High-throughput single-cell polyomics

Abstract

Provided herein, in some embodiments, are devices, systems and methods for high-throughput single-cell polyomics (e.g., genomic, epigenomic, proteomic and/or phenotypic profile) analyses.

Claims

1. A polyomic multiplexing device, comprising: a substrate comprising X rows intersecting (i) Y columns to form X*Y patches, and (ii) Z.sub.n+1 columns to form X*Z patches, wherein each of the X*Y patches comprises a unique nucleic acid barcode that is immobilized to the substrate and comprises a polyT sequence, wherein: each of the rows comprises a different subset of barcoded nucleic acid strands of a first set of nucleic acid strands, and each of the Y columns comprises a different subset of barcoded nucleic acid strands of a second set of nucleic acid strands, the nucleic acid strands of the first set are bound to nucleic acid strands of the second set to form the unique nucleic acid barcode at each of the X*Y patches, the unique nucleic acid barcode comprises, in the 5′ to 3′ direction, a barcode sequence of the first set, a barcode sequence of the second set, and a polyT sequence, each of the X*Z patches comprises an antibody immobilized to the substrate, and n is zero or greater; and an array of microwells coupled to the substrate, wherein each microwell comprises one of the unique molecular barcodes immobilized the substrate and at least one of the antibodies immobilized to the substrate.

2. The polyomic multiplexing device of claim 1, wherein X is at least 10.

3. The polyomic multiplexing device of claim 2, wherein X is 10 to 20000.

4. The polyomic multiplexing device of claim 1, wherein Y is at least 10.

5. The polyomic multiplexing device of claim 4, wherein Y is 10 to 20000.

6. The polyomic multiplexing device of claim 1, wherein n is at least 1, and each of the Z.sub.n+1 columns comprises a different antibody.

7. The polyomic multiplexing device of claim 6, wherein n is at least 2, and each of the Z.sub.n+1 columns comprises a different antibody.

8. The polyomic multiplexing device of claim 1, wherein the microwell array comprises at least 20 microwells.

9. The polyomic multiplexing device of claim 1, wherein the nucleic acid strands of the first set of nucleic acid strands comprise, in the 5′ to 3′ direction: a promoter sequence, a sequencing adaptor sequence, a first barcode sequence and a first anchor sequence.

10. The polyomic multiplexing device of claim 1, wherein the nucleic acid strands of the second set of nucleic acid strands comprise, in the 5′ to 3′ direction: a polyT sequence, a unique molecular identifier sequence, a second barcode sequence and a second anchor sequence, wherein the second anchor sequence is complementary to the first anchor sequence.

11. The polyomic multiplexing device of claim 1, wherein the unique nucleic acid barcode comprises, in the 5′ to 3′ direction: a promoter sequence, a sequencing adaptor sequence, a first barcode sequence, a second barcode sequence, a unique molecular identifier, and a polyT sequence.

12. The polyomic multiplexing device of claim 1, wherein the substrate comprises glass, silicon or silica.

13. The polyomic multiplexing device of claim 1, wherein the substrate is coated with poly-1-lysine.

14. The polyomic multiplexing device of claim 1, wherein each column and/or row has a width of 50-200 microns.

15. The polyomic multiplexing device of claim 1, wherein each patch has an area of 400-40,000 μm.sup.2.

16. The polyomic multiplexing device of claim 1, wherein patches within a single row and/or within a single column are separated from each other by 20-200 microns.

17. The polyomic multiplexing device of claim 1, wherein patches between adjacent rows and/or between adjacent columns are separated from each other by 20-200 microns.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1: Design and fabrication of a large-scale array of distinct molecular barcode patches (40,000 per chip) for single-cell transcriptome capture. The upper images illustrate an example flow patterning fabrication process. The lower left images show a cross-sectional view of microchannel-guided flow patterning. The lower right images shows an examples of composite molecular barcodes combining Barcode A, Barcode B and other sequences (e.g., promoter, sequencing adaptor, unique molecular identifier (UMI) and polyT sequence).

(2) FIG. 2: Microfluidic flow patterning chips designed to produce molecular barcode patches. This examples includes a 42×42 array (throughput ˜300-600 single cells) with 100 micron flow-patterning channels (barcoded patches are 100×100 μm.sup.2). The corresponding microwells are separated from each other by 100 μm such that no two wells can share the same barcode.

(3) FIG. 3: Barcode immobilization reactions. The oligonucleotide sequence including T7 promoter, sequencing adapter and row barcodes (Barcodes A) are patterned and immobilized on glass surface using the first microfluidic flow patterning chip. Afterwards, the column barcode sequences (Barcodes B) are hybridized and extended through primer extension reaction. The reactions are quenched with sodium hydroxide to strip off the shorter second strand, and complete the molecular barcodes terminal with poly(dT) sequences. The barcode sequences are known a priori to enable deterministic barcoding.

(4) FIG. 4: Flow patterning of molecular barcodes confirmed by fluorescently-labeled hybridization probes. Immobilization of row barcodes (Barcodes A, top left). Patterning and extension of column barcodes confirmed by a fluorescently labeled hybridization probe against the poly(dT) sequences (top right) illustrating the barcoded patches. Distinct barcode sequences are quality checked using fluorescently labeled hybridization probes against the specific barcode sequences against row barcode 20 (bottom right) and column barcode 19 (bottom right). Only the oligonucleotide sequences carrying column 19 and row 20 barcode sequences are specifically labeled.

(5) FIG. 5: Microwell array devices used for single cell capture. In one example, wells are created on the surface of a substrate (left image). In another example, a through-hole microwell array was created on a thin substrate (middle and right images). Substrates can be glass, polymer or plastic.

(6) FIG. 6: Alignment-free strategy for interfacing high-density barcode array with single cells. Microwells are loaded randomly with cells such that ˜50-70% of the wells are occupied with single cells. The microwells are then sealed with barcode array without alignment. After mechanical clamping to stabilize the seal, fluorescence and bright-field images are taken to determine the well and barcoded patch locations. Patches are patterned such that one patch can only contain up to 4 wells at most. This alignment-free approach yields ˜10-30% single cell correspondence with the distinctly barcoded patches.

