hybrid bonding or micro-bumps to achieve high-density vertical electrical interconnection.

[0100] The SPAD array 421 of the sensing and control chip is strictly aligned with the micro-reaction cell array above. Each SPAD module (or SPAD unit 624) is located directly below the center of the corresponding micro-reaction cell 613, ensuring that the fluorescence optical axis is vertically aligned.

[0101] The fluorescent filter lens layer 15 is located above the SPAD array layer 16, closely attached to the bottom of the gene synthesis core 2. It uses a quartz glass (JGS1) or a weakly fluorescent background polymer substrate, integrating a fluorescent filter array and a microlens array. In terms of filtering function, it filters out scattering of 365nm ultraviolet light (i.e., excitation light) and ambient stray light, allowing only the fluorescence signal generated by the base extension reaction to pass through. Bandpass filters can be designed, with the wavelengths of the four base fluorescent dyes corresponding to the four center wavelengths of the filters.

[0102] For focusing, the microlens focuses the fluorescence emitted by the microreactor 613 onto the SPAD photosensitive area, improving light collection efficiency. The focal length of the microlens is designed to match the distance from the bottom surface of the reaction chamber to the SPAD photosensitive area. For the 2×2 SPAD module, the fluorescence filter lens layer 15 is further refined into a four-color filter array. The four quadrants correspond to the characteristic wavelengths of the four base fluorescent dyes A, T, C, and G, respectively, with a detection wavelength range of 450nm to 700nm, compatible with commonly used fluorescent dyes such as FAM, TAMRA, ROX, and Cy5. The base A fluorescent lens 411, base T fluorescent lens 412, base C fluorescent lens 413, and base G fluorescent lens 414 correspond to the four quadrants, and the microlenses focus the fluorescence emitted by the microreactor 613 onto the photosensitive area of ​​the single photon avalanche diode (SPAD), improving light collection efficiency.

[0103] The above scheme enables the SPAD module to not only detect whether the synthesis is successful, but also to identify the type of bases in the current synthesis extension by analyzing the relative response intensity of the four SPADs, thus achieving in-situ verification of "synthesis-by-sequencing" (SBS).

[0104] The SPAD array layer 16 is located below the fluorescence filter lens layer 15 and is fabricated using silicon-based or GaAs-based single-photon avalanche diode technology. Operating in Geiger mode, it possesses single-photon level sensitivity and is suitable for detecting weak fluorescence signals released during enzymatic base extension reactions. In this embodiment, the SPAD array 421 has a size of 200×200 pixels, a pixel pitch of 100μm, a single SPAD photosensitive area size of 75μm×75μm, and a fill factor of 56.25%.

[0105] Directly below each microreactor 613 is a detection module consisting of four SPAD tubes in a 2×2 configuration. After the detection signals from the four SPAD tubes are combined, the photon count is approximately linearly superimposed, while the shot noise only increases by the square root. The signal-to-noise ratio can be improved to about twice the original value. At the same time, multi-path parallel detection can alleviate the limitation of the dead time of a single SPAD.

[0106] In this embodiment, the key parameters of the SPAD array layer 16 are: photon detection efficiency of not less than 30% (@550nm), dark count rate of less than 100cps (25℃), time resolution of less than 100ps, breakdown voltage of 20V to 30V, and quench recovery time of less than 50ns. A precise alignment process is used between the SPAD array layer 16 and the gene synthesis core 2 above it to ensure that the optical axis of each SPAD photosensitive area coincides with the corresponding microreactor 613, with an alignment accuracy of no more than 1μm.

[0107] It is understood that in the above embodiments, the SPAD device is not limited to silicon-based or GaAs-based, but can also be InGaAs-based (suitable for near-infrared fluorescent dyes) or other wide bandgap semiconductor materials, and the detection wavelength can cover the range of 400nm to 1000nm.

[0108] Similarly, in some embodiments, SPAD devices may not be used, and other detection modules may be used instead; these will not be exhaustively listed here.

[0109] The addressing control layer 17 is located below the SPAD array layer 16 and is manufactured using standard CMOS logic technology (e.g., 40nm or 28nm node). It integrates a SPAD address decoding module and a Geiger-quenching circuit module 431. The SPAD address decoding module corresponds spatially to the addressing module in the optical control chip 1. When a micro-reaction cell 613 undergoes photoprotection and base extension, the decoding module 211 addresses and activates the corresponding SPAD module, realizing spatiotemporal synchronous readout of "reaction-detection". The readout rate is not less than 1MHz and supports three modes: row scanning, column scanning, and random addressing.

[0110] In some embodiments, when the addressing control layer 17 is in parallel readout mode, all 10,000 SPAD modules are activated simultaneously for chip initialization calibration, dark count rate uniformity testing, and global fluorescence background detection.

[0111] In row / column parallel readout mode, the entire row or column of SPAD modules is activated simultaneously, which is suitable for batch detection of microreactors 613 in the same row or column.

[0112] In random addressing mode, the target SPAD module is precisely activated, enabling spatially specific detection at the single-base level.

