Methods for identifying disease-associated RNA and RNA binding protein (RBP) interactions for diagnostic and therapeutic selection

Abstract

The present disclosure provides methods for simultaneously identifying dynamic and disease- relevant RNA binding protein (RBP)-RNA interaction (PRI) sites throughout the transcriptome at single-nucleotide resolution. The methods disclosed herein recapitulate the PRI profiles obtained with several different conventional PRI methods with several orders of magnitude improvement in efficiency in a single assay and allow identification of the specific RBP bound at the PRI.

Description

[0001] Cross-reference to related applications

[0002] This application claims the benefit and priority of U.S. Provisional Patent Application No. 63 / 606,483, filed December 5, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to methods for simultaneously identifying dynamic and disease-associated RNA-binding protein (RBP)-RNA interaction (PRI) sites across the entire transcriptome at single-nucleotide resolution. The methods disclosed herein reproduce PRI profiles obtained using several different conventional PRI methods with orders of magnitude greater efficiency in a single assay and allow for the identification of specific RBPs binding at PRI sites.

[0004] Government support

[0005] This invention was completed with government support, based on GM124909 granted by the National Institutes of Health (NIH). The government holds certain rights to this invention. Background Technology

[0006] The following description of the technical background of the present invention is provided only to help understand the technology of the present invention, and does not acknowledge any prior art that describes or constitutes the technology of the present invention.

[0007] Site-specific interactions between RNA-binding proteins (RBPs) and regulatory RNA sequences control RNA stability, processing, and translation—processes crucial for the function of many genes essential for cancer cell survival and proliferation, as well as pathogenic processes in many benign diseases. 1-4 However, little is known about these interactions due to major methodological limitations in their identification. Current methods for identifying pathogenic interactions are largely limited to protein-centric approaches that identify interactions between a single RBP and RNA, or conversely, RNA-centric approaches that map multiple RBP binding sites on a single RNA molecule. 5-12 This presents a significant bottleneck, preventing the annotation of cancer-related interactions at the scale required for the speed needed to effectively identify diagnostic and therapeutic targets. Since RBP-RNA interactions (PRI) are, in principle, druggable, the major limitations of current approaches may lead to a severe underestimation of druggable mutations in cancer and therapeutic targets in benign diseases.

[0008] Therefore, there is an urgent need for high-throughput, unbiased, and efficient methods to identify dynamic and disease-associated PRIs at single-nucleotide resolution across the entire transcriptome. Summary of the Invention

[0009] In one aspect, this disclosure provides a method for identifying one or more RNA-binding protein (RBP)-RNA interaction (PRI) sites in at least one transcript derived from a biological sample comprising a plurality of cells, the method comprising: (a) crosslinking the plurality of cells present in the biological sample with short-wavelength ultraviolet light to stabilize a plurality of RBP-RNA complexes containing RNA transcripts, wherein each RBP-RNA complex corresponds to a PRI; (b) isolating RNA molecules from the crosslinked plurality of cells, wherein the RNA molecules comprise unbound RNA molecules and the RNA transcripts of the plurality of RBP-RNA complexes; (c) contacting the isolated RNA molecules with a protease under conditions of cleaving the RBPs of the plurality of RBP-RNA complexes to produce RNA transcripts bound by residual peptides, wherein each bound residual peptide (i) corresponds to a PRI site and (ii) comprises an α-amino group at its N-terminus and a carboxyl group at its C-terminus; (d) labeling the RNA transcripts bound by the residual peptides by coupling the α-amino group or the carboxyl group of each bound residual peptide to a chemically conjugate comprising an affinity ligand; (e) subjecting the RNA transcripts to heat and In the presence of a divalent cation, the RNA molecule is fragmented to generate an RNA fragment with a 5' end and a 3' end, wherein the RNA fragment comprises an unlabeled RNA fragment and an RNA fragment labeled with the chemically conjugate containing the affinity ligand; (f) a sequencing adaptor is ligated to the 3' end of the RNA fragment to generate an RNA fragment tagged with the adaptor; (g) the RNA fragment tagged with the adaptor and labeled with the chemically conjugate containing the affinity ligand is captured using affinity purification; (h) the tagged RNA fragment is reverse transcribed to generate multiple c DNA molecules, optionally wherein the tagged RNA fragment with an adaptor tag is attached to a solid surface or in solution; (i) a sequencing adaptor is ligated to the 3' end of the cDNA molecule to generate a plurality of cDNA molecules with an adaptor tag; (j) the plurality of cDNA molecules with an adaptor tag are amplified to generate a nucleic acid library of transcripts with RNA-RBP interactions; (k) the plurality of cDNA molecules with an adaptor tag in the nucleic acid library are sequenced and aligned to identify individual transcripts; (l) the frequency of the cDNA 3' end at each nucleotide in the individual transcript is determined; and (m) when the frequency of the cDNA 3' end at a specific nucleotide in the individual transcript meets a predetermined threshold, enriched RBP binding sites in the individual transcript are detected. In some embodiments, the plurality of cDNA molecules are reverse transcribed using a truncated reverse transcriptase (RT) or a read-through reverse transcriptase (RT). A truncated RTase is formed at the PRI site (i.e., the cross-linking site) or a readthrough RTase with an SNV or insertion / deletion is formed at the PRI site (i.e., the cross-linking site).Alternatively or alternatively, in some embodiments, the methods disclosed herein further include identifying in the individual transcript (i) short insertions or deletions (e.g., 1-3 bp) generated by PRI isolated by affinity purification, (ii) single nucleotide variants (SNVs) generated by PRI isolated by affinity purification, or (iii) reverse transcriptase (RT) termination sites generated by PRI isolated by affinity purification.

[0010] The biological sample may be derived from breast tissue, kidney tissue, cervical tissue, endometrial tissue, head or neck tissue, gallbladder tissue, parotid gland tissue, prostate tissue, brain tissue, pituitary gland tissue, kidney tissue, muscle tissue, esophageal tissue, stomach tissue, small intestine tissue, colon tissue, liver tissue, spleen tissue, pancreas tissue, thyroid tissue, heart tissue, lung tissue, bladder tissue, adipose tissue, lymph node tissue, uterine tissue, ovarian tissue, adrenal gland tissue, testicular tissue, tonsils, thymus, blood, hair, oral cavity, skin, serum, plasma, CSF, semen, prostatic fluid, seminal plasma, urine, feces, sweat, saliva, sputum, mucus, bone marrow, lymph, or tears. In some embodiments, the biological sample is a fresh or frozen sample.

[0011] In some embodiments, the crosslinking UV box operates at approximately 50-600 mJ / cm. 2 The dosage is applied for approximately 30 seconds to 10 minutes. In some embodiments, the crosslinking UV box is used at approximately 50 mJ / cm². 2 Approximately 100 mJ / cm 2 Approximately 150 mJ / cm 2 Approximately 200 mJ / cm 2 Approximately 250 mJ / cm 2 Approximately 300 mJ / cm 2 Approximately 350 mJ / cm 2 Approximately 400 mJ / cm 2 Approximately 450 mJ / cm 2 Approximately 500 mJ / cm 2 Approximately 550 mJ / cm 2 or about 600 mJ / cm 2 The dosage is applied for approximately 30 seconds, 45 seconds, 1 minute, 1.5 minutes, 2 minutes, 2.5 minutes, 3 minutes, 3.5 minutes, 4 minutes, 4.5 minutes, 5 minutes, 5.5 minutes, 6 minutes, 6.5 minutes, 7 minutes, 7.5 minutes, 8 minutes, 8.5 minutes, 9 minutes, 9.5 minutes, or approximately 10 minutes. In some embodiments, the crosslinking UV box operates at 300-500 mJ / cm². 2 The dose is administered over approximately 3 minutes or at a rate of 100-400 mJ / cm. 2The dosage is applied for approximately 10 minutes. In other embodiments, the crosslinking is performed using a UV laser for approximately 10-20 seconds.

[0012] Alternatively or concurrently, in some embodiments, the isolated RNA molecules are treated with the protease at approximately 25°C–65°C for approximately 5 to 60 minutes. In some embodiments, the isolated RNA molecules are treated with the protease at approximately 25°C, approximately 27.5°C, approximately 30°C, approximately 32.5°C, approximately 35°C, approximately 37.5°C, approximately 40°C, approximately 42.5°C, 45°C, approximately 47.5°C, approximately 50°C, approximately 52.5°C, approximately 55°C, approximately 57.5°C, approximately 60°C, approximately 62.5°C, or approximately 65°C for approximately 5 minutes, approximately 7.5 minutes, or approximately 10 minutes, respectively. The time intervals are approximately 12.5 minutes, 15 minutes, 17.5 minutes, 20 minutes, 22.5 minutes, 25 minutes, 27.5 minutes, 30 minutes, 32.5 minutes, 35 minutes, 37.5 minutes, 40 minutes, 42.5 minutes, 45 minutes, 47.5 minutes, 50 minutes, 52.5 minutes, 55 minutes, 57.5 minutes, or 60 minutes. In some embodiments, the isolated RNA molecule is treated with the protease at 50°C for approximately 30 minutes. In some embodiments, the protease is proteinase K. Alternatively or in some embodiments, the length of each bound residual peptide is no more than 10 amino acids, no more than 9 amino acids, no more than 8 amino acids, no more than 7 amino acids, no more than 6 amino acids, no more than 5 amino acids, no more than 4 amino acids, no more than 3 amino acids, no more than 2 amino acids, or a single amino acid.

[0013] Alternatively or additionally, in some embodiments, the method of this technology further includes removing genomic DNA after step (c) to obtain purified RNA molecules. In some embodiments, the method disclosed herein further includes contacting the purified RNA molecules with a protease at 50°C for about 30 minutes prior to step (d), optionally wherein the protease is proteinase K.

[0014] In any of the foregoing embodiments, the method of this technology further includes enriching the isolated or purified RNA molecule with a poly-dT oligonucleotide or a targeting decoy capture reagent after step (c) but before step (d). In other embodiments, the method of this technology further includes enriching the cDNA molecule with an adaptor tag with a targeting decoy capture reagent after step (j) but before step (k). In some embodiments, the targeting decoy capture reagent hybridizes with one or more disease-related genes, such as genes associated with cancer, neurodegenerative diseases, autoimmune diseases, metabolic diseases, etc. Examples of disease-related genes include, but are not limited to, AKT1, ALK, APC, AR, ARAF, ARID1A, ARID2, ATM, B2M, BCL2, BCOR, BRAF, BRCA1, BRCA2, CARD11, CBFB, CCND1, CDH1, CDK4, CDKN2A, CIC, CREBBP, CTCF, CTNNB1, DICER1, DIS3, DNMT3A, EGFR, EIF1AX, EP300, ERBB2, ER BB3, ERCC2, ESR1, ETV6, EZH2, FBXW7, FGFR1, FGFR2, FGFR3, FGFR4, FLT3, FOXA1, FOXL2, FOXO1, FUBP1, GATA3, GNA1 1. GNAQ, GNAS, H3F3A, HIST1H3B, HRAS, IDH1, IDH2, IKZF1, INPPL1, JAK1, KDM6A, KEAP1, KIT, KNSTRN, KRAS, MAP2K1 , MAPK1, MAX, MED12, MET, MLH1, MSH2, MSH3, MSH6, MTOR, MYC, MYCN, MYD88, MYOD1, NF1, NFE2L2, NOTCH1, NRAS, NTR K1, NTRK2, NTRK3, NUP93, PAK7, PDGFRA, PIK3CA, PIK3CB, PIK3R1, PIK3R2, PMS2, POLE, PPP2R1A, PPP6C, PRKCI, PTC H1, PTEN, PTPN11, RAC1, RAF1, RB1, RET, RHOA, RIT1, ROS1, RRAS2, RXRA, SETD2, SF3B1, SMAD3, SMAD4, SMARCA4, SMA RCB1, SOS1, SPOP, STAT3, STK11, STK19, TCF7L2, TGFBR1, TGFBR2, TP53, TP63, TSC1, TSC2, U2AF1, VHL, XPO1 and TERT.

[0015] In any of the above embodiments, the method of this technology further includes depleting the ribosomal RNA from the isolated or purified RNA molecule after step (c) but before step (d).