(7) FIG. 7: Deterministic alignment strategy for interfacing a high-density barcode array with single cells. Microwells are fabricated with through-hole wells which are then aligned and bonded with the high-density barcoded array using custom-made alignment platform such that each barcoded patch corresponds to a single well. Once the cells are loaded randomly, microwells are sealed with a second glass slide. After mechanical clamping to stabilize the seal, bright field images are taken to determine the well occupancy and to match barcodes to cells. The expected yield is ˜50-70% of the total number of barcoded patches.

(8) FIG. 8: Alternative deterministic strategy for interfacing high-density barcode array with single cells. Instead of patterning the molecular barcodes onto a flat glass slide, barcodes are patterned and generated onto a microwell platform such that distinct barcodes are directly immobilized within each well. Cell loading and assaying is similar as in other versions.

(9) FIG. 9: A cell trapping strategy is developed where two microwell platforms are used to increase the loading efficiency of the microwells. Cells are first loaded onto a smaller microwell array where almost all wells are loaded with single cells due to size exclusion (each well size˜cell diameter). Once saturated, the smaller microwell is aligned with the larger microwell array and cells are transferred to larger wells. This enables high-efficiency loading of larger microwell arrays, which offers a larger area for mRNA capture, for example.

(10) FIG. 10: Example of cell lysis in microwell arrays using freeze-thaw method. The top images depict a field of view of microwells that are loaded with single cells labeled with green fluorescent stain (cells are not visible in bright field images). The bottom two images show the cells after two freeze-thaw cycles. Cell lysis is apparent by the diffusion of fluorescent stain into the whole volume of individual wells. Sealing maintains cell lysate within each well, preventing leakage. Other lysis methods including but not limited to detergent-based, thermal-coating or spay-coating of lysis buffer may be used.

(11) FIG. 11: Quality check of library preparation procedures applied in the system described. Bulk cell lysate from ˜500,000 cells is applied to high-density barcode array. As a control, RNAse free water was applied onto the high-density barcode array. Libraries were prepared following the CelSeq2 protocol High sensitivity bio analyzer. Results showed successful library preparations from the lysate-applied barcode array, where sample RNA and DNA showed expected curves with average sizes at 300 bp and 400 bp. The control barcode array yielded no detectable RNA or DNA.

(12) FIG. 12: Example of polyomics measurements from same single cells via serial measurements. After cell loading, cells are imaged through various microscopy imaging techniques to record surface marker expression, morphology, motility and other relevant physical phenotypes. The wells are then sealed with an antibody barcode array (see International Publication No. WO 2014/031997, published Feb. 27, 2014, incorporated herein by reference) to measure secretion of ˜45 proteins. Next, the antibody barcode array is removed and replaced by a high-density barcode array for transcriptomic measurements. The data from same single cells are then integrated and analyzed together using custom software.

(13) FIG. 13: Combining multiple omics analyses via deterministic barcoding. Example 1: single-cell transcriptome and microRNA co-analysis. Left: Schematic depiction of cell loading and device assembly, and on-chip cell lysis. Right: Use of upper and lower slides for microRNA and transcriptome measurement of the same single cells. The mechanism that microRNAs and mRNAs can be linked and associated to the same single cell is due to the deterministic molecular barcoding, in which the sequence and location of each molecular barcode in each microchamber (microwell) is known.

(14) FIGS. 14A-14B: Combining polyomics analyses via deterministic barcoding. Example 2: co-detection of whole transcriptome and a panel of proteins either secreted from the cell (FIG. 14A) or intracellular/surface proteins released after cell lysis (FIG. 14B). The former is conducted using the aforementioned molecular barcode approach with the lower glass slides containing molecular barcode micropatch array. The latter is conducted with an array of antibody features patterned on the top glass slide. The mechanism that proteins and mRNAs can be linked and associated to the same single cell is due to the deterministic molecular barcoding, in which the sequence and location of each molecular barcode in each microchamber is known.

(15) FIGS. 15A-15D depict an example of a device of the present disclosure, which include a substrate comprising columns of barcoded nucleic acids of a first subset (FIG. 15A), rows of barcoded nucleic acids of a second subset (FIG. 15B), additional columns (formed by a ‘wavy’ flow pattern) of subsets of different antibodies (FIG. 15C), and microwells coupled to the substrate (black squares) (FIG. 15C).

DETAILED DESCRIPTION

(16) Single-cell sequencing, in particular, sequencing of a whole transcriptome for gene expression profiling and phenotype analysis, is an enabling scientific discovery tool in nearly all fields of biology. Nonetheless, several major problems still exist. First, in order to quantitatively dissect phenotypic and functional heterogeneity of complex cell populations, one must simultaneously sequence more than 10,000 single cells. To date, there is no technology to achieve this goal. Second, in order to be utilized in the clinical setting, the technology should work for low-input samples (e.g., <50,000 cells) and rare cell populations isolated from clinical specimens. Third, the field is still unable to measure polyomics information in the same cell, for example, to directly correlate gene expression (transcriptome sequencing) to regulatory elements (e.g., microRNAs, epigenetic modification), which is important for understanding the mechanism of cellular heterogeneity. The technology of the present disclosure addresses the three foregoing problems.