[0113] During the time window when the μLED array emits 365nm ultraviolet light for deprotection, the Geiger-quenching circuit module 431 forcibly biases the SPAD module below the corresponding microreactor 613 below the breakdown voltage, putting it in a quenched state. The timing alignment accuracy of the quenching synchronization does not exceed 10ns, and the quenching duration covers the entire light pulse width (e.g., 100ms) and the subsequent afterglow decay period (approximately 50ms). After the μLED array is turned off, the Geiger-quenching circuit module 431 controls the corresponding SPAD detection module to switch to Geiger mode for single-photon fluorescence detection.

[0114] The data processing layer 18 is located below the addressing control layer 17 and integrates the TDC module, histogram processing module and inference analysis module 441 to form an on-chip edge computing architecture.

[0115] The TDC module employs a vernier or tapped delay line architecture with a time resolution of 10 ps to 50 ps and a dynamic range of no less than 10 ns. It records the arrival timestamp of each photon event detected by the SPAD for fluorescence lifetime imaging (FLIM) analysis.

[0116] The histogram processing module performs statistical histogram analysis on the photon events detected in each microreactor 613 during the synthesis cycle. By fitting the time-correlated single-photon count (TCSPC) histogram, it extracts key parameters such as fluorescence intensity (the number of photons detected per unit time), photon count rate, fluorescence lifetime τ, and background noise level.

[0117] Fluorescence intensity is determined by the number of photons detected per unit time; the photon count rate is positively correlated with the doping concentration of fluorescently labeled dNTPs; fluorescence lifetime τ is obtained by fitting a time-correlated single-photon count histogram and is used to distinguish different fluorescent dyes; background noise level is calibrated using a blank cell without fluorescent labeling.

[0118] The reasoning and analysis module 441 performs real-time evaluation of the synthesis quality of each micro-reaction cell 613 based on a preset threshold or a lightweight machine learning model (such as a decision tree or a small neural network, which can be an existing model). The evaluation dimensions include: fluorescence intensity threshold determination, base type identification, and abnormal signal detection.

[0119] Specifically, in this embodiment, the fluorescence intensity threshold determination process is as follows: if the fluorescence signal is lower than the success threshold, the synthesis is determined to have failed.

[0120] The base type identification process involves analyzing the relative response intensity of the four quadrants of the 2×2 SPAD module to identify whether the type of the currently synthesized extended base is consistent with the expectation, thereby achieving in-situ base type verification.

[0121] The abnormal signal detection process includes: if the detection results show unexpected base fluorescence characteristics or abnormal signal fluctuations, it is determined to be an incorrect synthesis extension or cross-contamination.

[0122] In this embodiment, the response time of the data processing layer 18 is no more than 10ms to meet the requirements of real-time closed-loop control. The analysis results are sent to the servo core 4 through the communication interface to trigger the error correction process, including repeating the base extension step, locally re-deprotecting, or marking the pool as a waste pool.

[0123] In some implementations, refer to Figure 8 and Figure 9 The servo core 4 includes a light control core servo module 521, a gene synthesis core servo module 530, a fluorescence sensing control core servo module 522, a power management module 529, and an electrical signal interface module 528, all of which are integrated into a single-layer structure of the servo core 4.

[0124] Servo chip 4 is mainly used for: Based on the base type and target sequence information of the current synthesis cycle, precise μLED231 addressing and light pulse control commands are issued to the photocontrol chip 1; The chip communicates with external micro-precision pumps, valves, pressure sensors, and macro-temperature control units via an external electrical signal interface to achieve time-series control of reagent dispensing, waste liquid discharge, and product collection. It is responsible for the global quenching and activation timing scheduling of the SPAD array 421, and configures the readout mode during the fluorescence detection stage; Provides stable power supplies at multiple voltage levels for each functional chip; It interacts with the host computer 527 in real time through a high-speed serial communication interface, reports the synthesis status, and receives gene sequence data and timing control instructions.

[0125] Specifically, the photocontroller servo module 521 internally contains a gene synthesis timing control state machine, which is communicatively connected to the photocontroller 1. Based on the base type and target sequence information of the current synthesis cycle, a 16-bit addressing instruction is sent to the addressing chip 6 of the photocontroller 1 via the SPI / I²C bus. The high 8 bits identify the row address, and the low 8 bits identify the column address, enabling four addressing modes: single-point, row scan, column scan, and global scan. A light pulse control instruction is also sent to precisely specify the switching time, pulse width, and intensity level of the μLED231. In this embodiment, the μLED231 driving parameters in the light pulse control instruction include: peak wavelength 365nm±5nm, adjustable driving current 0.1mA–10mA, deprotection time 50ms–100ms, and adjustable pulse width 1ms–100ms.

[0126] The 521 light control chip servo module has a built-in DAC array, which independently configures the drive current level for each μLED231 pixel, with a total of 32 levels of quantization (5-bit resolution).

[0127] When receiving the synthesis error information fed back by the fluorescence sensing control chip 3, the light control chip servo module 521 dynamically adjusts the light control parameters at the error point, such as increasing the light pulse width to 100ms or increasing the drive current to 5mA to enhance the deprotection, skipping subsequent illumination of the abandoned microreactor 613 to save energy, and starting local adaptive compensation for uneven global illumination to achieve closed-loop error correction.