[0016] In any and all embodiments of the methods disclosed herein, labeling the RNA transcript bound by the residual peptide comprises thermally denaturing the isolated or purified RNA molecule and incubating the isolated or purified RNA molecule with the chemically conjugate containing the affinity ligand in a reaction buffer supplemented with about 5%–70% DMSO (e.g., 50% DMSO). Examples of suitable reaction buffers include, but are not limited to, phosphate buffers (e.g., PBS buffer), carbonate buffers (e.g., carbonate-bicarbonate buffer), HEPES, borate, 2-[morpholino]ethanesulfonic acid (MES), or any buffer free of free amine or free carboxylate groups. Additionally or alternatively, in some embodiments, thermal denaturation comprises heating the isolated or purified RNA molecule at about 50°C–100°C for 1–5 minutes. In some embodiments, thermal denaturation comprises heating the isolated or purified RNA molecule at 70°C for 2 minutes.

[0017] In some embodiments of the methods disclosed herein, the chemically moietraconjugate is coupled to the α-amino group of each bound residual peptide. In some embodiments, the chemically moietraconjugate comprises a succinimide ester, a carboxylic acid ester, a tetrafluorophenyl ester, a sulfonyl dichlorophenol ester, a carbonyl azide, an aldehyde, or an isothiocyanate. In other embodiments of the methods disclosed herein, the chemically moietraconjugate is coupled to the carboxyl group of each bound residual peptide. In some embodiments, the chemically moietraconjugate comprises a primary amine, a carbodiimide, or an isocyanate. Alternatively or additionally, in some embodiments, the chemically moietraconjugate containing the affinity ligand is added to a final concentration of about 0.1 mM to about 10 mM and incubated for about 3 minutes to about 5 hours.

[0018] In any and all embodiments of the methods disclosed herein, the affinity ligand comprises biotin, a biotin derivative, digoxigenin, dinitrophenyl, a click chemistry reagent, a sugar, or a peptide. In some embodiments, the click chemistry reagent comprises an alkyne group, an azide group, a dibenzocyclooctyne group (DBCO), or a bicyclic nonyne (BCN) group.

[0019] Alternatively or alternatively, in some embodiments, fragmenting the RNA molecule includes using approximately 10 mM-30 mM of a divalent cation (e.g., Mg). 2+ Zn 2+In the presence of [a specific substance], the RNA molecule is heated at 90°C-95°C for approximately 3-12 minutes. In some embodiments, fragmentation of the RNA molecule includes using approximately 10 mM, approximately 15 mM, approximately 20 mM, approximately 25 mM, or approximately 30 mM of divalent cations (e.g., Mg). 2+ Zn 2+ In the presence of [a specific substance], the RNA molecule is heated at 90°C, 91°C, 92°C, 93°C, 94°C, or 95°C for approximately 3 minutes, 4 minutes, 5 minutes, 6 minutes, 7 minutes, 8 minutes, 9 minutes, 10 minutes, 11 minutes, or 12 minutes, respectively. The length of the RNA fragment can be approximately 40-90 nucleotides or approximately 80-250 nucleotides.

[0020] In any and all embodiments of the methods disclosed herein, the biological sample is obtained from a subject diagnosed with a disease. In some embodiments, the disease is cancer, an autoimmune disease, a metabolic disease, or a neurodegenerative disease. Examples of cancer include, but are not limited to, adrenal cancer, bladder cancer, leukemia, bone cancer, brain cancer, breast cancer, carcinoma, cervical cancer, colon cancer, colorectal cancer, endometrial cancer, ear, nose, and throat (ENT) cancer, endometrial cancer, esophageal cancer, gastrointestinal cancer, head and neck cancer, Hodgkin's disease, colon cancer, kidney cancer, laryngeal cancer, leukemia, liver cancer, lymph node cancer, lymphoma, diffuse large B-cell lymphoma (DLBCL), lung cancer, melanoma, mesothelioma, myeloma, nasopharyngeal carcinoma, neuroblastoma, non-Hodgkin's lymphoma, oral cancer, ovarian cancer, pancreatic cancer, penile cancer, pharyngeal cancer, prostate cancer, rectal cancer, sarcoma, seminoma, skin cancer, gastric cancer, teratoma, testicular cancer, thyroid cancer, uterine cancer, vaginal cancer, vascular tumors and their metastases.

[0021] Examples of neurodegenerative diseases include, but are not limited to, age-related memory impairment (AAMI), mild cognitive impairment (MCI), Alzheimer's disease, Down syndrome, boxer's dementia, cognitive dysfunction syndrome, multiple system atrophy, inclusion body myositis, Dutch hereditary cerebral hemorrhage with amyloidosis, Niemann-Pick disease type C, cerebral β-amyloid angiopathy, dementia associated with corticobasal degeneration, amyloidosis, Creutzfeldt-Jakob disease, Gerstmann-Strauss syndrome, kuru, pruritus, Huntington's disease, Parkinson's disease, ataxia, motor neuron disease, progressive supranuclear palsy, and amyotrophic lateral sclerosis (ALS).

[0022] Examples of autoimmune diseases include, but are not limited to, vasculitis such as antineutrophil cytoplasmic antibodies (ANCA), ANCA-associated vasculitis (AAV) or giant cell arteritis (GCA), Sjögren's syndrome, inflammatory bowel disease (IBD), pemphigus vulgaris, lupus nephritis, psoriasis, thyroiditis, type 1 diabetes mellitus, idiopathic thrombocytopenic purpura (ITP), ankylosing spondylitis, multiple sclerosis, systemic lupus erythematosus (SLE), rheumatoid arthritis, Crohn's disease, myasthenia gravis, neuromyelitis optica (NMO), IgG4-related disease, systemic sclerosis, insulin-dependent diabetes mellitus (IDDM), ankylosing spondylitis, atopic dermatitis, uveitis, and graft-versus-host disease (GVHD).

[0023] Examples of metabolic disorders include, but are not limited to, familial hypercholesterolemia, Gaucher disease, Hunter syndrome, Krab disease, maple syrup diabetes, metachromatic leukodystrophy, mitochondrial encephalopathy, lactic acidosis-like stroke (MELAS), Niemann-Pick disease, phenylketonuria (PKU), porphyria, Ty-Sachs disease, and Wilson's disease.

[0024] In any and all embodiments of the methods disclosed herein, one or more of the enriched RBP binding sites detected in the individual transcript are allele-specific and / or associated with diseases such as cancer, neurodegenerative diseases, autoimmune diseases, metabolic diseases, etc.

[0025] In any of the foregoing embodiments, the method of this technology further includes identifying the RBPs that bind to one or more of the enriched RBP binding sites detected in the individual transcript. In some embodiments, the identity of the RBPs that bind to one or more of the detected enriched RBP binding sites in the individual transcript is determined by identifying RBP-specific crosslinking patterns within sequence motifs in the individual transcript (e.g., via computational analysis). In some embodiments, the RBP crosslinking patterns are identified using shared motifs defined by eCLIP, mCross crosslinking patterns, position-enriched k-mer analysis (PEKA), or sequence motifs defined by RBNS.

[0026] In one aspect, this disclosure provides a method for evaluating the efficacy of a drug in inhibiting the interaction between a specific RNA-binding protein (RBP) and an RNA transcript, the method comprising (a) contacting a first cell population with at least one repressive oligonucleotide that specifically inhibits the expression of RBP, wherein the first cell population is obtained from a biological sample; (b) contacting a second cell population with a test drug, wherein the second cell population is obtained from the biological sample and has not been treated with the at least one repressive oligonucleotide; (c) performing the ARORA method described herein to identify one or more RBP-RNA interaction (PRI) sites in at least one transcript derived from the first cell population, the second cell population, and a control cell population, wherein the control cell population is obtained from the biological sample and has not been treated with the at least one repressive oligonucleotide and the test drug; (d) defining an RBP-specific PRI by identifying a crosslinking pattern within one or more sequence motifs in the individual transcript, the crosslinking pattern being (i) present in the control cell population and (ii) absent in the first cell population; and determining that the test drug is effective when the RBP-specific PRI identified in step (d) is absent in the second cell population. In another aspect, this disclosure provides a method for evaluating the specificity of a drug in inhibiting the interaction between a specific RNA-binding protein (RBP) and RNA transcripts, the method comprising: (a) contacting a first cell population with at least one repressive oligonucleotide that specifically inhibits the expression of RBP, wherein the first cell population is obtained from a biological sample; (b) contacting a second cell population with a test drug, wherein the second cell population is obtained from the biological sample and has not been treated with the at least one repressive oligonucleotide; (c) performing the ARORA method described herein to identify one or more RBP-RNA interaction (PRI) sites in at least one transcript derived from the first cell population, the second cell population, and a control cell population, wherein the control cell population is obtained from the biological sample and has not been treated with the at least one repressive oligonucleotide and the test drug; and (d) identifying the nonspecific effect of the test drug by identifying a crosslinking pattern within one or more sequence motifs in the individual transcript, the crosslinking pattern being (a) present in the control cell population and the first cell population and (b) absent in the second cell population. The at least one repressive oligonucleotide may be siRNA, shRNA, antisense oligonucleotide, or sgRNA. In some embodiments, the test drug is a small molecule, nucleic acid, or peptide. Attached Figure Description

[0027] Figures 1A to 1I : A general orthogonal labeling strategy for protein-nucleic acid interactions. Figure 1A: Schematic diagram of universal labeling for amino acid-nucleic acid crosslinking. PRI, stabilized by UVC crosslinking, undergoes thorough proteolysis, generating a free N-terminal α-amino group and a free C-terminal carboxyl group regardless of the peptide sequence. These are orthogonal chemical portions of the nucleic acid sequence, which can be selectively labeled with existing chemicals, including N-hydroxysuccinimide labeling of the α-amino group. Figure 1B Dot blots showing the binding of streptavidin-IR800 to NHS-biotinylated RNA isolated from UVC cross-linked cells, RNA cross-linked with UVC in vitro in the absence of proteins, and RNA from non-cross-linked cells. Figure 1C Denatured agarose gel electrophoresis of NHS-biotinylated total RNA. Right panel: Streptavidin-IR800 binds to NHS-biotinylated total RNA from UVC cross-linked cells. Figure 1D Quantitative fluorescence measurements of NHS-Alexa488-labeled RNA isolated from cells after cross-linking with a specified dose of UVC. Figure 1E qPCR enrichment of selected transcripts was performed by increasing UVC dose via NHS biotinylation and streptavidin pull-down. Lack of recovery in the no-RT control (-RT) confirmed that RNA, rather than DNA, was recovered. Figure 1F Left panel: Infrared imaging of primer extension assays performed on a 100 bp oligonucleotide with a simulated crosslink (XL) at position 70, using the indicated RT-enzyme. The specific RT-enzymes used are listed above. (FL, full-length sequence; FL-del, full-length sequence with deletion; XL, crosslinked truncated sequence). Right panel: Quantification of the relative abundance of FL, FL-del, and XL fragments obtained with each RT-enzyme. Figure 1G : Deletion rate (left) and single nucleotide substitution rate (right) under four conditions of ARORA (NCL_IN, uncrosslinked RNA input; x400_IN, crosslinked RNA input; NCL_PD, uncrosslinked RNA after pull-down with streptavidin beads; x400_PD, crosslinked RNA after pull-down with streptavidin beads). Figure 1H A schematic diagram of identifying PRI in ARORA by identifying both significantly enriched cross-linked RT termination sites and cross-linked mutation sites. Figure 1I : Secondary structure diagram of PRI identified by RNP-MaP (left) and ARORA (right). Red annotations indicate significant RT termination sites, and blue annotations indicate significant crosslink-induced mutations at residues with significant RT termination sites.

[0028] Figures 2A to 2D ARORA identified dynamic, stress-induced PRI and variability in ribosome structure associated with different biological states. Figure 2AGraphs of helices 13 and 14 of the small ribosomal subunit (18S) and PRI identified by cryo-EM (left), ARORA (middle), and CLIP-seq (right). ARORA was performed under normal, non-stress conditions (untreated control) and after ribosomal stress (actinomycin or anisoxin). Black annotations indicate non-variable PRI associated with the RP binding site on 18S. Red annotations indicate dynamic, stress-induced PRI on h14, which is precisely mapped to the ZAKα binding site identified by CLIP-seq (red line, right panel). Figure 2B : PRI correlation matrix in approximately 1000 nt segments of 18S rRNA from 14 breast cancer cell lines. (Red = 1, White = 0, Blue = -1). Highly PRI-associated cohesive domains (red nodes) are marked and indicated by dense regions of RP-bound rRNA nucleotides (gray lines). Blue: PRI variation among the 14 breast cancer cell lines. Below: Annotations of the 18S helices of interest and RBPs interacting with these sites. Figure 2C ARORA PRI signal within the ZAKα binding site on h14 of the 18S (Inset: Nucleotides plotted as secondary structure identifiers of the region). Mean PRI signal at each nucleotide in this region across seven BRCA cell lines investigated by CCLE and ARORA, five with high ZAKα PRI and two with low ZAKα PRI. *P<0.01. Figure 2D Area under the curve (AUC) of omaxitazine sensitivity for each of the seven cell lines. P < 0.02.