(17) Methods

(18) Provided herein are methods of producing a barcoded array, comprising (a) flow patterning and immobilizing onto a surface of a substrate a first set of barcoded nucleic acid (e.g., DNA) strands of a first solution to produce columns that are parallel to and space apart relative to each other, wherein each column comprises X patches of barcoded nucleic acid strands of the first set, wherein the patches within each column are spaced apart relative to each other, wherein each column comprises a different subset of barcoded nucleic acid strands, and wherein X is a number greater than 2; and (b) flow patterning and immobilizing onto the surface of the substrate a second set of barcoded nucleic acid (e.g., DNA) strands of a second solution to produce rows that are parallel to and space apart relative to each other, wherein each row comprises Y patches of barcoded nucleic acid strands of the second set, wherein the patches within each row are spaced apart relative to each other, wherein each row comprises a different subset of barcoded nucleic acid strands, wherein the rows are perpendicular relative to the columns, and wherein Y is a number greater than 2, thereby producing a X*Y array of overlapping patches, each overlapping patch comprising (i) a subset of barcoded nucleic acid strands of the first set and (ii) a subset of barcoded nucleic acid strands of the second set.

(19) General methods of flow patterning are known, and include, for example, streamline flow patterning, which is the flow of fluid in which its velocity at any point is constant or varies in a regular manner.

(20) In some embodiments, the barcoded nucleic acid strands of the first set comprise, in the 5′ to 3′ direction: a promoter sequence, a sequencing adaptor sequence, a first barcode sequence and a first anchor sequence. In some embodiments, the barcoded nucleic acid strands of the first set comprise, in the 3′ to 5′ direction: a promoter sequence, a sequencing adaptor sequence (“sequencing adaptor”), a first barcode sequence and a first anchor sequence.

(21) Promoter sequences are DNA sequences that define where transcription of a gene or other downstream nucleotide sequence by polymerase (e.g., RNA polymerase) begins. Examples of promoter sequences include, but are not limited to, T7 promoter sequences, T3 promoter sequences, and SP6 promoter sequences.

(22) Sequence adaptors are short (known) nucleotide (e.g., DNA) sequences added to an end of a nucleic acid of interest. A complementary sequencing primer binds to the sequence adaptor. The length of a sequence adaptor may vary. For example, a sequence adaptor may have a length of 10 to 50 nucleotide. In some embodiments, a sequence adaptor has a length of 10, 20, 30, 40, or 50 nucleotides.

(23) Anchor sequences enable binding of barcoded nucleic acids to each other. As shown in FIG. 3 (Step 2), Barcode A has an anchor sequence, and Barcode B has an anchor sequence that is complementary to the anchor sequence of Barcode A such that Barcode A and B can bind to each other.

(24) A barcode sequence is a sequence of nucleotides (e.g., deoxyribonucleotides) that is specific to a set or a subset of nucleic acids strands. For example, as shown in FIG. 1, the nucleic acid strands of subset A1 (of Barcode A) are coded with a specific barcode sequence, while subsets A2, A3, A4, etc. are each coded with a different barcode sequence, each barcode specific to the subset. Likewise, the nucleic acid strands of subset B1 (of Barcode B) are coded with a specific barcode sequence, while subsets B2, B3, B4, etc. are each coded with a different barcode sequence, each barcode specific to the subset. Thus, each overlapping patch, which includes a unique combination of Barcode A subsets and Barcode B subsets, contains a unique composite barcode (Barcode A+Barcode B). For example, an overlapping patch containing A1+B1 barcodes is uniquely coded relative to its neighboring overlapping patches, which contain A2+B1 barcodes, A1+B2 barcodes, A2+B2 barcodes, etc., as depicted in FIG. 1.

(25) In some embodiments, the barcoded nucleic acid strands of the second set comprise, in the 5′ to 3′ direction: a polyT sequence (e.g., T19V), a unique molecular identifier (UMI) sequence, a second barcode sequence and a second anchor sequence, wherein the second anchor sequence is complementary to the first anchor sequence. In some embodiments, the barcoded nucleic acid strands of the second set comprise, in the 3′ to 5′ direction: a polyT sequence, a unique molecular identifier sequence, a second barcode sequence and a second anchor sequence, wherein the second anchor sequence is complementary to the first anchor sequence. Examples of UMIs are described by Kivioja T et al. Nature Methods 9, 72-74 (2012), incorporated herein by reference.

(26) The methods may further comprise maintaining (incubating) the array of overlapping patches under conditions that result in hybridization of the second set of barcoded nucleic acid strands to the first set of barcoded nucleic acid strands to produce patches that comprise partially double-stranded barcoded nucleic acids. Nucleic acid hybridization conditions are known.

(27) The methods may also comprise maintaining the array of overlapping patches in the presence of a polymerase, a primer that binds to the second barcode sequence, and dNTPs (e.g., dATP, dTTP, dCTP, and dGTP), under conditions that result in DNA polymerization (production/synthesis of strand of DNA) to produce a nucleic acid strand comprising (e.g., in the 5′ to 3′ direction): a promoter sequence, a sequencing adaptor sequence, a first barcode sequence, a first anchor sequence, a second barcode sequence, a unique molecular identifier sequence and a polyT sequence. Nucleic acid synthesis conditions are known.

(28) In some embodiments, the methods comprise removing (e.g., washing) from the array of overlapping patches the second set of barcoded nucleic acid. In some embodiments, the polymerization/synthesis reaction is quenched with sodium hydroxide to strip off the shorter second barcoded nucleic acid strand,

(29) The surface may be a glass surface, a silicon surface or a silica surface. Other surfaces are encompassed by the present disclosure. In some embodiments, the glass surface is coated with poly-1-lysine.

(30) In some embodiments, the substrate is a microwell array or is coupled to a microwell array, and wherein each overlapping patch occupies or is aligned with a single microwell of the microwell array.

(31) In some embodiments, X equals 20-20,000. For example, X may equal 20-50, 20-100, 20-500, 20-1000, 20-5000, 20-10000, 50-100, 50, 500, 50-1000, 50-5000, 50-10000, 50-20000, 100-500, 100-1000, 100-5000, 100-10000, or 100-20000. In some embodiments, X equals 100-20,000.

(32) In some embodiments, Y equals 20-20,000. For example, Y may equal 20-50, 20-100, 20-500, 20-1000, 20-5000, 20-10000, 50-100, 50, 500, 50-1000, 50-5000, 50-10000, 50-20000, 100-500, 100-1000, 100-5000, 100-10000, or 100-20000. In some embodiments, Y equals 100-20,000.