[0128] The gene synthesis core servo module 530 communicates with the precision pump valve, pressure sensor and temperature control unit 524 outside the chip through an external electrical signal interface to control the precision injection pump or electroosmotic pump. In some embodiments, the flow rate is 1 nL / min to 1 μL / min and the response time is less than 100 ms. The corresponding base reagent is pumped into the micro-reaction cell 613 according to the gene synthesis sequence. The reagent distribution is precisely adjusted by the servo core 4 to ensure that the inlet flow deviation of each cell does not exceed 5%.

[0129] After base extension is completed, the rinsing solution is injected and the waste liquid is discharged. The temperature of the microreactor 613 is adjusted to 37°C by the temperature control unit 524. After the entire synthesis cycle is completed, the gene synthesis chip servo module 530 controls the product collection valve 526 to open, exporting the target DNA product to the outside of the chip.

[0130] In some embodiments, the product collection pool 353 adopts a partitioned design, with each collection partition corresponding to a row of micro-reaction pools 613, and the volume is from nano-level to micro-level.

[0131] The fluorescence sensing and control chip servo module 522 is communicatively connected to the fluorescence sensing and control chip 3. It has a built-in quenching synchronization circuit. During the photo-controlled deprotection stage, it sends a global quenching command to the addressing control layer 17 to put the SPAD array 421 into a quenching state, so as to avoid the 365nm ultraviolet light emitted by μLED231 from triggering avalanche breakdown. During the base extension and fluorescence detection stage, it sends a region activation command and configures the readout mode of the SPAD array 421, which can adopt single-point readout, line scan readout or full-frame readout mode.

[0132] In this embodiment, the quenching synchronization circuit completes global SPAD quenching within 10 ns after being triggered by the rising edge of the μLED231 driving signal. The quenching duration covers the entire optical pulse width (100 ms) and the subsequent afterglow decay period (approximately 50 ms). After quenching is lifted, the fluorescence sensing and control chip servo module 522 configures the SPAD array 421 for fluorescence detection. The TDC records the photon arrival time with a time resolution of 10 ps, ​​and the success of base extension is determined by fluorescence lifetime analysis (FLIM).

[0133] The fluorescence sensing and control chip servo module 522 summarizes the fluorescence intensity, fluorescence lifetime, and photon count rate of each cell in real time and compares them with the gene library. If the fluorescence lifetime of a certain cell deviates from the normal value, it is determined that the synthesis of that cell is abnormal, and an error correction request is sent to the light control chip servo module 521 via the I²C bus to trigger local adaptive illumination compensation or mark the cell as abandoned.

[0134] The power management module 529 integrates multiple LDO low-dropout linear regulators and DC-DC converters to provide stable power supplies at multiple voltage levels for each chip: the μLED driver power supply is 3.3V to 5V; the SPAD high-voltage bias is 20V to 30V, provided by the ChargePump DC-DC converter, with an output ripple of no more than 10mV; the servo logic digital core power supply is 1.2V, and the I / O interface power supply is 1.8V, provided by the LDO regulators. The module also integrates power sequence control and power-on reset circuitry. The power-on sequence first enables the 1.2V digital power supply, then enables the 1.8V and 3.3V I / O power supplies, and finally enables the 5V μLED driver and the 20V to 30V SPAD high voltage, avoiding gate-locking effect and voltage overshoot; the power-down sequence is the reverse.

[0135] The electrical signal interface module 528 is located on the top side of the chip, integrating a high-speed serial communication interface and a low-speed control interface, and is managed uniformly by the microprocessor 523. The microprocessor 523 adopts a RISC-V architecture, integrates 4GB RAM and 4MB FLASH, has a main frequency of 100MHz, runs an embedded real-time operating system (RTOS), and is responsible for task scheduling, interrupt management, and closed-loop control algorithm execution.

[0136] In some embodiments, the high-speed interface uses USB 3.0 (5Gbps) for real-time transmission of gene sequence data, synthesis progress reports, and quality scores for each pool; the low-speed interface includes SPI (50MHz), I²C (400kHz), and GPIO (32 channels) for connecting external precision pumps and valves, temperature control unit 524, and sensor array. The I / O interface uses high-speed SerDes with a bandwidth greater than 10Gbps and a latency less than 1μs, meeting the high-speed data return requirements of the SPAD array 421. The electrical signal interface module 528 communicates with the external host computer 527, reagent dispensing station 525, and synthesis operation control console, reporting the gene synthesis status in real time, such as synthesis progress, quality scores for each pool, and error statistics, and receiving gene sequence data and timing control commands sent by the host computer 527.

[0137] In some implementations, refer to Figure 1 The 3D stacked packaging of the chip adopts a rectangular three-dimensional packaging form, and vertical interconnection between chips is achieved through TSV through silicon vias and μBump microbump technology.

[0138] The specific packaging process includes the following steps: Chip fabrication: Wafer-level fabrication and testing of the following chips were completed: light control chip 1 (GaN μLED array + CMOS driver / addressing), gene synthesis chip 2 (silicon-based MEMS microreactor + glass microfluidic channel network), fluorescence sensing control chip 3 (SPAD array + CMOS data processing), and servo chip 4 (CMOS logic + interface). Each chip was fabricated independently at its optimal process node.