[0029] Figures 3A to 3F ARORA identifies allele-specific PRIs in regulatory RNAs. Figure 3A This diagram illustrates how the position of the preferred cross-link site relative to a single nucleotide variant affects the interpretation of allele-specific sequencing results from ARORA and when using an RT-enzyme that preferentially truncates at the cross-link site. When the cross-link preference occurs at the 5' cross-link site, the RT-enzyme can still capture the identity of the variant nucleotide before the RT is truncated at the cross-link site. However, if the cross-link preference occurs at the 3' of the variant nucleotide, the truncated read will not capture the identity of the variant nucleotide. The use of a read-through RT-enzyme overcomes this limitation by reading through the cross-link and capturing the identity of the variant nucleotide. Figure 3BAllele frequencies of the normal variant GAPDH rs1065691 and the HepG2-specific MYC 5'UTR variant C184T were determined under four ARORA conditions using the truncated RT enzyme Maxima H-. Crosslinking and pull-down conditions significantly depleted C184T, indicating allele-specific binding, but had no effect on GAPDH rs1065691. Figure 3C ARORA was detected by qPCR in the 3' region immediately adjacent to C184 in three cell lines. Significant enrichment was observed only in HepG2 cells. Figure 3D ARORA enrichment using AffinityScript revealed a region centered on C184T, which corresponds to the DDX3X eCLIP site in the 5'UTR of MYC. Figures 3E to 3F Allele fractions in ARORA input and drop-down samples in HepG2 cells using AffinityScript ( Figure 3E ) and allele enrichment ( Figure 3F ARORA using AffinityScript significantly enriches C184T, indicating that C184T achieves PRI compared to the wild type.

[0030] Figures 4A to 4E Key technical points of ARORA sequencing analysis. Figure 4A Dot blots of streptavidin-IR800 binding to EDC / amino-biotin-labeled RNA isolated from UVC cross-linked cells, RNA cross-linked in vitro with UVC in the absence of protein, and RNA from non-cross-linked cells. Figure 4B A schematic diagram of the ARORA sequencing workflow. Protein cross-linking sites are biotinylated (pink circles), and RNA is fragmented. After streptavidin enrichment, reverse transcriptase initiation is performed at the 3' linker with a unique barcode to identify the sample. NGS sequencing is performed, and PRI is identified using RT termination sites or cross-linked induced mutation sites. Figure 4C Nucleotide identification within + / -10 nucleotides at significantly enriched marker sites (RT termination, deletion, and substitution). U>C>G>A enriched at + / -1 positions of significantly enriched markers. Figure 4D The median relative ARORA signal (reactivity) of all nucleotides with specific identities in the NCL_PD and x400_PD samples indicates a slight nonspecific reactivity with A residues. Figure 4ENormalization from x400_PD to NCL_PD samples corrected for this modest artifact and allowed for comparisons between nucleotides. Notably, nucleotides significantly enriched at PRI typically had reactivity scores >10X to 1000X at this scale, indicating that the 1.4X bias for A in NCL_PD and x400_PD samples and the 1.6X bias for U in x400_PD samples were minor compared to the reactivity scale for significant PRI.

[0031] Figures 5A to 5C Optimization of RNA fragmentation in ARORA-tagged RNA. Figure 5A The denaturation PAGE of RNA labeled with ARORA after increasing the fragmentation time with divalent cation Mg2+ at 94℃. Figure 5B : A bioanalyzer for fragmenting RNA into 4-minute (long fragment) or 9-minute (short fragment) fragments for library construction. Gray bars indicate fragments in the range of 50-150 nucleotides. Figure 5C : Input (top image) and dropdown (bottom image) dsDNA ARORA library. The gray bar indicates a significant truncation of cDNA in the dropdown library compared to the library generated with the input material, as expected.

[0032] Figures 6A to 6C The effect of RT enzyme and RNA fragment size on RT termination identified by ARORA. Figure 6A : Correlation matrix of RT termination counts in ARORA libraries generated using total RNA when mapped to 45S (>95% reads). Three conditions were tested (AffinityScript RT using a long RNA fragment "4 min", Maxima H-RT using a long RNA fragment "4 min", and Maxima H-RT using a short RNA fragment "9 min"). NCL_IN, x400_IN, NCL_PD, and x400_PD libraries are color-coded. Bottom right: Four comparisons between the three conditions. Figure 6B Under each condition, the relevance of biological replicates of the x400_PD library was very high and did not change significantly with RT or fragment size (C1, i.e., reproducibility). Figure 6C Compared to AffinityScript, the correlation between x400_PD and negative controls was significantly reduced in libraries generated by Maxima H-, and further reduced by using short fragments instead of long fragments.

[0033] Figures 7A to 7C Detailed data on RT termination counts mapped to RMRP under various conditions, as well as signal robustness and appropriate sequencing depth. Figure 7AThe normalized RT termination count mapped to RMRP in four ARORA libraries. The vertical gray line indicates the corresponding... Figure 1I The RT termination sites are statistically significantly enriched on the RMRP. Figure 7B The original RT termination count at RMRP in two biological replicates of each of the four ARORA libraries. Figure 7C x400_PD RT termination counts in libraries with sequencing depths differing by nearly two orders of magnitude (sequencing depth in the mapped reads of each library listed on the right).

[0034] Figures 8A to 8E The low-depth whole transcriptome ARORA reproduced the PRI identified by the reference method. Figure 8A : Schematic diagram of data interpretation. In ARORA, the X400_PD sample generated with a truncated RT enzyme produces a peak in read stacking (read depth peak, gray), but PRI identifies it by a narrow window of highly enriched RT terminator (red) at the 5' end of the read depth peak. Figure 8B Compared to the three negative control libraries, the significantly enriched read depth peaks of x400_PD were evident even in intermediate / low abundance transcripts (such as G6PC). Some RT termination peaks could be resolved at this low depth if RT termination occurred in the short sequence space (grey bars), but more dispersed RT terminations were difficult to resolve at low sequencing depths. Figure 8C A full-length genome browser view of the abundant housekeeping gene ACTB reveals that most enrichment occurs in the 3' UTR (expanded above). Gray bars indicate sequences enriched in one or more of the eight RBPs >16X in the eCLIP, which significantly enrich the ACTB 3' UTR in HepG2 cells (equivalent to read depth peaks in ARORA). Multiple significant RT termination peaks enriched in the x400_PD sample can be seen within these eCLIP enriched regions and at the 5' end of each region. Figure 8D RT termination count in the high-abundance transcript RPL30 encoding the ribosomal protein L30. Highly enriched PRI is present in both the CDS and UTR of this transcript. Notably, high enrichment is observed in the 5' TOP domain of the 5' UTR, an essential regulator of RP translation, which controls ribosomal biogenesis and RP stoichiometry. Enrichment at the 5' TOP domain is observed in almost all transcripts encoding RP and several other translation regulators, but is rare in other mRNAs. Figure 8EThe RT termination peak and read depth peak in the 3' UTR of the TFRC, where the interaction with RBP has been well characterized by two existing methods, eCLIP and SHAPE footprint analysis. Above, a significantly enriched RT termination peak near the 5' end of the significantly enriched eCLIP sequence was identified. Peaks were also identified at the precise locations of iron reactive elements (IREs) as indicated by SHAPE footprint analysis of the ferroreactive protein.

[0035] Figure 9 RT termination counts associated with RP binding sites in small ribosomal subunits. ARORA analysis using total RNA identified significantly enriched RT termination peaks that were highly correlated with RP-interacting nucleotides as defined by cryo-EM (P < 10). -6 This figure depicts the 18S nucleotides in contact with the R-protein, as defined by Cryo-EM, in the small subunit region not visualized in Figure 2.

[0036] Figure 10 ARORA in frozen tissue samples reproduced the results of cell culture. Correlation matrices of ARORA biological repeatability were obtained from modified protocols of ARORA x400_IN (input) and x400_PD (drop-down) libraries of UVC-crosslinked HepG2 cells in culture (original method) and UVC-crosslinked material cells recovered from cryoprecipitates of HepG2 cells. These results demonstrate the consistency of ARORA in samples from both live and frozen cells, indicating the excellent potential of this method for ARORA in frozen clinical tumor samples.

[0037] Figures 11A to 11C ARORA identifies RBP interactions on RNA at single nucleotide resolution. Figure 11A A model of the interaction between RNA and three unique RBPs at different sites on the transcript. Figure 11B ARORA workflow and data output: (1) Cross-link cells and covalently label RBP-RNA interaction sites and remove proteins. (2) Reproducibly fragment covalently modified RNA with heat and divalent cations, followed by purification with RBP interaction sites labeled with affinity ligands. (3) Separate labeled fragments by affinity purification. (4) Connect sequencing adapters to the 3' end of purified RNA, followed by first-strand synthesis and library amplification. (5) Delineate RBP binding sites on RNA by strand-specific sequencing and first-strand alignment. (6) Calculated annotation of multiple significantly enriched RBP binding sites on individual RNAs. Figure 11CeCLIP workflow: (1) Cell lysates with cross-linked RBP-RNA complexes are treated with RNase to fragment the RNA; (2) RNA cross-linked with individual specific RBPs is immunoprecipitated with antibodies; (3) Sequencing adapters are ligated to the 3' end of the purified RNA, followed by first-strand synthesis and library amplification; (4) Strand-specific sequencing and first-strand alignment depict the sites where individual RBPs bind to RNA; (5) Computational annotation of interactions between individual RBPs and RNA.

[0038] Figures 12A to 12C ARORA defines the regional landscape of RBP-mRNA interactions. Figures 12A to 12C The genome browser trajectory of four conditions of reverse transcription truncation sites (RBP-mRNA crosslinking sites) was used by ARORA sequencing to define the ribosomal protein RPL30 in HepG2 cells. Figure 12A ), GAPDH Figure 12B ) and ACTB ( Figure 12C RBP-RNA interactions on the transcripts of [samples]. "Pulldown_x400" indicates a sample where RBP-mRNA interactions were cross-linked, biochemically modified, and enriched via affinity purification prior to sequencing. "Pulldown_NCL" indicates a sample that underwent biochemical modification and enrichment without prior cross-linking (negative control). "Input_x400" and "Input_NCL" represent libraries generated from RNA inputs under both conditions but without pull-down enrichment. Among all tracks, highly specific and unique sites in the "Pulldown_x400" sample represent sites of highly significant enrichment of RBP-mRNA interactions relative to both "Pulldown_NCL" and input samples. The x400 and NCL RNA samples were indexed separately and then merged before enrichment; therefore, the substantial enrichment of RBP-mRNA interactions in the "Pulldown_x400" sample compared to the "Pulldown_NCL" sample is a direct measure of relative mRNA recovery under each condition. The interaction between RBP and RPL30 is enriched in the 5' UTR, particularly at the 5' TOP motif that controls ribosomal protein translation, and additionally throughout the CDS. The interaction between RBP and GAPDH mRNA is most prominent in exons 3 and 5 of the CDS. The interaction between RBP and ACTB is most enriched in the 3' UTR. Importantly, RBP-mRNA interactions are observed in the 5' UTR, CDS, and 3' UTR for all three transcripts. These scaled trajectories suggest that certain interactions are more enriched relative to others within certain regions of the mRNA. See also Figure 15ATo view the trajectory details scaled up, and to see the RBP-mRNA interaction within the 3'UTR of GAPDH mRNA.

[0039] Figures 13A to 13F Analysis of cross-linking patterns within the sequence motif identified the specific RBP that binds at the RBP-mRNA interaction site. Figures 13A to 13B ARORA crosslinking site and eCLIP on GAPDH exon 5 ( Figure 13A ) and TOMM6's 3'UTR ( Figure 13B A comparison of the annotations for the crosslinking sites of YBX3 and PUM2 in the diagram. The gray boxes indicate the common sequence motif for each RBP. Figures 13C to 13D YBX3 eCLIP ( Figure 13C ) and PUM2 eCLIP ( Figure 13D The mCross results show the specific nucleotide positions of crosslinks in the sequence. Similar to fingerprints, YBX3 is typically crosslinked at nucleotide 2 of the GUCANC sequence motif, while PUM2 is crosslinked more at position 0 of the PUM2 motif UGUANANA than at position 1. Figures 13E to 13F : Detected by ARORA, containing YBX3 ( Figure 13E ) or PUM2 ( Figure 13F A heatmap of crosslinking patterns within the first 200 sites of RBP-RNA interaction within the sequence motif of a common RBP, with the scale bar indicating the fraction of all crosslinks at each nucleotide within an 11-nucleotide window. A subset of these RBP interaction sites contains crosslinking patterns consistent with YBX3 or PUM2. The presence of a crosslinking pattern consistent with the typical crosslinking pattern of an RBP is associated with approximately a 75% probability of an eCLIP peak annotated at that precise site, while approximately 25% of sites without a consistent crosslinking pattern have an eCLIP peak, but this is often in the context of many other RBPs having peaks at that site. Therefore, the presence of a crosslinking pattern consistent with a specific RBP within its shared sequence motif strongly predicts whether the RBP will bind at that specific site (P < 10). -5 Crucially, the consistent crosslinking pattern of RBP was observed only under Pulldown_x400 conditions, and not under negative control or input conditions.