(33) At least one (e.g., at least 2, 3, 4, 5, 10, 20) column, or each (all) row, may have a width of 10-500 microns, or 50-200 microns. For example, a column may have a width of 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190 or 200 microns. In some embodiments, a column has a width of 100 microns.

(34) At least one (e.g., at least 2, 3, 4, 5, 10, 20) column, or each (all) row, may have a width of 10-500 microns, or 50-200 microns. For example, a row may have a width of 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190 or 200 microns. In some embodiments, a row has a width of 100 microns.

(35) Typically, at least one, or each (all), overlapping patch has an area of 100-40,000 μm.sup.2. For example, an overlapping path may have an area of 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 20000, 30000 or 40000 μm.sup.2. In some embodiments, an overlapping patch has an area of 10,000 μm.sup.2. Thus, in some embodiments, the dimensions of a patch may be 10×10 μm to 200×200 μm. Larger or smaller overlapping patches are encompassed by the present disclosure.

(36) In some embodiments, the overlapping patches within a single row are separated from each other by 10-500 microns, or 20-200 microns. For example, the overlapping patches within a single row may be separated from each other by 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 125, 150, 175 or 200 microns. In some embodiments, the overlapping patches within a single row are separated from each other by 100 microns.

(37) In some embodiments, the overlapping patches within a single column are separated from each other by 10-500 microns, or 20-200 microns. For example, the overlapping patches within a single column may be separated from each other by 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 125, 150, 175 or 200 microns. In some embodiments, the overlapping patches within a single column are separated from each other by 100 microns.

(38) In some embodiments, the overlapping patches between adjacent rows are separated from each other by 20-200 microns. For example, the overlapping patches between adjacent rows may be separated from each other by 20, 50, 75, 100, 125, 150, 175 or 200 microns. In some embodiments, the overlapping patches between adjacent rows are separated from each other by 20-30, 20-50, 20-100, 50-100 or 50-200 microns. In some embodiments, the overlapping patches between adjacent rows are separated from each other by (about) 100 microns.

(39) In some embodiments, the overlapping patches between adjacent columns are separated from each other by 20-200 microns. For example, the overlapping patches between adjacent columns may be separated from each other by 20, 50, 75, 100, 125, 150, 175 or 200 microns. In some embodiments, the overlapping patches between adjacent columns are separated from each other by 20-30, 20-50, 20-100, 50-100 or 50-200 microns. In some embodiments, the overlapping patches between adjacent columns are separated from each other by (about) 100 microns.

(40) The first set of barcoded nucleic acid strands may be patterned and immobilized onto the surface of the substrate using, for example, a microfluidic flow patterning chip (see, e.g., FIG. 2). Likewise, the second set of barcoded nucleic acid strands may be patterned and immobilized onto the surface of the substrate using, for example, a microfluidic flow patterning chip (see, e.g., FIG. 2).

(41) Barcoded Arrays and Multiplexing Devices

(42) Also provided herein are barcoded arrays, for example, produced by any of the methods described herein. For example, a barcoded array may be produced by a method, comprising: (a) flow patterning and immobilizing onto a surface of a substrate a first set of barcoded nucleic acid strands of a first solution to produce columns that are parallel to and space apart relative to each other, wherein each column comprises X patches of barcoded nucleic acid strands of the first set, wherein the patches within each column are spaced apart relative to each other, wherein each column comprises a different subset of barcoded nucleic acid strands, and wherein X is a number greater than 2; and (b) flow patterning and immobilizing onto the surface of the substrate a second set of barcoded nucleic acid strands of a second solution to produce rows that are parallel to and space apart relative to each other, wherein each row comprises Y patches of barcoded nucleic acid strands of the second set, wherein the patches within each row are spaced apart relative to each other, wherein each row comprises a different subset of barcoded nucleic acid strands, wherein the rows are perpendicular relative to the columns, and wherein Y is a number greater than 2, thereby producing a X*Y array of overlapping patches, each overlapping patch comprising (i) a subset of barcoded nucleic acid strands of the first set and (ii) a subset of barcoded nucleic acid strands of the second set.

(43) Also provided herein are multiplexing devices comprising the barcoded array coupled to a microwell array, wherein each overlapping patch is aligned with a single microwell of the array such that each overlapping patch corresponds to a single microwell.

(44) Further provided herein are multiplexing devices comprising a barcoded array, wherein the substrate is a microwell array, and wherein each overlapping patch occupies a single microwell of the microwell array.

(45) In some embodiments, each microwell of the microwell array contains no more than 5 cells. For example, each microwell of the microwell array may contain no more than 4, no more than 3, or no more than 2 cells. In some embodiments, each microwell of the microwell array contains no more than 2 cells. some embodiments, each microwell of the microwell contains a single cell.

(46) The microwell array may be located, for example, between the barcoded array and another substrate such that microwells of the microarray are sealed (e.g., fluid cannot leave or enter the microwell).

(47) In some embodiments, the other substrate is coated with dried (e.g., lyophilized) lysis buffer.

(48) In some embodiments, the other substrate comprises a nucleic acid capture array, such as a microRNA capture array.

(49) In some embodiments, the other substrate comprises an antibody capture array (see, e.g., U.S. Pat. No. 9,188,586, incorporated herein by reference).