[0139] TSV etching and filling: TSV vias are formed in the interconnection region of each core through deep reactive ion etching. The via diameter is 5μm to 10μm and the depth penetrates the core thickness. Vertical conductivity is achieved by filling with copper or tungsten electroplating. μBump fabrication: Tin-silver-copper (SAC) or copper pillar μBumps are fabricated on the core bonding surface by photolithography and electroplating. In this embodiment, the diameter is 10 μm to 20 μm and the height is 5 μm to 10 μm.

[0140] Precision alignment and bonding: The four core layers are sequentially aligned and bonded under a nitrogen atmosphere using hot-press bonding or reflow soldering processes, with an alignment accuracy of no more than 1μm; Housing encapsulation: The 3D stacked core assembly is mounted on the substrate 5. The substrate 5 can be made of a ceramic substrate, a glass substrate, or a PCB substrate. A metal or ceramic top cover is installed and hermetic sealing is performed to prevent external contaminants from entering.

[0141] Fluid interface and electrical interface integration: Fluid channel openings are machined on the side of the package housing and sealed connectors are installed, and electrical pin arrays are arranged at the edge of the housing.

[0142] The chip's optical path is arranged vertically and coaxially: the 365nm ultraviolet light emitted by the μLED array chip 8 is focused by the focusing microlens array chip 9 and then vertically downwards illuminates the central region of the micro-reaction cell 613; the fluorescence signal generated by the base extension reaction passes vertically downwards through the bottom transparent window of the micro-reaction cell 613 and the fluorescence filter lens layer 15, and is then focused onto the photosensitive area of ​​the corresponding SPAD tube. The three layers—μLED array chip 8, micro-reaction cell array layer 12, and SPAD array layer 16—achieve optical axis alignment through wafer-level bonding and precision alignment technology, with an alignment error not exceeding 1μm. The microfluidic liquid channel is arranged in a horizontal plane, completely avoiding the vertical optical path channel, preventing the liquid from absorbing and scattering ultraviolet light or fluorescence.

[0143] The chip's liquid path adopts a single-sided inlet / outlet layout: the right side is the liquid path input end, integrating a standardized Luer connector or micro / nanofluidic interface, with 6 input ports (dATP, dTTP, dCTP, dGTP, TdT / polymerase, buffer / rinsing solution), with an outer diameter of 100 μm and an inner diameter of 50 μm. The interface material is PDMS / glass bonding, which is biocompatible; the left side is the liquid path output end, with three output ports (corresponding to products, waste liquid, and bubbles). The liquid flows from right to left within the microfluidic channel layer 10, sequentially completing reagent dispensing, reaction, rinsing, and product recovery, with a flow stability coefficient of variation not exceeding 5%.

[0144] The electrical interface layout of the chip follows the functional zoning principle: the upper side of the chip is the IO communication interface area, where high-speed differential signal lines, clock lines and control buses are centrally arranged, which facilitates connection with the flat cable of the host computer 527; the lower side of the chip is the power and ground interface area, where multiple power supply pins and large-area ground pins are arranged to reduce power supply noise and grounding impedance.

[0145] Power-sensitive lines (such as high-current lines for μLED driving and high-voltage bias lines for SPAD) and signal-sensitive lines (such as high-speed communication lines and SPAD readout signal lines) are isolated by ground plane and shielded by routing design to avoid crosstalk.

[0146] Reference Figure 10 The chip's optical path is arranged vertically and coaxially. The 365nm ultraviolet light emitted by the μLED light source 611 in the light-controlled chip 1 is focused by the condenser lens 612 and then vertically downwards to illuminate the central region of the micro-reaction cell 613; the fluorescence signal generated by the base extension reaction passes vertically downwards through the transparent window at the bottom of the micro-reaction cell 613, and after the excitation light scattering is filtered out by the fluorescence filter 614, it is focused onto the photosensitive area of ​​the SPAD fluorescence detection region 615.

[0147] Reference Figure 11The 100×100 μLED array-microreactor array-SPAD array 621 are aligned vertically one by one. Each μLED unit 622 corresponds to a microreactor unit 623 directly below it, and each microreactor unit 623 corresponds to a SPAD unit 624 directly below it. The alignment error of the three optical axes does not exceed 1μm, ensuring that the ultraviolet irradiation and fluorescence detection of each microreactor 613 are completed on the same vertical optical axis.

[0148] The timing control state machine solidified within the servo core 4 schedules each core in a unified manner according to the following stage sequence, and executes it cyclically until the synthesis of all bases of the target gene sequence is completed.

[0149] The first stage involves simultaneous photo-controlled deprotection and quenching. Servo chip 4 sends an addressing command to photo-controlled chip 1. The addressing module performs pixel-level address decoding on the μLED231 directly above the target microreactor 613. The μLED driving module drives the corresponding μLED231 to output 365nm ultraviolet light. The light pulse irradiates the target microreactor 613, triggering a photochemical deprotection reaction of NPPOC and other protective groups within the cell. Simultaneously, servo chip 4 sends a quenching command to fluorescence sensing control chip 3. The Geiger-quenching control circuit completes global SPAD quenching within 10ns after the rising edge of the μLED231 driving signal, ensuring the corresponding SPAD detection module is biased below its breakdown voltage and in a non-sensitive state, preventing false triggering or saturation of the SPAD due to the strong 365nm excitation light. The quenching duration covers the entire light pulse width and the subsequent afterglow decay period.