[0040] Figures 14A to 14D ARORA identified the relative enrichment of specific RBPs on mRNA and identified proteins that act as key regulators of transcripts. Figure 14AThe interaction between YBX3-congruent RBPs and GAPDH mRNA is the most enriched RBP-mRNA interaction to date, with an enrichment >10X compared to other RBPs. The advantage of YBX3-congruent RBP interactions is their common feature among the mRNAs of most proteins involved in glycolysis and subsets of proteins involved in translation. This is consistent with recent reports that YBX proteins are essential for regulating the mRNA stability and translation of some components of glycolytic proteins and the translation apparatus. Figure 14B The cell cycle regulator cell cycle protein B2 (CCNB2) mRNA has a single dominant RBP mRNA interaction in the 3'UTR of the mRNA, and was identified as PUM2 by analysis of the cross-linking pattern in the sequence motif, which is consistent with eCLIP. Figure 14C Compared to the eCLIP results for each RBP, the interactions at the two YBX3 congruent sites and the two PUM2 congruent sites on RPL14 mRNA showed similar relative stoichiometry. Figure 14D However, unlike eCLIP, ARORA can directly compare the relative proportions of interactions between two RBPs on the same transcript.

[0041] Figures 15A to 15C Differences between cell lines: RBP-mRNA interactions occur at sites of single nucleotide variants. Figure 15A RBP-mRNA interaction sites in different GAPDH 3'UTRs between HepG2 and K562 cells (gray boxes). Figure 15B The RBP-mRNA interaction in the 3'UTR of the proto-oncogene CTNNB1 is specific to HepG2 cells and is associated with single nucleotide variants at this site (*). Figure 15C The unique RBP interaction with the 5'UTR of MYC in HepG2 cells, relative to K562 cells, is also associated with single nucleotide variants (SNPs) in the 5'UTR of this cell line (*). The SNP alleles in the HepG2 5'UTR are highly enriched relative to the wild-type alleles, suggesting that this allele may acquire RBP interactions in the MYC 5'UTR.

[0042] Figure 16A PUM2 crosslinking pattern identified on lncRNA NORAD by ARORA. Figure 16B YBX3 crosslinking mode identified on GAPDH by ARORA. Figure 16C TARDBP crosslinking mode identified on SMAP2 by ARORA. Detailed Implementation

[0043] It should be understood that certain aspects, methods, embodiments, variations and features of the methods of the present invention are described below at different levels of detail in order to provide a substantial understanding of the technology of the present invention.

[0044] In practicing the methods of this invention, many conventional techniques from molecular biology, protein biochemistry, cell biology, immunology, microbiology, and recombinant DNA were used. See, for example, Sambrook and Russell, eds. (2001) *Molecular Cloning: A Laboratory Manual*, 3rd edition; Ausubel et al., eds., series (2007) *Current Protocols in Molecular Biology*; *Methods in Enzymology* series (AcademicPress, Inc., NY).MacPherson et al. (1991) PCR 1: A Practical Approach (IRL Press at Oxford University Press); MacPherson et al. (1995) PCR 2: A Practical Approach; Harlow and Lane (eds.) (1999) Antibodies, A Laboratory Manual; Freshney (2005) Culture of Animal Cells: A Manual of Basic Technique, 5th Edition; Gait (ed.) (1984) Oligonucleotide Synthesis; US Patent No. 4,683,195; Hames and Higgins (eds.) (1984) Nucleic Acid Hybridization; Anderson (1999) Nucleic Acid Hybridization; Hames and Higgins (eds.) (1984) Transcription and Translation; Immobilized Cells and Enzymes (IRL Press (1986)); Perbal (1984) A Practical Guide to Molecular Cloning; Miller and Calos (eds., 1987) Gene Transfer Vectors for Mammalian Cells (ColdSpring Harbor Laboratory); Makrides (ed., 2003) Gene Transfer and Expression in Mammalian Cells; Mayer and Walker (ed., 1987) Immunochemical Methods in Cell and Molecular Biology (Academic Press, London); and Herzenberg et al. (eds., 1996) Weir's Handbook of Experimental Immunology. Methods for detecting and measuring peptide gene expression product levels (i.e., gene translation levels) are well known in the art and include peptide detection methods using techniques such as antibody detection and quantitative techniques. (See also Strachan and Read, Human Molecular Genetics, 2nd Edition.)(John Wiley and Sons, Inc., NY, 1999)). .

[0045] This article discloses a method (aka, RNA-bound protein-occupied annotation (ARORA)) for simultaneously identifying hundreds of RBPs at single nucleotide resolution throughout the transcriptome for RBP-RNA interaction (PRI) sites.

[0046] ARORA requires several key technological advancements to achieve scalable and sensitive detection of PRI with high signal-to-noise ratios (even at moderate read depths), including: (1) efficient and unbiased labeling of RNA by maintaining RNA in an unstructured state at high DMSO concentrations during covalent labeling reactions; (2) reducing the molecular complexity of RBP-RNA crosslinked biological samples using extensive proteolysis while preserving PRI information, making RNA processing identical to standard uncrosslinked RNA and compatible with high-throughput processing and automation; and (3) utilizing heat and divalent cations for highly reliable RNA processing. Reproducible, unbiased RNA fragmentation to achieve consistent short fragments is optimal for effective annotation of PRI. (4) Enrichment of cross-linked RNA fragments by streptavidin affinity purification and cDNA synthesis while RNA templates are immobilized on streptavidin beads significantly improves the efficiency of PRI identification. (5) Attaching uniquely indexed 3' adapters to RNA fragments before merging RNA from UVC cross-linked and non-cross-linked samples into a single sample before pull-down and cDNA synthesis allows for precise, quantitative comparison of enrichment between UVC cross-linked and non-cross-linked control samples.

[0047] This technique reproduces PRI profiles obtained using three conventional orthogonal methods (eCLIP (enhanced crosslinking immunoprecipitation), RNP-MaP (ribonucleoprotein network analysis via mutation profiling), and SHAPE footprinting analysis) with orders of magnitude higher efficiency in a single assay. ARORA is also capable of identifying PRI interactions in dynamic and stress responses and distinguishing biological states that may be associated with different drug sensitivities. ARORA also identifies allele-specific sequence variations in regulatory RNAs that affect RBP molecule interactions (e.g., variations in the MYC 5'UTR that promote MYC protein translation and cancer cell growth). The examples described herein further demonstrate the successful implementation of this technique in frozen tissue samples. Therefore, the methods disclosed herein can be used as platform technologies for diagnostic and drug target discovery. Unlike conventional methods for identifying PRI, ARORA is also suitable for automated, high-throughput, and unbiased discovery.

[0048] definition

[0049] Unless otherwise defined, all technical and scientific terms used herein generally have the same meaning as commonly understood by one of ordinary skill in the art to which this art pertains. As used in this specification and the appended claims, unless otherwise expressly indicated, the singular forms “a / an” and “the” include plural referents. For example, reference to “cell” includes a combination of two or more types of cells, etc. Generally, the nomenclature used herein and the laboratory procedures in cell culture, molecular genetics, organic chemistry, analytical chemistry, and nucleic acid chemistry and hybridization described below are those well-known and commonly used in the art.

[0050] Generally speaking, mentioning an element (such as hydrogen or H) means including all isotopes of that element. For example, if the R group is defined as including hydrogen or H, it also includes deuterium and tritium. Therefore, it includes radioactive isotopes (such as tritium, carbon, and nitrogen). 14 P 32 and S 35 Compounds of this technology are within the scope of this technology. Based on the disclosure herein, the procedure for inserting such markings into compounds of this technology will be apparent to those skilled in the art.

[0051] Generally, "substituted" means an organic group (e.g., an alkyl group) as defined below, wherein one or more bonds to a hydrogen atom contained therein are replaced by bonds to non-hydrogen or non-carbon atoms. A substituted group also includes a group in which one or more bonds to a carbon or hydrogen atom are replaced by one or more bonds to a heteroatom (including double or triple bonds). Therefore, unless otherwise specified, a substituted group is substituted by one or more substituents. In some embodiments, the substituted group is substituted by one, two, three, four, five, or six substituents. Examples of substituent groups include: halogens (i.e., F, Cl, Br, and I); hydroxyl groups; alkoxy, alkenoxy, aryloxy, arylalkoxy, heterocyclic, heterocyclic alkyl, heterocyclic oxy, and heterocyclic alkoxy groups; carbonyl (oxo); carboxylic acid esters; esters; carbamates; oximes; hydroxylamines; alkoxyamines; arylalkoxyamines; thiols; thioethers; sulfoxides; sulfones; sulfonyl groups; pentafluorothiols (i.e., SF5); sulfonamides; amines; N-oxides; hydrazines; acyl hydrazides; hydrazones; azides; amides; ureas; amidines; guanidines; enamines; imides; isocyanates; isothiocyanates; cyanates; thiocyanates; imides; nitro groups; and nitriles (i.e., CN).

[0052] Substituted cyclic groups (such as substituted cycloalkyl, aryl, heterocyclic, and heteroaryl groups) also include rings and ring systems in which the bonds to hydrogen atoms are replaced by bonds to carbon atoms. Therefore, substituted cycloalkyl, aryl, heterocyclic, and heteroaryl groups can also be substituted with substituted or unsubstituted alkyl, alkenyl, and alkynyl groups as defined below.

[0053] Alkyl groups include straight-chain and branched alkyl groups having 1 to 12 carbon atoms, typically 1 to 10 carbon atoms, or in some embodiments 1 to 8, 1 to 6, or 1 to 4 carbon atoms. Alkyl groups may be substituted or unsubstituted. Examples of straight-chain alkyl groups include groups such as methyl, ethyl, n-propyl, n-butyl, n-pentyl, n-hexyl, n-heptyl, and n-octyl groups. Examples of branched alkyl groups include, but are not limited to, isopropyl, isobutyl, sec-butyl, tert-butyl, neopentyl, isopentyl, and 2,2-dimethylpropyl groups. Representative substituted alkyl groups may be substituted once or multiple times with substituents such as those listed above, and include, but are not limited to, haloalkyl (e.g., trifluoromethyl), hydroxyalkyl, thioalkyl, aminoalkyl, alkylaminoalkyl, dialkylaminoalkyl, alkoxyalkyl, carboxylalkyl, etc.

[0054] Cycloalkyl groups comprise monocycloalkyl, bicycloalkyl, or tricycloalkyl groups having 3 to 12 carbon atoms in the ring (or, in some embodiments, 3 to 10, 3 to 8, or 3 to 4, 5, or 6 carbon atoms). Cycloalkyl groups may be substituted or unsubstituted. Exemplary monocycloalkyl groups include, but are not limited to, cyclopropyl, cyclobutyl, cyclopentyl, cyclohexyl, cycloheptyl, and cyclooctyl groups. In some embodiments, the cycloalkyl group has 3 to 8 ring members, while in other embodiments, the number of ring carbon atoms ranges from 3 to 5, 3 to 6, or 3 to 7. Bicycloalkyl and tricycloalkyl systems comprise both bridged cycloalkyl groups and fused rings, such as, but not limited to, bicyclo[2.1.1]hexane, adamantyl, decahydronaphthyl, etc. Substituted cycloalkyl groups may be substituted once or multiple times with non-hydrogen and non-carbon groups as defined above. However, substituted cycloalkyl groups also comprise rings substituted with straight-chain or branched alkyl groups as defined above. Representative substituted cycloalkyl groups can be monosubstituted or substituted more than once, such as, but not limited to, 2,2-, 2,3-, 2,4-, 2,5- or 2,6-disubstituted cyclohexyl groups, which can be substituted by substituents such as those listed above.