(50) The present disclosure further encompasses the embodiments described in the following numbered paragraphs:

(51) 1. A method of producing a barcoded array, comprising:

(52) (a) flow patterning and immobilizing onto a surface of a substrate a first set of barcoded nucleic acid strands of a first solution to produce columns that are parallel to and space apart relative to each other, wherein each column comprises X patches of barcoded nucleic acid strands of the first set, wherein the patches within each column are spaced apart relative to each other, wherein each column comprises a different subset of barcoded nucleic acid strands, and wherein X is a number greater than 2;

(53) (b) flow patterning and immobilizing onto the surface of the substrate a second set of barcoded nucleic acid strands of a second solution to produce rows that are parallel to and space apart relative to each other, wherein each row comprises Y patches of barcoded nucleic acid strands of the second set, wherein the patches within each row are spaced apart relative to each other, wherein each row comprises a different subset of barcoded nucleic acid strands, wherein the rows are perpendicular relative to the columns, and wherein Y is a number greater than 2,

(54) thereby producing a X*Y array of overlapping patches, each overlapping patch comprising (i) a subset of barcoded nucleic acid strands of the first set and (ii) a subset of barcoded nucleic acid strands of the second set.

(55) 2. The method of paragraph 1, wherein the barcoded nucleic acid strands of the first set comprise, optionally in the 5′ to 3′ direction: a promoter sequence, a sequencing adaptor sequence, a first barcode sequence and a first anchor sequence.

(56) 3. The method of paragraph 1 or 2, wherein the barcoded nucleic acid strands of the second set comprise, in the 5′ to 3′ direction: a polyT sequence, a unique molecular identifier sequence, a second barcode sequence and a second anchor sequence, wherein the second anchor sequence is complementary to the first anchor sequence.

(57) 4. The method of any one of paragraphs 1-3, further comprising maintaining the array of overlapping patches under conditions that result in hybridization of the second set of barcoded nucleic acid strands to the first set of barcoded nucleic acid strands to produce patches that comprise partially double-stranded barcoded nucleic acids.

(58) 5. The method of paragraph 4, further comprising maintaining the array of overlapping patches in the presence of a polymerase, a primer that binds to the second barcode sequence, and dNTPs, under conditions that result in DNA polymerization to produce a nucleic acid strand comprising, in the 5′ to 3′ direction: a promoter sequence, a sequencing adaptor sequence, a first barcode sequence, a first anchor sequence, a second barcode sequence, a unique molecular identifier sequence and a polyT sequence.

(59) 6. The method of paragraph 5 further comprising removing from the array of overlapping patches the second set of barcoded nucleic acid.

(60) 7. The method of any one of paragraphs 1-6, wherein the surface is a glass surface, silicon or silica.

(61) 8. The method of paragraph 7, wherein the glass surface is coated with poly-1-lysine.

(62) 9. The method of any one of paragraph 1-6, wherein the substrate is a microwell array, and were each overlapping patch occupies a single microwell of the microwell array.

(63) 10. The method of any one of paragraphs 1-9, wherein X equals 20-20,000 and/or Y equals 20-20,000.

(64) 11. The method of paragraph 10, wherein X equals 100-20,000 and/or Y equals 100-20,000.

(65) 12. The method of paragraph 11, wherein X equals 1000-20,000 and/or Y equals 1000-20,000.

(66) 13. The method of paragraph 12, wherein X equals 10,000-20,000 and/or Y equals 10,000-20,000.

(67) 14. The method of any one of paragraphs 1-13, wherein each column and/or row has a width of 50-200 microns.

(68) 15. The method of paragraph 14, wherein each column and/or row has a width of 100 microns.

(69) 16. The method of any one of paragraphs 1-15, wherein each overlapping patch has an area of 400-40,000 μm.sup.2.

(70) 17. The method of paragraph 16, wherein each overlapping patch has an area of 10,000 μm.sup.2.

(71) 18. The method of any one of paragraphs 1-17, wherein the overlapping patches within a single row and/or within a single column are separated from each other by 20-200 microns.

(72) 19. The method of paragraph 18, wherein the overlapping patches within a single row and/or within a single column are separated from each other by 100 microns.

(73) 20. The method of any one of paragraphs 1-19, wherein the overlapping patches between adjacent rows and/or between adjacent columns are separated from each other by 20-200 microns.

(74) 21. The method of paragraph 20, wherein the overlapping patches between adjacent rows and/or between adjacent columns are separated from each other by 100 microns.

(75) 22. The method of any one of paragraphs 1-21, wherein the first and/or second set of barcoded nucleic acid strands are patterned and immobilized onto the surface of the substrate using a microfluidic flow patterning chip.

(76) 23. A barcoded array produced by a method, comprising:

(77) (a) flow patterning and immobilizing onto a surface of a substrate a first set of barcoded nucleic acid strands of a first solution to produce columns that are parallel to and space apart relative to each other, wherein each column comprises X patches of barcoded nucleic acid strands of the first set, wherein the patches within each column are spaced apart relative to each other, wherein each column comprises a different subset of barcoded nucleic acid strands, and wherein X is a number greater than 2;

(78) (b) flow patterning and immobilizing onto the surface of the substrate a second set of barcoded nucleic acid strands of a second solution to produce rows that are parallel to and space apart relative to each other, wherein each row comprises Y patches of barcoded nucleic acid strands of the second set, wherein the patches within each row are spaced apart relative to each other, wherein each row comprises a different subset of barcoded nucleic acid strands, wherein the rows are perpendicular relative to the columns, and wherein Y is a number greater than 2,

(79) thereby producing a X*Y array of overlapping patches, each overlapping patch comprising (i) a subset of barcoded nucleic acid strands of the first set and (ii) a subset of barcoded nucleic acid strands of the second set.

(80) 24. The barcoded array of claim 23, wherein the barcoded nucleic acid strands of the first set comprise, optionally in the 5′ to 3′ direction: a promoter sequence, a sequencing adaptor sequence, a first barcode sequence and a first anchor sequence.

(81) 25. The barcoded array of claim 23 or 24, wherein the barcoded nucleic acid strands of the second set comprise, in the 5′ to 3′ direction: a polyT sequence, a unique molecular identifier sequence, a second barcode sequence and a second anchor sequence, wherein the second anchor sequence is complementary to the first anchor sequence.

(82) 26. The barcoded array of any one of claims 23-25, further comprising hybridizing the second set of barcoded nucleic acid strands to the first set of barcoded nucleic acid strands and producing patches that comprise partially double-stranded barcoded nucleic acids.