[0150] The second stage is enzymatic elongation. After the protecting group is removed, the servo core 4 controls an external precision pump valve to inject a reaction solution containing fluorescently labeled deoxyribonucleotides (dNTP-NPPOC) into the microreactor 613. The TdT or its engineered variant immobilized within the microreactor 613 catalyzes base elongation, synthesizing the fluorescently labeled dNTP into the growth chain. The reaction temperature is maintained at 37℃±0.1℃ by a temperature control module via PID closed-loop control, providing an optimal catalytic environment for the TdT enzyme.

[0151] The third stage involves rinsing off residual dNTPs and fluorescence detection. After base extension is complete, servo core 4 controls the injection of rinsing solution into microreactor 613 to remove residual unsynthesized dNTPs and byproducts. After rinsing, servo core 4 deactivates the SPAD quenching state, and the SPAD array 421 of the fluorescence sensing control core 3 switches to Geiger mode to perform single-photon detection of the fluorescence signal generated by base extension in microreactor 613. Data processing layer 18 performs real-time quality assessment of the detection results, with a single-cell assessment response time not exceeding 10 μs and a full array assessment cycle not exceeding 100 ms.

[0152] The fourth stage involves feedback error correction and fluorescent group removal. Data processing layer 18 sends the quality assessment results to servo chip 4 via the communication interface. Based on the assessment results, servo chip 4 triggers error correction operations on the erroneous microreactor 613, such as increasing the light pulse width or driving current to enhance deprotection, repeating the base extension step, or marking the discarded microreactor 613 to skip subsequent operations. After error correction, the fluorescence background is eliminated, and the next base synthesis cycle begins. These four stages are uniformly scheduled by the timing control state machine of servo chip 4, integrating synthesis, detection, and error correction into an on-chip real-time closed-loop process, with error detection and correction cycles reaching millisecond levels.

[0153] Example 2 A photoelectric co-enzymatic gene synthesis method is applied to the three-dimensional photoelectric co-enzymatic gene synthesis chip provided in Example 1. The method is driven by the timing control state machine of the servo chip 4 and is executed according to the following steps.

[0154] Step S1 is the synchronization of photo-controlled deprotection and quenching. Servo chip 4 sends an addressing command to photo-controlled chip 1. The addressing command includes the two-dimensional coordinates of the target microreactor 613, the optical pulse width (adjustable from 1ms to 100ms), the optical pulse sequence mode, and the light intensity level information. The addressing response time does not exceed 10μs. μLED array chip 8 emits 365nm ultraviolet light to the target microreactor 613, triggering the removal of the protecting groups within the target microreactor 613. The addressing mode can be at least one of global scan mode, row parallel mode, column parallel mode, and arbitrary unit independent decoding mode, flexibly adapting to the parallel synthesis requirements of different scales. Simultaneously, servo chip 4 sends a quenching command to fluorescence sensing and control chip 3. The Geiger-quenching control circuit controls the corresponding SPAD detection module to enter the quenching state. The time alignment accuracy of quenching synchronization does not exceed 10ns, and the quenching duration covers the optical pulse width and the subsequent afterglow decay period, preventing the SPAD from being falsely triggered or saturated due to the strong 365nm excitation light.

[0155] Step S2 is enzymatic extension. After the protecting group is removed, the servo core 4 controls an external precision pump valve to inject a reaction solution containing fluorescently labeled deoxynucleotides into the microreactor 613. TdT or its engineered variants catalyze base extension, incorporating the fluorescently labeled dNTPs into the DNA growth chain. The reaction temperature is maintained at 37℃±0.1℃ by a temperature control module through PID closed-loop control.

[0156] In this embodiment, the fluorescently labeled deoxynucleotide is a 3'-O-NPPOC-protected deoxynucleoside triphosphate, wherein the base is covalently coupled to a characteristic fluorescent dye via an enzymatically cleavable linker. The core cleavage group of the enzymatically cleavable linker is an o-nitrobenzyl derivative, with an absorption peak at 350 nm-370 nm and a photolysis quantum yield of approximately 0.2-0.5. The o-nitrobenzyl group remains stable under elongation reaction conditions, ensuring that the fluorescent dye and the base are simultaneously synthesized to the primer terminus under TdT catalysis.

[0157] Step S3 involves rinsing off residual dNTPs and fluorescence detection. After base extension, the servo core 4 controls the injection of rinsing solution into the microreactor 613 to remove any remaining unincorporated dNTPs and byproducts. After rinsing, the servo core 4 deactivates the SPAD quenching state, and the SPAD array 421 switches to Geiger mode to perform single-photon detection of the fluorescence signal within the microreactor 613. The TDC module records photon arrival timestamps at a time resolution of 10 ps to 50 ps, ​​the histogram processing module extracts fluorescence lifetimes through TCSPC histogram fitting, and the inference analysis module 441 performs real-time evaluation of the synthesis quality of each microreactor 613 based on fluorescence intensity threshold determination, base type identification, and abnormal signal detection. The single-cell evaluation response time does not exceed 10 μs.