[0055] A cycloalkyl group is an alkyl group as defined above, wherein the hydrogen or carbon bond of the alkyl group is replaced by a bond to the cycloalkyl group as defined above. The cycloalkyl group may be substituted or unsubstituted. In some embodiments, the cycloalkyl group has 4 to 16 carbon atoms, 4 to 12 carbon atoms, and typically 4 to 10 carbon atoms. A substituted cycloalkyl group may be substituted at the alkyl, cycloalkyl, or both alkyl and cycloalkyl moieties of the group. Representative substituted cycloalkyl groups may be monosubstituted or substituted more than once, such as, but not limited to, monosubstituted, disubstituted, or trisubstituted by substituents such as those listed above.

[0056] In addition to at least one double bond between two carbon atoms, the alkenyl group comprises straight-chain and branched alkyl groups as defined above. The alkenyl group may be substituted or unsubstituted. The alkenyl group has 2 to 12 carbon atoms, typically 2 to 10 carbon atoms, or in some embodiments 2 to 8, 2 to 6, or 2 to 4 carbon atoms. In some embodiments, the alkenyl group has one, two, or three carbon-carbon double bonds. Examples include, but are not limited to, vinyl, allyl, -CH=CH(CH3), -CH=C(CH3)2, -C(CH3)=CH2, -C(CH3)=CH(CH3), -C(CH2CH3)=CH2, etc. Representative substituted alkenyl groups may be monosubstituted or substituted more than once, such as, but not limited to, monosubstituted, disubstituted, or trisubstituted groups (such as those listed above).

[0057] A cycloalkenyl group includes a cycloalkyl group as defined above having at least one double bond between two carbon atoms. The cycloalkenyl group may be substituted or unsubstituted. In some embodiments, the cycloalkenyl group may have one, two, or three double bonds, but does not include aromatic compounds. The cycloalkenyl group has 4 to 14 carbon atoms, or in some embodiments, 5 to 14 carbon atoms, 5 to 10 carbon atoms, or even 5, 6, 7, or 8 carbon atoms. Examples of cycloalkenyl groups include cyclohexenyl, cyclopentenyl, cyclohexadienyl, cyclobutadienyl, and cyclopentadienyl.

[0058] A cycloalkenylalkyl group is an alkyl group as defined above, wherein the hydrogen or carbon bond of the alkyl group is replaced by a bond to a cycloalkenyl group as defined above. The cycloalkenylalkyl group can be substituted or unsubstituted. A substituted cycloalkenylalkyl group can be substituted at the alkyl moiety, at the cycloalkenyl moiety, or at both the alkyl and cycloalkenyl moiety. Representative substituted cycloalkenylalkyl groups can be substituted once or multiple times with substituents (such as those listed above).

[0059] The alkynyl group includes straight-chain and branched alkyl groups as defined above, differing in that there is at least one triple bond between two carbon atoms. The alkynyl group can be substituted or unsubstituted. The alkynyl group has 2 to 12 carbon atoms, typically 2 to 10 carbon atoms, or in some embodiments 2 to 8, 2 to 6, or 2 to 4 carbon atoms. In some embodiments, the alkynyl group has one, two, or three carbon-carbon triple bonds. Examples include, but are not limited to, –C≡CH, -C≡CCH3, -CH2C≡CCH3, and -C≡CCH2CH(CH2CH3)2. Representative substituted alkynyl groups can be monosubstituted or substituted more than once, such as, but not limited to, monosubstituted, disubstituted, or trisubstituted groups (such as those listed above).

[0060] An aryl group is a cyclic aromatic hydrocarbon that does not contain heteroatoms. The aryl groups described herein include monocyclic, bicyclic, and tricyclic ring systems. Aryl groups can be substituted or unsubstituted. Therefore, aryl groups include, but are not limited to, phenyl, azulel, heptenyl, biphenyl, fluorenyl, phenanthryl, anthracel, indene, indanyl, cyclopentadienyl, and naphthyl groups. In some embodiments, the aryl group contains 6 to 14 carbons in the ring portion of the group, and in other embodiments, it contains 6 to 12 or even 6 to 10 carbon atoms. In some embodiments, the aryl group is phenyl or naphthyl. The phrase "aryl group" includes groups containing fused rings, such as fused aromatic-aliphatic ring systems (e.g., indene, tetrahydronaphthyl, etc.). Representative substituted aryl groups can be monosubstituted (e.g., tolyl) or substituted more than once. For example, monosubstituted aryl groups include, but are not limited to, phenyl or naphthyl groups substituted with 2, 3, 4, 5 or 6, which can be substituted by substituents such as those listed above.

[0061] An aralkyl group is an alkyl group as defined above, wherein the hydrogen or carbon bond of the alkyl group is replaced by a bond to an aryl group as defined above. The aralkyl group may be substituted or unsubstituted. In some embodiments, the aralkyl group contains 7 to 16 carbon atoms, 7 to 14 carbon atoms, or 7 to 10 carbon atoms. A substituted aralkyl group may be substituted at the alkyl, aryl, or both alkyl and aryl moieties of the group. Representative aralkyl groups include, but are not limited to, benzyl and phenethyl groups, as well as fused (cycloalkylaryl) alkyl groups such as 4-indenylethyl. A representative substituted aralkyl group may be substituted once or multiple times with substituents such as those listed above.

[0062] Heterocyclic groups include aromatic (also called heteroaryl) and non-aromatic cyclic compounds containing three or more ring members, one or more of which are heteroatoms, such as, but not limited to, N, O, and S. Heterocyclic groups may be substituted or unsubstituted. In some embodiments, a heterocyclic group contains one, two, three, or four heteroatoms. In some embodiments, a heterocyclic group includes monocyclic, bicyclic, and tricyclic groups having 3 to 16 ring members, while other such groups have 3 to 6, 3 to 10, 3 to 12, or 3 to 14 ring members. Heterocyclic groups include aromatic, partially unsaturated, and saturated ring systems, such as, for example, imidazolyl, imidazolinyl, and imidazoalkyl groups. The phrase “heterocyclic group” includes fused ring species, including those containing fused aromatic and non-aromatic groups, such as, for example, benzotriazolyl, 2,3-dihydrobenzo[1,4]dioxadienyl, and benzo[1,3]dioxadienyl. This phrase also includes bridged polycyclic ring systems containing heteroatoms, such as, but not limited to, quinine cyclic groups. This phrase includes heterocyclic groups having other groups (such as alkyl, oxo, or halogenated groups) bonded to one of the ring members, referred to as "substituted heterocyclic groups." Heterocyclic groups include, but are not limited to, acridinel, acridinel, pyrrolidine, imidazolyl, pyrazolidine, thiazolyl, tetrahydrothiophene, tetrahydrofuranyl, dioxacyclopentenyl, furanyl, thiophene, pyrrolyl, pyrrololinyl, imidazolyl, imidazolinyl, pyrazolyl, pyrazolyl, triazolyl, tetraazolyl, oxazolyl, isoxazolyl, thiazolyl, thiazolinyl, isothiazolyl, thiazolinyl, isothiazolyl, thiadiazolyl, oxadiazolyl, piperidinyl, piperazinel, etc. Morpholinyl, thiomorpholinyl, tetrahydropyranyl, tetrahydrothiaranyl, oxothiacyclohexane, dioxacyclohexyl, dithiaranyl, pyranyl, pyridyl, pyrimidinyl, pyrazinyl, triazinyl, dihydropyridyl, dihydrodithiaranyl, dihydrodithionyl, peripetaazinyl, quininecycloyl, indoleyl, indololinyl, isoindoleyl, azaindoleyl (pyrrolopyridyl), indazoleyl, inazinyl, benzotriazolyl, benzimidazoleyl, benzene benzo[1,3]dioxadiazolyl, benzo[1,3]pyrazolyl, imidazopyridyl (azabenzimidazolyl), triazolylpyridyl, isoxazolylpyridyl, purine, xanthine, adenine, guanine, quinolinyl Isoquinolinyl, quinazinyl, quinoxalinyl, quinazolinyl, cyclophosphinyl, phthalazinyl, naphridinyl, pteridinyl, thianaphthyl, dihydrobenzothiazinyl, dihydrobenzofuranyl, dihydroindolyl, dihydrobenzodioxanedienyl, tetrahydroindolyl, tetrahydroindazoleyl, tetrahydrobenzimidazolyl, tetrahydrobenzotriazolyl, tetrahydropyrrolopyridyl, tetrahydropyrazolopyridyl, tetrahydroimidazopyridyl, tetrahydrotriazolopyridyl, and tetrahydroquinolinyl groups.Representative substituted heterocyclic groups can be monosubstituted or substituted more than once, such as, but not limited to, pyridyl or morpholinyl groups substituted or disubstituted at the 2-, 3-, 4-, 5-, or 6- positions by various substituents (such as those listed above).

[0063] A heteroaryl group is an aromatic ring compound containing five or more ring members, one or more of which are heteroatoms, such as, but not limited to, N, O, and S. The heteroaryl group can be substituted or unsubstituted. Heteroaryl groups include, but are not limited to, groups such as pyrroloyl, pyrazolyl, triazolyl, tetrazolyl, oxazolyl, isoxazolyl, thiazolyl, pyridinyl, pyridinyl, pyrazinyl, thiophenyl, benzothiophene, furanyl, benzofuranyl, indolyl, azaindolyl (pyrrolopyridinyl), inzolyl, benzimidazolyl, imidazopyridinyl (azabenzimidazolyl), pyrazolopyridinyl, triazolopyridinyl, benzotriazolyl, benzooxazolyl, benzothiazolyl, benzothiadiazolyl, imidazopyridinyl, isoxazolopyridinyl, thianaphthyl, purine, xanthine, adenine, guanine, quinolinyl, isoquinolinyl, tetrahydroquinolinyl, quinoxolinyl, and quinazolinyl groups. A heteroaryl group comprises fused-ring compounds in which all rings are aromatic, such as an indole group, and fused-ring compounds in which only one ring is aromatic, such as a 2,3-dihydroindole group. Representative substituted heteroaryl groups can be substituted once or multiple times with various substituents, such as those listed above.

[0064] A heterocyclic alkyl group is an alkyl group as defined above, wherein the hydrogen or carbon bond of the alkyl group is replaced by a bond to the heterocyclic group as defined above. The heterocyclic alkyl group can be substituted or unsubstituted. A substituted heterocyclic alkyl group can be substituted at the alkyl moiety, at the heterocyclic moiety, or at both the alkyl and heterocyclic moieties. Representative heterocyclic alkyl groups include, but are not limited to, morpholino-4-yl-ethyl, furan-2-yl-methyl, imidazol-4-yl-methyl, pyridin-3-yl-methyl, tetrahydrofuran-2-yl-ethyl, and indol-2-yl-propyl. A representative substituted heterocyclic alkyl group can be substituted once or multiple times with substituents (such as those listed above).

[0065] A heteroarylalkyl group is an alkyl group as defined above, wherein the hydrogen or carbon bond of the alkyl group is replaced by a bond to a heteroaryl group as defined above. The heteroarylalkyl group can be substituted or unsubstituted. A substituted heteroarylalkyl group can be substituted at the alkyl moiety, at the heteroaryl moiety, or at both the alkyl and heteroaryl moieties. Representative substituted heteroarylalkyl groups can be substituted once or multiple times by substituents (such as those listed above).

[0066] Groups having two or more connection points (i.e., divalent, trivalent, or polyvalent) in compounds described herein are named using the suffix "ene." For example, a divalent alkyl group is an alkylene group, a divalent aryl group is an arylene group, a divalent heteroaryl group is a divalent heteroarylene group, and so on. Substituted groups having a single connection point in compounds of this technology are not referred to using the name "ene." Thus, for example, chloroethyl is not referred to as vinyl chloride herein.

[0067] An alkoxy group is a hydroxyl group (-OH) in which the bond to a hydrogen atom is replaced by a bond to a carbon atom of a substituted or unsubstituted alkyl group as defined above. Alkoxy groups can be substituted or unsubstituted. Examples of straight-chain alkoxy groups include, but are not limited to, methoxy, ethoxy, propoxy, butoxy, pentoxy, and hexoxy. Examples of branched-chain alkoxy groups include, but are not limited to, isopropoxy, sec-butoxy, tert-butoxy, isopentoxy, and isohexoxy. Examples of cycloalkoxy groups include, but are not limited to, cyclopropoxy, cyclobutoxy, cyclopentoxy, and cyclohexoxy. Representative substituted alkoxy groups can be substituted once or multiple times by substituents such as those listed above.

[0068] As used herein, the terms “alkanoyl” and “alkanoyloxy” can refer to –C(O)–alkyl groups and –O–C(O)–alkyl groups, respectively. Similarly, “aromaticyl” and “aromaticoxy” refer to –C(O)-aryl groups and –OC(O)-aryl groups, respectively.