(83) 27. The barcoded array of claim 26, further comprising combining the array of overlapping patches with a polymerase, a primer that binds to the second barcode sequence, and dNTPs, and producing a nucleic acid strand comprising, in the 5′ to 3′ direction: a promoter sequence, a sequencing adaptor sequence, a first barcode sequence, a first anchor sequence, a second barcode sequence, a unique molecular identifier sequence and a polyT sequence.

(84) 28. The barcoded array of claim 27 further comprising removing from the array of overlapping patches the second set of barcoded nucleic acid.

(85) 29. The barcoded array of any one of claims 23-28, wherein the surface is a glass surface, silicon or silica.

(86) 30. The barcoded array of claim 9, wherein the glass surface is coated with poly-1-lysine.

(87) 31. The barcoded array of any one of claim 23-30, further comprising applying a microwell array to the surface of the substrate to produce a device wherein each overlapping patch, each row or overlapping patches, or each column of overlapping patches occupies a single microwell of the microwell array.

(88) 32. The barcoded array of any one of claims 23-31, wherein X equals 20-20,000 and/or Y equals 20-20,000, X equals 100-20,000 and/or Y equals 100-20,000, X equals 1000-20,000 and/or Y equals 1000-20,000, or X equals 10,000-20,000 and/or Y equals 10,000-20,000.

(89) 33. The barcoded array of any one of claims 23-32, wherein each column and/or row has a width of 50-200 microns, or each column and/or row has a width of 100 microns.

(90) 34. The barcoded array of any one of claims 23-33, wherein each overlapping patch has an area of 400-40,000 μm.sup.2, or each overlapping patch has an area of 10,000 μm.sup.2.

(91) 35. The barcoded array of any one of claims 23-34, wherein the overlapping patches within a single row and/or within a single column are separated from each other by 20-200 microns, or the overlapping patches within a single row and/or within a single column are separated from each other by 100 microns.

(92) 36. The barcoded array of any one of claims 23-35, wherein the overlapping patches between adjacent rows and/or between adjacent columns are separated from each other by 20-200 microns, or the overlapping patches between adjacent rows and/or between adjacent columns are separated from each other by 100 microns.

(93) 37. The barcoded array of any one of claims 23-36, wherein the first and/or second set of barcoded nucleic acid strands are patterned and immobilized onto the surface of the substrate using a microfluidic flow patterning chip.

(94) 38. A multiplexing device comprising the barcoded array of any one of paragraphs 23-37 coupled to a microwell array, wherein each overlapping patch is aligned with a single microwell of the array such that each overlapping patch corresponds to a single microwell.

(95) 39. A multiplexing device comprising the barcoded array of any one of paragraphs 23-37, wherein the substrate is a microwell array, and wherein each overlapping patch occupies a single microwell of the microwell array.

(96) 40. The multiplexing device of paragraph 38 or 39, wherein each microwell of the microwell array contains no more than 5 cells.

(97) 41. The multiplexing device of paragraph 40, wherein each microwell of the microwell array contains no more than 2 cells.

(98) 42. The multiplexing device of paragraph 41, wherein each microwell of the microwell array contains a single cell.

(99) 43. The multiplexing device of any one of paragraphs 23-42, wherein the microwell array is located between the barcoded array and another substrate such that microwells of the microarray are sealed.

(100) 44. The multiplexing device of paragraph 43, wherein the other substrate is coated with lyophilized lysis buffer.

(101) 45. The multiplexing device of paragraph 43 or 44, wherein the other substrate comprises a nucleic acid capture array.

(102) 46. The multiplexing device of paragraph 45, wherein the nucleic acid capture array is a microRNA capture array.

(103) 47. The multiplexing device of paragraph 45, wherein the other substrate comprises an antibody capture array.

EXAMPLES

Example 1. Fabrication of a High-Density Array Comprising 40,000 Distinct Molecular Barcodes

(104) Two hundred unique DNA barcodes (FIG. 1, Barcode A's: A1, A2, A3 . . . A200), each having a 5′ linker (T7 promoter+sequencing adapter) were immobilized on a substrate using a microfluidic flow patterning technique. The first set of Barcode A molecules (A1-A200; see, e.g., FIG. 3 step 1) were patterned onto a poly-1-lysine-coated glass slide using a microfluidic patterning chip (FIG. 2, top) and incubated overnight to form “Barcode A stripes”. The barcode concentration was greater than 100 μM in order to attach billions of barcodes corresponding to each single cell. The patterned Barcode A molecules were covalently linked to the glass slide surface by UV exposure and incubation at 80° C. for at least 2 hours. The 3′ ends of the immobilized Barcode A molecules were free for subsequent hybridization and extension reactions. After patterning the first set of Barcode A molecules, the original microfluidic chip was removed, and a second microfluidic patterning chip (FIG. 2, bottom) was placed on top of the patterned glass slide with the microchannels perpendicular to the patterned Barcode A stripes to add a second set of Barcodes B molecules (FIG. 1, Barcode B's: B1, B2, B3 . . . B200), each having a poly(dT) sequence at the 3′ end for mRNA capture, thereby forming “Barcode B stripes.” This was achieved using a primer extension reaction where the Barcode B molecules were first hybridized to the previously patterned Barcode A molecules via a complementary anchor sequence (see, e.g., FIG. 3 step 2). After hybridization for more than 1 hour, the microfluidic patterning chip was removed, unbound barcodes were washed away and the hybridized barcodes were extended by DNA polymerase I (see, e.g., FIG. 3 step 3). The reaction was then quenched with sodium hydroxide to remove the shorter strands (see, e.g., FIG. 3 step 4). The resulting immobilized nucleic acid strand included both Barcode A and Barcode B sequences terminated at the 3′ end with a poly(dT) tail for mRNA capture.