[0158] Step S4 involves feedback error correction and fluorescent group removal. Data processing layer 18 sends the quality assessment results to servo core 4. Based on the assessment results, servo core 4 triggers error correction operations on the erroneous microreactor 613. Error correction operations include at least one of the following: increasing the light pulse width or driving current to enhance deprotection, repeating the base extension step, and labeling the discarded microreactor 613. After error correction, the fluorescence background is eliminated, and the next base synthesis cycle begins. After all synthesis cycles are completed, servo core 4 controls the product collection valve 526 to open, exporting the target DNA product outside the chip.

[0159] For microreactor 613 determined to be synthesized normally, a phosphor removal operation is performed: using the same μLED array described in step S1, a phosphor removal light pulse dose is applied to the target microreactor 613. The phosphor removal light pulse dose falls into the low-dose region, which is different from the deprotection light pulse dose that falls into the high-dose region in step S1. The relative value of the light pulse dose in the low-dose region falls into the interval [1, 10], and the relative value of the light pulse dose in the high-dose region falls into the interval [20, 100], with a dose isolation band (10, 20) set between them.

[0160] It is understood that in the above embodiments, the protecting group is not limited to NPPOC (p-nitrophenylpropoxycarbonyl), but can also be other photolytically cleavable protecting groups such as NVOC (nitroveratroloxycarbonyl), and the deprotection wavelength can be adjusted accordingly to the appropriate wavelength range around 365nm.

[0161] It is understood that in the above embodiments, the SPAD device is not limited to silicon-based or GaAs-based, but can also use InGaAs-based (suitable for near-infrared fluorescent dyes) or other wide bandgap semiconductor materials, and the detection wavelength can cover the range of 400nm to 1000nm.

[0162] It is understood that in the above embodiments, the 3D vertical stacking interconnect method is not limited to TSV+μBump, and copper-copper hybrid bonding (Cu-Cu Hybrid Bonding) can also be used to achieve higher density vertical interconnects, which is suitable for ultra-high throughput chip solutions with smaller pitch.

[0163] It is understood that in the above embodiments, the reasoning and analysis module 441 of the data processing layer 18 is not limited to decision trees or small neural networks, but may also use other lightweight machine learning models such as support vector machines or convolutional neural networks.

[0164] It is understood that in the above embodiments, the material of substrate 5 is not limited to ceramic substrate, glass substrate or PCB substrate, but can also be aluminum nitride ceramic substrate (with better thermal conductivity, suitable for high power μLED driving scenarios) or silicon substrate (with better compatibility with CMOS process).

[0165] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of one or more computer-usable storage media (including, but not limited to, disk storage, etc.) containing computer-usable program code. CD - ROM It takes the form of a computer program product implemented on (such as optical memory, etc.).

[0166] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0167] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0168] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0169] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made by those skilled in the art without creative effort within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip, characterized in that, include: The light-controlled chip contains multiple light-emitting sources arranged in an array. The light-emitting sources are used to emit excitation light to the target microreactor according to a preset addressing signal, triggering the removal of the protecting groups in the target microreactor. The gene synthesis core contains multiple microreactors arranged in an array, each microreactor being used to immobilize terminal deoxynucleotidyl transferase or its engineered variant to catalyze a base elongation reaction after the protection group is removed; The fluorescence sensing control chip includes multiple photodetectors arranged in an array, used to detect the fluorescence signal of the base extension reaction in the target microreactor after the triggering of the control chip is turned off, and to obtain the fluorescence detection result; The servo chip is communicatively connected to the optical control chip, the gene synthesis chip, and the fluorescence sensing control chip, respectively, and is used to control the timing of the actions of the optical control chip, the gene synthesis chip, and the fluorescence sensing control chip according to the closed-loop logic of gene synthesis, detection, and error correction. The light control chip, the gene synthesis chip, the fluorescence sensing control chip, and the servo chip are vertically stacked and packaged together, and the array formed by the light source, the microreactor, and the photodetector is aligned with each other in the vertical direction to form multiple independent units.

2. The three-dimensional photoelectric synergistic enzymatic gene synthesis chip as described in claim 1, characterized in that, The optical control chip includes addressing chips, light source driving chips, light source array chips, and focusing microlens array chips arranged vertically from top to bottom, wherein: The addressing core is communicatively connected to the servo core and is used to receive the addressing command sent by the servo core, convert the two-dimensional coordinates of the target microreactor into a gating signal, and perform pixel-level address decoding on the light source corresponding to the target microreactor. The light source driving chip is signal-connected to the addressing chip and is used to drive the corresponding light source to output excitation light according to the received light pulse control command and the gating signal; The light source array core is electrically connected to the light source driving core, and is used to emit excitation light with a predetermined peak wavelength; The focusing microlens array core is located below the light source array core, and is used to focus the excitation light onto the central region of the target microreactor and suppress optical crosstalk between adjacent microreactors; The addressing core supports four addressing modes: global scan, row parallel, column parallel, and independent decoding of any unit.