[0069] The terms "aryloxy" and "arylalkoxy" refer to a substituted or unsubstituted aryl group bonded to an oxygen atom and a substituted or unsubstituted aralkyl group bonded to an oxygen atom at an alkyl group, respectively. Examples include, but are not limited to, phenoxy, naphthoxy, and benzyloxy. Representative substituted aryloxy and arylalkoxy groups may be substituted once or multiple times by substituents such as those listed above.

[0070] As used in this article, the term "carboxylate group" refers to the -COOH group.

[0071] As used in this article, the term "ester" refers to –COOR 70 And –C(O)OG group. R 70It is a substituted or unsubstituted alkyl, cycloalkyl, alkenyl, alkynyl, aryl, aralkyl, heterocyclic alkyl, or heterocyclic group as defined herein. G is a carboxylic acid ester protecting group. Carboxylic acid ester protecting groups are well known to those skilled in the art. A comprehensive list of protecting groups for the functional groups of carboxylic acid ester groups can be found in Protective Groups in Organic Synthesis, Greene, TW; Wuts, PGM, John Wiley & Sons, New York, NY, (3rd edition, 1999), which may be added or removed using the procedures described herein, and is incorporated herein by reference in its entirety and for any and all purposes, as fully set forth herein.

[0072] The term "amide" (or "amide group") includes C- and N-amide groups, namely -C(O)NR, respectively. 71 R 72 and –NR 71 C(O)R 72 Group. R 71 and R 72 Independently, it is hydrogen, or a substituted or unsubstituted alkyl, alkenyl, alkynyl, cycloalkyl, aryl, aralkyl, heterocyclic alkyl, or heterocyclic group as defined herein. Therefore, the amide group includes, but is not limited to, carbamoyl groups (-C(O)NH2) and formamide groups (-NHC(O)H). In some embodiments, the amide is –NR. 71 C(O)-(C 1-5 Alkyl), and the group is referred to as "carbonylamino", and in other embodiments, the amide is -NHC(O)-alkyl, and the group is referred to as "alkanoylamino".

[0073] As used herein, the terms “nitrile” or “cyano” refer to the -CN group.

[0074] Carbamate groups include N- and O-carbamate groups, namely -NR and -NR respectively. 73 C(O)OR 74 and -OC(O)NR 73 R 74 Group. R 73 and R 74 Independently, it refers to substituted or unsubstituted alkyl, alkenyl, ynyl, cycloalkyl, aryl, aralkyl, heterocyclic alkyl, or heterocyclic groups as defined herein. 73 It can also be H.

[0075] As used in this article, the term "amine" (or "amino group") refers to –NR 75 R 76Group, wherein R 75 and R 76 Independently, it is hydrogen, or a substituted or unsubstituted alkyl, alkenyl, alkynyl, cycloalkyl, aryl, aralkyl, heterocyclic alkyl, or heterocyclic group as defined herein. In some embodiments, the amine is an alkylamino, dialkylamino, arylamino, or alkylarylamino. In other embodiments, the amine is NH2, methylamino, dimethylamino, ethylamino, diethylamino, propylamino, isopropylamino, phenylamino, or benzylamino.

[0076] The term "sulfonamide group" includes S- and N-sulfonamide groups, namely -SO2NR and -SO2NR, respectively. 78 R 79 and –NR 78 SO2R 79 Group. R 78 and R 79 Independently, it is hydrogen, or a substituted or unsubstituted alkyl, alkenyl, alkynyl, cycloalkyl, aryl, aralkyl, heterocyclic alkyl, or heterocyclic group as defined herein. Therefore, the sulfonamide group includes, but is not limited to, an aminosulfonyl group (-SO2NH2). In some embodiments herein, the sulfonamide group is –NHSO2-alkyl and is referred to as an “alkylsulfonylamino” group.

[0077] The term "thiol" refers to the -SH group, while "thioether" includes the -SR group. 80 Groups, "sulfoxide" include –S(O)R 81 Groups, sulfones include -SO2R 82 Groups, including sulfonyl groups, include –SO2OR 83 And “sulfonates” include –SO3 − R 80 R 81 R 82 and R 83 Each is independently a substituted or unsubstituted alkyl, cycloalkyl, alkenyl, alkynyl, arylalkyl, heterocyclic, or heterocyclic alkyl group as defined herein. In some embodiments, the thioether is an alkylthio group or an -S-alkyl group.

[0078] The term "urea" refers to –NR 84 -C(O)-NR 85 R 86 Group. R 84 R 85 and R 86 The group is independently hydrogen, or a substituted or unsubstituted alkyl, alkenyl, alkynyl, cycloalkyl, aryl, aralkyl, heterocyclic or heterocyclic alkyl group as defined herein.

[0079] The term "ammonia" refers to –C(NR)87 )NR 88 R 89 and –NR 87 C(NR 88 )R 89 , where R 87 R 88 and R 89 Each is independently hydrogen, or a substituted or unsubstituted alkyl, cycloalkyl, alkenyl, alkynyl, arylaralkyl, heterocyclic or heterocyclic alkyl group as defined herein.

[0080] The term "guanidine" refers to –NR 90 C(NR 91 )NR 92 R 93 , where R 90 R 91 R 92 and R 93 Each is independently hydrogen, or a substituted or unsubstituted alkyl, cycloalkyl, alkenyl, alkynyl, arylaralkyl, heterocyclic or heterocyclic alkyl group as defined herein.

[0081] The term "enamine" refers to –C(R) 94 )=C(R 95 )NR 96 R 97 and –NR 94 C(R 95 )=C(R 96 )R 97 , where R 94 R 95 R 96 and R 97 Each is hydrogen, as defined herein, substituted or unsubstituted alkyl, cycloalkyl, alkenyl, alkynyl, arylaralkyl, heterocyclic or heterocyclic alkyl groups.

[0082] As used herein, the term "halogen" or "halogenated" refers to bromine, chlorine, fluorine, or iodine. In some embodiments, the halogen is fluorine. In other embodiments, the halogen is chlorine or bromine.

[0083] As used herein, the term "hydroxyl group" can refer to –OH or its ionized form –O. – A "hydroxyalkyl" group is an alkyl group that has been substituted with a hydroxyl group, such as HO-CH2-.

[0084] The term "imide" refers to –C(O)NR 98 C(O)R 99 , where R 98 and R 99Each is, independently, hydrogen, or a substituted or unsubstituted alkyl, cycloalkyl, alkenyl, alkynyl, arylaralkyl, heterocyclic or heterocyclic alkyl group as defined herein.

[0085] The term "imine" refers to –CR 100 (NR 101 ) and –N(CR 100 R 101 ) group, wherein R 100 and R 101 Each is independently hydrogen or a substituted or unsubstituted alkyl, cycloalkyl, alkenyl, ynyl, arylaralkyl, heterocyclic, or heterocyclic alkyl group as defined herein, provided that R 100 and R 101 They are not both hydrogen at the same time.

[0086] As used in this article, the term "nitro" refers to the –NO2 group.

[0087] As used in this article, the term "trifluoromethyl" refers to –CF3.

[0088] As used in this article, the term "trifluoromethoxy" refers to –OCF3.

[0089] The term "azido group" refers to –N3.

[0090] The term "trialkylammonium" refers to the –N(alkyl)3 group. Trialkylammonium groups are positively charged and therefore typically have associated anions, such as halide anions.

[0091] The term "isocyanate" refers to –NC.

[0092] The term "isothiocyanate" refers to –NCS.

[0093] The term "pentafluorosulfuryl" refers to –SF5.

[0094] As will be understood by those skilled in the art, “molecular weight” (also known as “relative molar mass”) is a dimensionless quantity, but can be converted to molar mass by multiplying by 1 g / mol or by 1 Da. For example, a compound with a weight-average molecular weight of 5,000 has a weight-average molar mass of 5,000 g / mol and a weight-average molar mass of 5,000 Da.

[0095] As used herein, the term “about” when referring to a number means a number that is generally included in the range of 1%, 5%, or 10% in either direction of the number (greater or less), unless otherwise stated or otherwise obvious from the context (unless such a number is less than 0% or more than 100% of the possible value).

[0096] The term "adapter" refers to a short, chemically synthesized nucleic acid that can be attached to the 3' or 5' end of a nucleic acid sequence to facilitate attachment to another molecule. Adapters can be single-stranded or double-stranded. Adapters can be incorporated into short (e.g., less than 55 base pairs) sequences used for PCR amplification or sequencing. Adapters can contain known sequences, degenerate sequences (sequences without precise definitions), or both. Double-stranded adaptors can contain two hybridizable strands. Alternatively, double-stranded adaptors can contain a hybridizable portion and a non-hybridizable portion. The non-hybridizable portion of a double-stranded adaptor contains two single-stranded regions that cannot hybridize with each other. Within the non-hybridizable portion, the strand containing the unhybridized 5' end is called the 5'-strand, and the strand containing the unhybridized 3' end is called the 3'-strand. In some embodiments, a double-stranded adaptor has a hybridizable portion at one end and a non-hybridizable portion at the opposite end. In some embodiments, the non-hybridizable portion of a double-stranded adaptor can be open (Y-shaped adaptor). In some embodiments, the adaptor can be a U-shaped adaptor.

[0097] The term "amino acid" refers to naturally occurring and non-naturally occurring amino acids, as well as amino acid analogs and amino acid mimics that function in a manner similar to naturally occurring amino acids. Naturally encoded amino acids include 20 common amino acids (alanine, arginine, asparagine, aspartic acid, cysteine, glutamine, glutamic acid, glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, and valine), as well as pyrrolidone and selenocysteine. Amino acid analogs are reagents that have the same basic chemical structure as naturally occurring amino acids, namely, an α-carbon linked to a hydrogen, carboxyl, amino, and R group, such as homoserine, ortholeucine, methionine sulfoxide, and methionine methylsulfonium. Such analogs have modified R groups (such as ortholeucine) or modified peptide backbones, but retain the same basic chemical structure as naturally occurring amino acids. In some embodiments, the amino acid forming the polypeptide is in the D form. In some embodiments, the amino acid forming the polypeptide is in the L form. In some embodiments, the first plurality of amino acids forming the polypeptide are in the form of D, and the second plurality of amino acids are in the form of L.

[0098] Amino acids are represented in this paper using their commonly known three-letter symbols or single-letter symbols recommended by the IUPAC-IUB Biochemistry Nomenclature Committee. Similarly, nucleotides are represented using their commonly accepted single-letter codes.

[0099] As used herein, the term "amplify" or "amplification" in relation to nucleic acid sequences refers to a method for increasing the expression of a population of nucleic acid sequences in a sample. Nucleic acid amplification methods (such as PCR, isothermal methods, rolling circle methods, etc.) are well known to those skilled in the art. A copy of a specific nucleic acid sequence produced in vitro during an amplification reaction is called an "amplifier" or "amplification product".

[0100] As used herein, a "decoy" is a hybrid capture agent that recovers a target nucleic acid sequence for sequencing. The decoy can be a nucleic acid molecule, such as a DNA or RNA molecule, that can hybridize (e.g., complement) with the target nucleic acid, thereby allowing the capture of the target nucleic acid. In one embodiment, the decoy is an RNA molecule (e.g., a naturally occurring or modified RNA molecule); a DNA molecule (e.g., a naturally occurring or modified DNA molecule); or a combination thereof. In other embodiments, the decoy includes a binding entity, such as an affinity tag, which allows, for example, the capture and separation of a hybrid formed by the decoy and the nucleic acid hybridized with the decoy, by binding to the binding entity. In one embodiment, the decoy is suitable for solution-phase hybridization.

[0101] The term "barcode" refers to a nucleotide sequence within a polynucleotide used to identify a nucleic acid molecule. For example, a barcode can be used to identify molecules when molecules from several groups are combined in multiple ways for processing or sequencing. The barcode can be located at a position within the polynucleotide (e.g., at the 3' end, 5' end, or middle of the polynucleotide) and can contain sequences of any length (e.g., 1-100 or more nucleotides). Additionally, a barcode can contain one or more predefined sequences. The term "predefined" means that the sequence of the barcode is predetermined or known before identification or when identification of the entire sequence of the nucleic acid containing the barcode is not required. In some cases, a predefined barcode can be attached to a nucleic acid for sorting the nucleic acid into groups. In some embodiments, the barcode may contain an artificial sequence, such as a designed or engineered sequence not present in the subject's unaltered (wild-type) genome. In other embodiments, the barcode may contain an endogenous sequence, such as a sequence present in the subject's unaltered (wild-type) genome. In some embodiments, the barcode may be an endogenous barcode. Endogenous barcodes can be sequences of genomic nucleic acids, which serve as barcodes or identifiers for the genomic nucleic acids. One or more sequences of a genomic DNA fragment can be endogenous barcodes. Different types of barcodes can be used in combination. For example, an endogenous genomic nucleic acid fragment can be attached to an artificial sequence, which can be used as a unique identifier for the genomic nucleic acid fragment. "Sample-specific barcodes" or "patient barcodes" refer to polynucleotide sequences used to identify the origin or source of nucleic acid molecules. For example, the sequence "AAAA" can be attached to identify nucleic acids isolated from patient A.