(105) An array of 200×200 (40,000) square distinct barcode patches were formed by overlapping Barcode A and Barcode B areas on the slide (intersection of A and B) (FIGS. 1 and 4). This array interfaces with a single-cell-loaded microchamber array for capture of single-cell-derived messenger RNAs for whole transcriptome sequencing. The stripes are 100 um in width and, thus, the barcodes patches are 100 μm X 100 μm. All molecular barcodes were chosen from a validated oligomer library and all the sequences in this database are orthogonal to human mRNAs.

Example 2. Cell Capture Platforms

(106) Microwell devices (individual wells˜picoliter to nanoliter volume) were used as single cell capture platforms. Two types of microwell devices were used (FIG. 5). The first device has microwells created on the surface of a substrate (glass, polymer or plastic). This device can be used with the alignment-free strategy described in Example 3. The second device is a through-hole microwell array created in a thin substrate (glass, polymer or plastic) which can be used with the deterministic alignment strategy described in Example 4.

Example 3. An Alignment-Free Strategy for Interfacing Single Cells with a High-Density Barcode Array

(107) For reliable mRNA capture from single cells using the high density barcode array, each barcoded patch is interfaced with a single cell. At least two techniques may be used to interface the microwell device with a barcode array: an alignment-free technique, described in this Example, and a deterministic alignment technique, described in Example 4.

(108) With the alignment-free technique, the first version of microwell arrays is used, and the microwell dimensions and cell loading protocols are set such that when the high-density barcode array is randomly overlaid on top of the microwell devices, 10-30% of the barcodes were interfaced with single cells (FIG. 6). The microwell dimensions and cell loading densities are chosen such that when cells are randomly loaded into microwells, 50-60% of the wells receive a single cell while the rest either remains empty or receives more than one cell (based on Poisson distribution). The size of wells is also such that only 1 to 4 individual wells can fit into a footprint of 100×100 μm.sup.2 region (corresponding to the size of barcoded patches). With these dimensions, when the high-density barcode array is overlaid on top of the microwell device without alignment, 10-30% of the barcoded patches are interfaced with single cells, while the rest of the barcodes either get no cell or more than one cell. The barcode locations corresponding to the single cells are recorded using imaging. For this purpose, the barcoded oligos are pre-tagged with a fluorophore labeled hybridization probe and their locations are determined using fluorescence microscopy once the microwell and barcode arrays are overlaid together. Microwell locations are determined using bright field imaging. The corresponding wells and barcodes are matched using a custom software. The overall random cell loading and alignment-free approach enables a straightforward operation to utilize the high-density barcode arrays.

Example 4. A Deterministic Alignment Strategy for Interfacing Single Cells with High-Density Barcode Array

(109) In this approach, the second version of microwell arrays (through-hole) are used and the microwells are aligned onto the barcode array using a precision alignment tool (FIG. 7). The alignment tool provides translational control in three dimensions (x, y and z) along with rotational control to controllably match the individual microwells to barcode array and bring the two substrates together for bonding. The microwell array is designed with similar pitch size as in a barcode array, such that each barcode patch contains only one microwell after proper alignment. During alignment, the barcode locations are determined using fluorescently labeled probes as described in Example 3, while the microwell locations are determined using the bright field images. Overall, this approach generates a microwell platform where each well is pre-patterned with a distinct barcode. For experiments, cells can be loaded randomly as described Example 3 or using a more deterministic approach as described in Example 5.

(110) An alternative deterministic approach was also developed (FIG. 8). Instead of aligning and bonding the through-hole microwells onto barcode array patterned on glass slides, the barcodes A and B are patterned directly onto a microwell array fabricated in different substrates (glass, plastic or polymer) following the similar procedures described in Example 1. For this purpose, the microfluidic flow patterning chips are first aligned onto the microwell arrays such that flow channels are overlaid onto the wells and barcodes are immobilized inside the individual wells. Cell loading can be performed based on random distribution or using a more deterministic approach as in Example 5.

Example 5. Deterministic Cell Trapping and Transfer Approach for High-Throughput Interfacing of ˜30,000 Single Cells to High-Density Barcode Array

(111) While the random cell loading such as the one described in Example enables a straightforward operation, a more deterministic approach helps improve throughputs by ensuring that almost all barcoded patches are interfaced with a single cell. For this purpose, a cell trapping and transfer method to reliable placing single cells into microwell/chambers for molecular analysis is demonstrated (FIG. 9), which involves two set of microwell devices, one where the width of microwells is approximately equal to the diameter of cells of interest and another with larger-sized microwells. Once the small microwell PDMS chip is saturated with cells, the cell enter the microwells capturing only one cell per well due to size exclusion. Afterwards, extra cells are washed out and this small microwell with cells trapped is placed against the array of larger-sized microchambers where cells are transferred to the larger chambers by gravity and pushing the back of the small microwell silicone chip. This process can be optimized and repeated such that nearly all microwells (>95%) are occupied by single cells. This process thereby enables improved throughputs while at the same time using larger wells (to provide an increased surface area coated with barcoded oligos for more efficient mRNA capture). Approximately 30,000 single cell microchambers per chip can be achieved, and this performance is well above the requirement and ensures ultra-high-throughput single-cell transcriptome analysis.

Example 6. Cell Lysis and mRNA Capture from Single Cells

(112) Once the cells are captured in microwells and overlaid with the high-density barcode array (or sealed with a second glass slide on top in case of devices described in Example 4), they can be lysed using a few cycles of freezing and thawing (FIG. 10). During the freeze-thaw cycles, cell membranes are broken due to ice-formation and intracellular contents are released into microwells. Similarly, more selective cell lysis methods can also be used to controllably lyse the cytoplasm only (or both cytoplasm and nucleus) by controlling the concentration of detergents used in the lysis solution. Alternatively, w the dried lysis buffer may be spray coated onto the high-density barcode array (or another sealing glass slide) to initiate lysis once the microwells are sealed. Upon lysis, the devices are incubated for 1 h to capture the mRNA onto the barcoded oligos terminated with poly(dT) sequences.