3. The three-dimensional photoelectric synergistic enzymatic gene synthesis chip as described in claim 2, characterized in that, The light source driver chip is a μLED driver chip, which integrates multiple independent constant current driving units. Each constant current driving unit includes a digital-to-analog converter, a constant current source, and a gating switch. The digital-to-analog converter is data-connected to the addressing chip and is used to output a reference voltage according to the light intensity level information. The constant current source is electrically connected to the digital-to-analog converter and is used to convert the reference voltage into a driving current. The gating switch is controlled to be turned on or off by the gating signal. The adjustment range of the drive current is 0.1mA to 10mA, and the rise time and fall time of the current do not exceed the predetermined values. Alternatively, the μLED driver chip may also integrate an overcurrent protection circuit, a thermal shutdown circuit, and an electrostatic discharge (ESD) protection circuit. The overcurrent protection circuit provides overcurrent protection, the thermal shutdown circuit executes a thermal shutdown mechanism, and the ESD protection circuit clamps abnormal voltages.

4. The three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip as described in claim 1, characterized in that, The gene synthesis core comprises, from top to bottom, a vertically stacked microfluidic channel layer, an upper sealing layer, a microreactor array layer, a lower sealing layer, and a product collection-waste treatment-temperature control layer, wherein: The microfluidic channel layer includes several reagent supply channels, enzyme channels, rinsing solution channels, and bubble discharge channels; The upper sealing layer is provided with a reagent injection port, an enzyme injection port, a rinsing solution injection port and a bubble discharge port. The reagent injection port is connected to the reagent supply channel, the enzyme injection port is connected to the enzyme channel, the rinsing solution injection port is connected to the rinsing solution channel, and the bubble discharge port is connected to the bubble discharge channel. The lower sealing layer is provided with a waste liquid discharge port and a product collection port; The product collection-waste liquid treatment-temperature control layer includes a waste liquid collection channel, a product collection pool, and a temperature control module. The waste liquid collection channel is connected to the waste liquid outlet, the product collection pool is connected to the product collection port, and the temperature control module is used to maintain the reaction temperature in the micro-reaction pool in the micro-reaction pool array layer within a predetermined range.

5. The three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip as described in claim 4, characterized in that, The spacing between adjacent micro-reaction cells in the micro-reaction cell array layer is 20μm to 50μm, and the depth of the micro-reaction cell is 5μm to 10μm. A high-reflectivity metal thin film is deposited on the inner wall surface of the micro-reaction cell and covered with a protective layer, so that the reflectivity of the wall surface to the excitation light is not less than a set percentage, and the optical crosstalk between adjacent micro-reaction cells does not exceed a set value. The water contact angle on the inner wall surface of the microreactor does not exceed 30°; The microreactor has a primer anchoring point at the bottom center for covalently immobilizing or affinity capturing deoxyribonucleic acid primers; Alternatively, the temperature control module includes a thin-film thermal sensor array and a microfluidic coolant channel. The thin-film thermal sensor array is distributed on the back of the product collection-waste liquid treatment-temperature control layer and is used to collect reaction temperature information. The microfluidic coolant channel contains circulating temperature control liquid. The microfluidic coolant channel works in conjunction with the thin-film thermal sensor array to regulate the reaction temperature by using PID closed-loop control of the circulating temperature control liquid.

6. The three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip as described in claim 1, characterized in that, The fluorescence sensing and control chip includes a fluorescence filter lens layer, a photodetector array layer, an addressing control layer, and a data processing layer, which are vertically stacked from top to bottom, wherein: The fluorescent filter lens layer is located above the photodetector array layer and is used to focus the fluorescence emitted by the microreactor onto the photodetector photosensitive area and filter out excitation light scattering and ambient stray light. The photoelectric detection array layer is optically coupled to the fluorescent filter lens layer, and includes multiple detection modules, with one detection module corresponding to the area directly below each microreactor. The addressing control layer is located below the photoelectric detection array layer and integrates a photoelectric detection address decoding module and a Geiger-quenching circuit module. The photoelectric detection address decoding module is spatially mapped to the addressing chip of the optical control chip. During the illumination of the light source, the Geiger-quenching circuit module controls the corresponding detection module to be in a quenched state. After the light source is turned off, the Geiger-quenching circuit module controls the corresponding detection module to switch to Geiger mode for single-photon fluorescence detection. The data processing layer, located below the addressing control layer, is used to perform quality assessment on the single-photon fluorescence detection results and send the assessment results to the servo chip.

7. The three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip as described in claim 6, characterized in that, The fluorescent filter lens layer adopts a four-color filter array, with the four quadrants corresponding to the characteristic wavelengths of four basic fluorescent dyes, namely A, T, C, and G. The photoelectric detection array layer includes multiple detection modules arranged in an array. Each detection module includes 2×2 single-photon avalanche diodes. Each single-photon avalanche diode corresponds to one of the four quadrants of the fluorescent filter lens layer. By analyzing the relative response intensity of the four single-photon avalanche diodes, the type of base currently synthesized and extended can be identified, thereby achieving in-situ base type verification. The data processing layer includes a time-to-digital transformation module, a histogram processing module, and an inference analysis module, wherein: The time-to-digital conversion module is used to record the arrival timestamp of each photon event; The histogram processing module is used to perform statistical histogram analysis on the photon events detected in each microreactor during the synthesis cycle, and extract fluorescence intensity, photon count rate and fluorescence lifetime; The inference analysis module is used to evaluate the synthesis quality of each microreactor in real time based on the extracted data, using a preset threshold or a lightweight machine learning model. The evaluation results are sent to the servo chip via a communication interface.