[0102] As used herein, the term "biological sample" means sample material derived from living cells. Biological samples can include tissues, cells, cellular protein or membrane extracts, and biological fluids (e.g., ascites or cerebrospinal fluid (CSF)) isolated from a subject, as well as tissues, cells, and fluids present within the subject. Biological samples of the technology of this invention include, but are not limited to, samples taken from: breast tissue, kidney tissue, cervix, endometrium, head or neck, gallbladder, parotid gland tissue, prostate, brain, pituitary gland, kidney tissue, muscle, esophagus, stomach, small intestine, colon, liver, spleen, pancreas, thyroid tissue, heart tissue, lung tissue, bladder, adipose tissue, lymph node tissue, uterus, ovarian tissue, adrenal gland tissue, testicular tissue, tonsils, thymus, blood, hair, oral cavity, skin, serum, plasma, CSF, sperm, prostatic fluid, semen, urine, feces, sweat, saliva, sputum, mucus, bone marrow, lymph, and tears. Biological samples can also be obtained from biopsies of internal organs or cancer. Biological samples can be obtained from subjects for diagnosis or research, or from non-disease-prone individuals as controls or for basic research. Samples can be obtained using standard methods, including, for example, venipuncture and surgical biopsy. In some embodiments, biological samples are tissue samples obtained via venipuncture biopsy.

[0103] The terms “cancer” and “tumor” are used interchangeably and refer to cells exhibiting typical characteristics of cancerous cells, such as uncontrolled proliferation, immortality, metastatic potential, rapid growth and proliferation rates, and certain distinctive morphological features. Cancer cells typically take the form of tumors, but such cells can exist alone in an animal or can be non-tumorigenic cancer cells. As used herein, the term “cancer cell” includes precancerous (e.g., benign), malignant, pre-metastatic, metastatic, and non-metastatic cells. Cancers in virtually every tissue are known to those skilled in the art, including solid tumors such as carcinoma, sarcoma, glioblastoma, melanoma, lymphoma, myeloma, etc., and circulating cancers such as leukemia. Examples of cancers include, but are not limited to, ovarian cancer, breast cancer, colon cancer, lung cancer, prostate cancer, stomach cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, urinary tract cancer, thyroid cancer, kidney cancer, carcinoma, melanoma, head and neck cancer, and brain cancer. The term “cancer cell” refers to a cell that exhibits cancer-like characteristics (e.g., uncontrolled proliferation, resistance to growth signals, metastatic ability, and loss of the ability to undergo programmed cell death (e.g., apoptosis)) or a cell derived from a cancer cell, such as a clone of a cancer cell.

[0104] As used herein, the term "complementary" or "complementarity" in relation to polynucleotides (i.e., nucleotide sequences, such as oligonucleotides or target nucleic acids) refers to the base pairing rule. A complementary sequence of a nucleic acid sequence, as used herein, refers to an oligonucleotide that is in "antiparallel association" when aligned with a nucleic acid sequence such that the 5' end of one sequence pairs with the 3' end of another. For example, the sequence "5'-AGT-3'" is complementary to the sequence "3'-TCA-5". Certain bases not commonly found in naturally occurring nucleic acids can be included in the nucleic acids described herein. These include, for example, inosine, 7-denitroguanine, locked nucleic acids (LNA), and peptide nucleic acids (PNA). Complementarity does not have to be perfect; stable duplexes can contain mismatched base pairs, degenerate bases, or unmatched bases. Those skilled in the art of nucleic acid technology can empirically determine duplex stability by considering several variables, including, for example, the length of the oligonucleotide, the base composition and sequence of the oligonucleotide, ionic strength, and the incidence of mismatched base pairs. The complementary sequence can also be an RNA sequence that is complementary to a DNA sequence or its complementary sequence, and it can also be cDNA.

[0105] As used herein, the term "association" refers to the association of two molecules by any method known to those skilled in the art. Suitable types of association include chemical bonds and physical bonds. Chemical bonds include, for example, covalent bonds and coordinate bonds. Physical bonds include, for example, hydrogen bonds, dipole interactions, van der Waals forces, electrostatic interactions, hydrophobic interactions, and aromatic stacking.

[0106] As used herein, a “control” is a substitute sample used in an experiment for comparative purposes. A control can be “positive” or “negative.” As used herein, a “control nucleic acid sample” or “reference nucleic acid sample” refers to a nucleic acid molecule derived from a control or reference sample. In some embodiments, the reference or control nucleic acid sample is a wild-type or non-mutated DNA or RNA sequence. In some embodiments, the reference nucleic acid sample is purified or isolated (e.g., removed from its native state).

[0107] As used in this article, "detection" refers to determining the presence of RBP-RNA interaction (PRI) sites in the nucleic acids of interest in a sample. Detection does not require the method to provide 100% sensitivity.

[0108] As used herein, a "detectable marker" refers to a molecule or compound, or a group of molecules or compounds, used to identify nucleic acids or proteins of interest. In some embodiments, the detectable marker can be detected directly. In other embodiments, the detectable marker may be part of a binding pair, which can then be detected subsequently. Signals from a detectable marker can be detected by a variety of means and will depend on the nature of the detectable marker. Detectable markers may be isotopes, fluorescent portions, colored substances, etc. Examples of means for detecting detectable markers include, but are not limited to, spectroscopic, photochemical, biochemical, immunochemical, electromagnetic, radiochemical, or chemical means such as fluorescence, chemiluminescence, or chemiluminescence, or any other suitable means.

[0109] As used herein, “expression” includes one or more of the following: gene transcription into precursor mRNA; splicing and other processing of precursor mRNA to produce mature mRNA; mRNA stability; translation of mature mRNA into protein (including codon use and tRNA availability); and glycosylation and / or other modifications of the translation product (if required for normal expression and function).

[0110] As used herein, a "gene" means a DNA sequence comprising the regulatory and coding sequences necessary to produce RNA that may have non-coding functions (e.g., ribosomes or transfer RNA) or may comprise polypeptides or polypeptide precursors. The RNA or polypeptide may be encoded by a full-length coding sequence or by any portion of the coding sequence, as long as the desired activity or function is preserved. Although nucleic acid sequences may be shown in DNA form, those skilled in the art will recognize that the corresponding RNA sequences will have similar sequences in which thymine is replaced by uracil, i.e., "T" is replaced by "U".

[0111] The term "genetic region" can refer to a series of sequences within or around a gene, such as introns, exons, promoters, 3' untranslated regions, etc.

[0112] As used herein, the term "hybridization" refers to the process by which two substantially complementary nucleic acid chains (at least about 65%, 75%, or 90% complementary over a segment of at least 14 to 25 nucleotides) anneal to each other under appropriately stringent conditions to form a duplex or heteroduplex by hydrogen bonding between complementary base pairs. Hybridization is typically and preferably performed with nucleic acid molecules of probe length, preferably 15-100 nucleotides, more preferably 18-50 nucleotides. Nucleic acid hybridization techniques are well known in the art. See, for example, Sambrook et al., 1989, Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Press, Plainview, NY. Hybridization and hybridization strength (i.e., the strength of association between nucleic acids) are influenced by factors such as the degree of complementarity between the nucleic acids, the stringency of the conditions involved, and the thermal melting point (T0) of the resulting hybrid. m Factors such as […] influence […]. Those skilled in the art understand how to estimate and adjust the stringency of hybridization conditions so that sequences with at least the desired level of complementarity will hybridize stably, while those with lower complementarity will not hybridize stably. For examples of hybridization conditions and parameters, see, for example, Sambrook et al., 1989, Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Press, Plainview, NY; Ausubel, FM et al., 1994, Current Protocols in Molecular Biology, John Wiley & Sons, Secaucus, NJ. In some embodiments, specific hybridization occurs under stringent hybridization conditions. Oligonucleotides or polynucleotides specific to the target nucleic acid (e.g., probes or primers) will “hybridize” with the target nucleic acid under appropriate conditions.

[0113] The term "hybridizable" means that the two polynucleotide chains of a nucleic acid are complementary at one or more nucleotide positions, such as when the nitrogenous bases of the two polynucleotide chains can form two or more Crick-Watson hydrogen bonds. For example, if a polynucleotide contains 5'ATGC 3', it can hybridize with the sequence 5'GCAT 3'. Under some experimental conditions, if a polynucleotide contains 5'GGGG 3', it can also hybridize with the incompletely complementary sequences 5'CCAC 3' and 5'CCCA 3'.

[0114] The term "non-hybridizable" means that the two polynucleotide chains of a nucleic acid are not complementary; for example, the nitrogenous bases of two separate polynucleotide chains do not form two or more Crick-Watson hydrogen bonds under strict hybridization conditions.

[0115] As used herein, the terms “individual,” “patient,” or “subject” can refer to an individual organism, a vertebrate, a mammal, or a human. In some embodiments, an individual, patient, or subject is a human.

[0116] As used herein, the term "library" refers to a collection of nucleic acid sequences, such as a collection of nucleic acids derived from the whole genome, subgenomic fragments, cDNA, cDNA fragments, cfDNA, RNA, RNA fragments, or combinations thereof. In one embodiment, some or all of the nucleic acid sequences in the library include adaptor sequences. Adaptor sequences may be located at one or both ends. Adaptor sequences can be used, for example, by sequencing methods (e.g., NGS methods), amplification, reverse transcription, sequencing, or cloning into a vector.

[0117] The library may contain a collection of nucleic acid sequences, such as target nucleic acid sequences (e.g., tumor nucleic acid sequences), reference nucleic acid sequences, or combinations thereof. In some embodiments, the nucleic acid sequences of the library may be derived from a single subject. In other embodiments, the library may contain nucleic acid sequences from more than one subject (e.g., 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, or more subjects). In some embodiments, two or more libraries from different subjects may be combined to form a library having nucleic acid sequences from more than one subject. In one embodiment, the subject has cancer or a tumor, or is at risk of having cancer or a tumor.

[0118] "Library nucleic acid sequence" refers to a nucleic acid molecule, such as DNA, RNA, or a combination thereof, that is a member of a library. In some embodiments, the library nucleic acid sequence is a DNA molecule, such as genomic DNA, cfDNA, or cDNA. In some embodiments, the library nucleic acid sequence is fragmented, such as spliced ​​or enzymatically prepared genomic DNA. In some embodiments, the library nucleic acid sequence comprises sequences derived from a subject and sequences not derived from a subject, such as adaptor sequences, primer sequences, or other sequences that allow for identification, such as "barcode" sequences.

[0119] The term "link" refers to connecting two molecules by a chemical bond to create a new molecule. For example, linking an adaptor polynucleotide to another polynucleotide can refer to forming a chemical bond between the adaptor and the polynucleotide (e.g., using a ligase or any other method) to produce a single new molecule containing both the adaptor and the polynucleotide.

[0120] As used in this article, "next-generation sequencing or NGS" refers to high-throughput parallel sequencing (e.g., simultaneously sequencing more than 10...). 3 10 4 10 5 Any sequencing method that determines the nucleotide sequence of an individual nucleic acid molecule (e.g., in single-molecule sequencing) or a clonal expansion agent of an individual nucleic acid molecule (sequencing one or more molecules). In one embodiment, the relative abundance of nucleic acid species in a library can be estimated by counting the relative number of times their homologous sequences appear in the data generated by the sequencing experiment. Next-generation sequencing methods are known in the art and are described, for example, in Metzker, M. Nature Biotechnology Reviews 11:31-46 (2010).