Example 7. Library Preparation, Sequencing, and Bioinformatics

(113) Library preparation follows the Cel-seq2 protocol. After the mRNA capture, the mRNA is reverse-transcribed followed by second strand synthesis. The generated cDNA is then in vitro transcribed to amplify the material captured. Amplified RNA (aRNA) is reverse transcribed and sequencing adapters are added through PCR amplification to finalize the libraries. The quality of library preparation is checked using a high sensitivity bioanalyzer (FIG. 11). The sequencing is done using the Illumina sequencers and data is analyzed using both available and custom bioinformatics tools.

Example 8. Multi-Omic Measurements

(114) The approach described here affords the capability to obtain polyomic measurements from same single cells. This can be achieved either through serial measurements where, for example, first up to 45 secreted proteins can be measured by interfacing the microwells with an antibody barcode followed by transcriptomic measurement by interfacing the same cells with high-density barcode array for mRNA capture (FIG. 12). It can also be achieved through parallel measurements by interfacing the microwell array with two glass slides, one is the high-density barcode array for transcriptomic measurements and the other is a second glass slide for protein, microRNA or epigenetic measurements (FIGS. 13 and 14). In addition, this approach allows for imaging of individual cells trapped in each microchamber and thus permits simultaneous measurement of live cell behaviors (size, morphology, migration, etc.) in conjunction with gene expression and proteomic profiling, all on the same single cells.

Example 9. Barcode and Sequence Design

(115) Row and column barcodes are separated by a constant sequence. Row barcodes are designed to be between 8 and 11 bases, such that the constant region will shift a base with each longer row barcode and this will prevent any sequencing issues related to sequencing problems with constant regions.

(116) Individual sequences containing row barcodes and column barcodes

(117) TABLE-US-00001 Barcode Name Sequence Barcodes complement Row_V1_ CGATTGAGCCGGTTTTTTTAAGCAGTGGTATC AGTACATC GATGTACT 8 bp AACGCAGAGTACAGTACATCGAGTGATTGCT TGTGACG (SEQ ID NO: 1) Row_V1_ CGATTGAGCCGGTTTTTTTAAGCAGTGGTATC CACGTCAGT ACTGACGTG 9 bp AACGCAGAGTACCACGTCAGTGAGTGATTGC TTGTGACG (SEQ ID NO: 2) Row_V1_ CGATTGAGCCGGTTTTTTTAAGCAGTGGTATC GTACGTGAGC GCTCACGTAC 10 bp AACGCAGAGTACGTACGTGAGCGAGTGATTG (SEQ ID (SEQ ID CTTGTGACG (SEQ ID NO: 3) NO: 9) NO: 11) Row_V1_ CGATTGAGCCGGTTTTTTTAAGCAGTGGTATC TCGTAGCTCGT ACGAGCTACGA 11 bp AACGCAGAGTACTCGTAGCTCGTGAGTGATT (SEQ ID (SEQ ID GCTTGTGACG (SEQ ID NO: 4) NO: 10) NO: 12) Column_ AAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AGCTCGTA TACGAGCT V1_1 NNNNNNNNTACGAGCTGTCATCAGCGTCACA AGCAATCACTC (SEQ ID NO: 5) Column_ AAAAAAAAAAAAAAAAAAAAAAAAAAAAAA CTGAGTCG CGACTCAG V1_2 NNNNNNNNCGACTCAGGTCATCAGCGTCACA AGCAATCACTC (SEQ ID NO: 6) Column_  AAAAAAAAAAAAAAAAAAAAAAAAAAAAAA GCTCACAT ATGTGAGC V1_3 NNNNNNNNATGTGAGCGTCATCAGCGTCACA AGCAATCACTC (SEQ ID NO: 7) Column_ AAAAAAAAAAAAAAAAAAAAAAAAAAAAAA V1_4 NNNNNNNNGCGACATAGTCATCAGCGTCACA TATGTCGC GCGACATA AGCAATCACTC (SEQ ID NO: 8)
Fully extended sequences with both row and column barcode sequences

(118) TABLE-US-00002 Full 1 CGATTGAGCCGGTTTTTTTAAGCAGTGGTATCAACGCAGA GTACAGTACATCGAGTGATTGCTTGTGACGCTGATGACAG CTCGTANNNNNNNNTTTTTTTTTTTTTTTTTTTTTTTTTT TTTT (SEQ ID NO: 13) Full 2 CGATTGAGCCGGTTTTTTTAAGCAGTGGTATCAACGCAGA GTACCACGTCAGTGAGTGATTGCTTGTGACGCTGATGACC TGAGTCGNNNNNNNNTTTTTTTTTTTTTTTTTTTTTTTTT TTTTT (SEQ ID NO: 14) Full 3 CGATTGAGCCGGTTTTTTTAAGCAGTGGTATCAACGCAGA GTACGTACGTGAGCGAGTGATTGCTTGTGACGCTGATGAC GCTCACATNNNNNNNNTTTTTTTTTTTTTTTTTTTTTTTT TTTTTT (SEQ ID NO: 15) Full 4 CGATTGAGCCGGTTTTTTTAAGCAGTGGTATCAACGCAGA GTACTCGTAGCTCGTGAGTGATTGCTTGTGACGCTGATGA CTATGTCGCNNNNNNNNTTTTTTTTTTTTTTTTTTTTTTT TTTTTTT (SEQ ID NO: 16)

(119) All references, patents and patent applications disclosed herein are incorporated by reference with respect to the subject matter for which each is cited, which in some cases may encompass the entirety of the document.

(120) The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.”

(121) It should also be understood that, unless clearly indicated to the contrary, in any methods claimed herein that include more than one step or act, the order of the steps or acts of the method is not necessarily limited to the order in which the steps or acts of the method are recited.

(122) In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of” and “consisting essentially of” shall be closed or semi-closed transitional phrases, respectively, as set forth in the United States Patent Office Manual of Patent Examining Procedures, Section 2111.03.