8. The three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip as described in claim 1, characterized in that, The servo chip includes a light-controlled chip servo module, a gene synthesis chip servo module, a fluorescence sensing control chip servo module, a power management module, and an electrical signal interface module, wherein: The photocontrol particle servo module is communicatively connected to the photocontrol particle and is used to send corresponding addressing instructions and light pulse control instructions to the photocontrol particle according to the base type and target sequence information of the current synthesis cycle. After receiving the synthesis error information fed back by the fluorescence sensing photocontrol particle, it dynamically adjusts the photocontrol parameters at the error point. The gene synthesis core servo module communicates with an external pump valve through an external electrical signal interface, and is used to send control commands for reagent injection, rinsing solution injection, waste liquid discharge and product collection in a time sequence. The fluorescence sensing and control chip servo module is communicatively connected to the fluorescence sensing and control chip and is equipped with a quenching synchronization circuit, which is used to complete the global quenching of the photodetector within a predetermined time after the rising edge of the light source driving signal is triggered, and to configure the readout mode of the photodetector during the fluorescence detection stage. The power management module includes a multi-channel voltage regulator and a DC-DC converter, which are used to provide stable power supplies of multiple voltage levels to each chip and control the power-on and power-off of each chip according to a safe timing sequence. The electrical signal interface module is used to communicate with an external host computer, report the synthesis status in real time, and receive gene sequence data and timing control commands.

9. A three-dimensional photoelectric co-catalyzed enzymatic gene synthesis chip as described in claim 1, characterized in that, The excitation light emitted by the light source is focused by a focusing microlens and then vertically downwards to illuminate the central region of the micro-reaction cell. The fluorescence signal generated by the base extension reaction in the micro-reaction cell passes vertically downwards through the transparent window at the bottom of the micro-reaction cell and the fluorescence filter lens layer and is then focused onto the photosensitive area of ​​the corresponding photodetector. The optical axis alignment error of the corresponding light source, micro-reaction cell and photodetector is within the set range. Each microfluidic channel is arranged in the horizontal plane and does not intersect with the vertical optical path; The liquid inlet is located on one side of the chip, and the liquid outlet is located on the other side of the chip; The communication interfaces are centrally located on the top side of the chip, while the power and ground interfaces are centrally located on the bottom side of the chip.

10. A method for photoelectric co-enzyme-catalyzed gene synthesis based on a three-dimensional photoelectric co-enzyme-catalyzed gene synthesis chip according to any one of claims 1-9, characterized in that, Includes the following steps: S1. Synchronization of optically controlled deprotection and quenching: The servo chip sends an addressing command to the optically controlled chip to control the light source to emit excitation light to the target microreactor, triggering the removal of the protective group in the target microreactor; the servo chip sends a quenching command to the fluorescence sensing and control chip, and the fluorescence sensing and control chip controls the corresponding photodetector to enter the quenching state, with the quenching duration covering the light pulse width and the subsequent afterglow decay period. S2, Enzymatic Elongation: After the protecting group is removed, the servo core controls the external pump valve to inject a reaction solution containing fluorescently labeled deoxynucleotides into the target microreactor, and the terminal deoxynucleotide transferase or its engineered variant catalyzes base elongation; S3. Rinsing Residue and Fluorescence Detection: After base extension is completed, the servo core controls the external pump valve to inject rinsing solution into the target microreactor to remove residual reagents; After rinsing, the servo chip releases the quenching state of the photodetector, the photodetector switches to Geiger mode, and performs single-photon detection on the fluorescence signal in the target microreactor. The fluorescence sensing and control chip performs real-time quality assessment on the detection results. S4. Feedback Error Correction and Fluorescent Group Removal: The fluorescence sensing and control chip sends the quality assessment result to the servo chip. The servo chip triggers an error correction operation on the erroneous microreactor based on the quality assessment result. After the error correction is completed, the fluorescence background is eliminated, and the next base synthesis cycle begins.

11. The photoelectric co-catalyzed enzymatic gene synthesis method as described in claim 10, characterized in that, In step S1, the light pulse width of the light source is 1ms to 100ms, and the driving current is 0.1mA to 10mA. The addressing command includes the two-dimensional coordinates of the target microreactor, the pulse width, the pulse sequence mode, and the intensity level information. Addressing modes include one of the following: global scan mode, row parallel mode, column parallel mode, and arbitrary unit independent decoding mode; Alternatively, in step S3, the photodetector array operates in Geiger mode, the time-to-digital conversion module of the data processing layer of the fluorescence sensing and control chip records the photon arrival timestamps at a time resolution of 10 ps to 50 ps, ​​the histogram processing module extracts fluorescence intensity, photon count rate and fluorescence lifetime by fitting the time-correlated single photon count histogram, and the inference analysis module evaluates the synthesis quality of each microreactor in real time based on the extracted data by using fluorescence intensity threshold determination, base type identification and abnormal signal detection. Alternatively, in step S4, the error correction operation includes at least one of increasing the optical pulse width or driving current to enhance deprotection, repeating the base extension step, and labeling the discarded microreactor. After all synthesis cycles are completed, the servo chip controls the gene synthesis chip to collect the product and export the target DNA product to the outside of the chip.

Images