[0121] As used herein, an "oligonucleotide" refers to a molecule having a backbone consisting primarily of a sequence of nucleic acid bases composed of identical monomeric units spaced at defined intervals. The bases are arranged in such a manner on the backbone that they can bind to nucleic acids having a base sequence complementary to the oligonucleotide's bases. The most common oligonucleotides have a backbone of sugar-phosphate units. Oligodeoxyribonucleotides without a hydroxyl group at the 2' position can be distinguished from oligoribonucleotides with a hydroxyl group at the 2' position. Oligonucleotides may also include derivatives in which the hydrogen atom of the hydroxyl group is replaced by an organic group (e.g., an allyl group). One or more bases of an oligonucleotide may also be modified to include a thiophosphate bond (e.g., one of the two oxygen atoms in the phosphate backbone that does not participate in the internucleotide bridge is replaced by a sulfur atom) to increase resistance to nuclease degradation. Oligonucleotides acting as primers or probes in this method are typically at least about 10-15 nucleotides long, and more preferably at least about 15 to 55 nucleotides long, although shorter or longer oligonucleotides may be used in this method. The exact size will depend on many factors, which in turn depend on the final function or use of the oligonucleotide. Oligonucleotides can be generated in any manner, including, for example, chemical synthesis, DNA replication, restriction endonuclease digestion of plasmid or phage DNA, reverse transcription, PCR, or a combination thereof. Oligonucleotides can be modified, for example, by adding a methyl group, biotin or digoxigenin moiety, fluorescent tagging, or by using radioactive nucleotides.

[0122] As used herein, the term "peptide" refers to a short polymer of 2-50 amino acids linked together by peptide bonds or modified peptide bonds (i.e., peptide isosteres). In some embodiments, the length of a peptide does not exceed 10 amino acids. Peptides may contain amino acids other than the 20 genetically encoded amino acids. Peptides comprise an amino acid sequence modified by natural processes such as post-translational processing or by chemical modification techniques well known in the art.

[0123] As used herein, the term "polynucleotide" or "nucleic acid" means any RNA or DNA, which may be unmodified or modified. Polynucleotides include, but are not limited to, single-stranded and double-stranded DNA, DNA as a mixture of single-stranded and double-stranded regions, single-stranded and double-stranded RNA, RNA as a mixture of single-stranded and double-stranded regions, and hybrid molecules comprising DNA and RNA that may be single-stranded or more typically double-stranded or a mixture of single-stranded and double-stranded regions. Additionally, a polynucleotide refers to a triple-stranded region comprising RNA or DNA, or both RNA and DNA. The term polynucleotide also includes DNA or RNA containing one or more modified bases, and DNA or RNA with a backbone modified for stability or other reasons.

[0124] As used herein, the term "strict hybridization conditions" refers to hybridization conditions that are at least as strict as the following: overnight hybridization at 42°C in salmon sperm DNA sonicated with 50% formamide, 5x SSC, 50 mM NaH2PO4 (pH 6.8), 0.5% SDS, 0.1 mg / mL, and 5x Denhart solution; washing at 45°C with 2x SSC and 0.1% SDS; and washing again at 45°C with 0.2x SSC and 0.1% SDS. In another instance, strict hybridization conditions should not allow hybridization of two nucleic acids differing by more than two bases in a 20-nucleotide sequence.

[0125] As used herein, the terms “target sequence” and “target nucleic acid sequence” refer to specific nucleic acid sequences in a biological sample to be analyzed.

[0126] Limitations of conventional methods for identifying PRI

[0127] Various methods have been developed for the interaction between RNA-binding proteins (RBPs) and RNA molecules. Most of these require cross-linking of RBPs with RNA to stabilize the originally weak intermolecular interactions before isolation and in vitro manipulation, or require covalent modification of unbound RNA residues with chemicals to generate an RBP “footprint.” Short-wavelength ultraviolet (UVC) light cross-links amino acids with directly contacting nucleic acids at sub-nanometer distances and can identify precise sites of PRIs at single-nucleotide resolution. 6 UVC crosslinking of PRI forms an irreversible covalent link between two macromolecules and is the primary method used in most current methods for identifying PRI to stabilize it prior to downstream biochemical steps. These include crosslinked immunoprecipitation (CLIP) and its derivatives for identifying protein-binding sites on RNA, orthogonal organic phase separation (OOPS) and related methods for identifying all proteins capable of directly binding RNA, or mass spectrometry methods for identifying specific amino acid residues and nucleotides involved in PRI. 7 .

[0128] The use of chemical cross-linking agents (such as formaldehyde, glutaraldehyde, and heterobifunctional cross-linking agents) or RNA-modified SHAPE reagents has also been applied in the study of PRI. 8,10 However, sensitivity is limited due to several constraints: (1) the limited solvent accessibility of many PRIs to chemical cross-linking agents, (2) the immobilization of indirect PRIs due to extensive protein-protein cross-linking, and (3) the long interspace distances (7 nm and longer) of PRIs that do not actually interact in vivo, thus preventing the achievement of single nucleotide resolution for PRIs. Therefore, CLIP and derivative methods have become the current standard for identifying PRIs of specific RNAs across the transcriptome. A consortium of researchers recently published the results of their nearly decade-long work mapping PRIs for 150 unique RBPs in one or two common cancer cell lines for the ENCODE project using enhanced CLIP (eCLIP). 13,14 However, OOPS and other related methods have recently identified over 1,000 human proteins that robustly interact with RNA, demonstrating the significant limitations of eCLIP and related technologies in mapping the complete spectrum of PRIs in the human transcriptome. 7,15 Furthermore, the eCLIP study initially reported a failure rate of 1 / 3 to 1 / 4 for its initial data storage. 14 And the final statistics showed a failure rate of 1 / 2. 13

[0129] While UVC or chemical cross-linking is essential for identifying PRIs, these pose significant challenges to material handling and storage, hindering high-throughput PRI annotation and automation. PRI cross-linking produces extremely large and chemically diverse macromolecular complexes, where large hydrophilic nucleic acid polymers attach to massive protein complexes with hydrophobic properties. This results in the formation of insoluble aggregates with significant material loss and requires the use of strong surfactants that can interfere with the enzymatic reactions required for library construction. Most importantly, to our knowledge, all currently available techniques are incompatible with standard liquid handling procedures used in automated library preparation, thus drastically limiting throughput, increasing cost, and increasing the likelihood of errors or contamination. Due to these limitations, the ability of current methods to define dynamic or disease-related PRI profiles, which can only be identified through systematic, whole-transcriptome PRI annotation of all RBPs under a variety of biological conditions, is fundamentally limited.

[0130] This technique utilizes the RNA-binding protein occupancy annotation (ARORA) method on RNA.

[0131] In one aspect, this disclosure provides a method for identifying one or more RNA-binding protein (RBP)-RNA interaction (PRI) sites in at least one transcript derived from a biological sample comprising a plurality of cells, the method comprising: (a) crosslinking the plurality of cells present in the biological sample with short-wavelength ultraviolet light to stabilize a plurality of RBP-RNA complexes containing RNA transcripts, wherein each RBP-RNA complex corresponds to a PRI; (b) isolating RNA molecules from the crosslinked plurality of cells, wherein the RNA molecules comprise unbound RNA molecules and the RNA transcripts of the plurality of RBP-RNA complexes; (c) contacting the isolated RNA molecules with a protease under conditions of cleaving the RBPs of the plurality of RBP-RNA complexes to produce RNA transcripts bound by residual peptides (e.g., single amino acids), wherein each bound residual peptide (e.g., single amino acid) (i) corresponds to a PRI site and (ii) comprises an α-amino group at its N-terminus and a carboxyl group at its C-terminus; and (d) labeling the residual peptide bound by coupling the α-amino group or the carboxyl group of each bound residual peptide (e.g., single amino acid) to a chemically conjugated compound comprising an affinity ligand. (e) fragmenting the RNA molecule in the presence of heat and divalent cations to generate an RNA fragment having a 5' end and a 3' end, wherein the RNA fragment comprises an unlabeled RNA fragment and an RNA fragment labeled with the chemical partial conjugate containing the affinity ligand; (f) ligating a sequencing adaptor to the 3' end of the RNA fragment to generate an RNA fragment tagged with the adaptor; (g) capturing the RNA fragment tagged with the adaptor and labeled with the chemical partial conjugate containing the affinity ligand using affinity purification; and (h) reverse transcribing the tagged RNA fragment with the adaptor. The method involves: (i) attaching a sequencing adaptor to the 3' end of the cDNA molecule to generate multiple cDNA molecules; (j) amplifying the multiple cDNA molecules with the adaptor tag to generate a nucleic acid library of transcripts with RNA-RBP interactions; (k) sequencing and aligning the multiple cDNA molecules with the adaptor tag in the nucleic acid library to identify individual transcripts; (l) identifying the frequency of the 3' end of the cDNA at each nucleotide in the individual transcript; and (m) detecting enriched RBP binding sites in the individual transcript when the frequency of the 3' end of the cDNA at a specific nucleotide in the individual transcript meets a predetermined threshold. In some embodiments, the multiple cDNA molecules are reverse transcribed using a truncated reverse transcriptase (RT) or a read-through reverse transcriptase (RT).Truncated RTases are generated at the PRI site (i.e., the cross-linking site), or read-through RTases generate SNVs or insertions / deletions at the PRI site (i.e., the cross-linking site). Alternatively or in some embodiments, the methods disclosed herein further include identifying in the individual transcript (i) short insertions / deletions (e.g., 1-3 bp) generated by PRI isolated by affinity purification, (ii) single nucleotide variants (SNVs) generated by PRI isolated by affinity purification, or (iii) reverse transcriptase (RT) termination sites generated by PRI isolated by affinity purification.

[0132] The biological sample may be derived from breast tissue, kidney tissue, cervical tissue, endometrial tissue, head or neck tissue, gallbladder tissue, parotid gland tissue, prostate tissue, brain tissue, pituitary gland tissue, kidney tissue, muscle tissue, esophageal tissue, stomach tissue, small intestine tissue, colon tissue, liver tissue, spleen tissue, pancreas tissue, thyroid tissue, heart tissue, lung tissue, bladder tissue, adipose tissue, lymph node tissue, uterine tissue, ovarian tissue, adrenal gland tissue, testicular tissue, tonsils, thymus, blood, hair, oral cavity, skin, serum, plasma, CSF, semen, prostatic fluid, seminal plasma, urine, feces, sweat, saliva, sputum, mucus, bone marrow, lymph, or tears. In some embodiments, the biological sample is a fresh or frozen sample.

[0133] In some embodiments, the crosslinking UV box operates at approximately 50-600 mJ / cm. 2 The dosage is applied for approximately 30 seconds to 10 minutes. In some embodiments, the crosslinking UV box is used at approximately 50 mJ / cm². 2 Approximately 100 mJ / cm 2 Approximately 150 mJ / cm 2 Approximately 200 mJ / cm 2 Approximately 250 mJ / cm 2 Approximately 300 mJ / cm 2 Approximately 350 mJ / cm 2 Approximately 400 mJ / cm 2 Approximately 450 mJ / cm 2 Approximately 500 mJ / cm 2 Approximately 550 mJ / cm 2 or about 600 mJ / cm 2 The dosage is applied for approximately 30 seconds, 45 seconds, 1 minute, 1.5 minutes, 2 minutes, 2.5 minutes, 3 minutes, 3.5 minutes, 4 minutes, 4.5 minutes, 5 minutes, 5.5 minutes, 6 minutes, 6.5 minutes, 7 minutes, 7.5 minutes, 8 minutes, 8.5 minutes, 9 minutes, 9.5 minutes, or approximately 10 minutes. In some embodiments, the crosslinking UV box operates at 300-500 mJ / cm².2 The dose is administered over approximately 3 minutes or at a rate of 100-400 mJ / cm. 2 The dosage is applied for approximately 10 minutes. In other embodiments, the crosslinking is performed using a UV laser for approximately 10-20 seconds.

[0134] Alternatively or concurrently, in some embodiments, the isolated RNA molecules are treated with the protease at approximately 25°C–65°C for approximately 5 to 60 minutes. In some embodiments, the isolated RNA molecules are treated with the protease at approximately 25°C, approximately 27.5°C, approximately 30°C, approximately 32.5°C, approximately 35°C, approximately 37.5°C, approximately 40°C, approximately 42.5°C, 45°C, approximately 47.5°C, approximately 50°C, approximately 52.5°C, approximately 55°C, approximately 57.5°C, approximately 60°C, approximately 62.5°C, or approximately 65°C for approximately 5 minutes, approximately 7.5 minutes, or approximately 10 minutes, respectively. The time intervals are approximately 12.5 minutes, 15 minutes, 17.5 minutes, 20 minutes, 22.5 minutes, 25 minutes, 27.5 minutes, 30 minutes, 32.5 minutes, 35 minutes, 37.5 minutes, 40 minutes, 42.5 minutes, 45 minutes, 47.5 minutes, 50 minutes, 52.5 minutes, 55 minutes, 57.5 minutes, or 60 minutes. In some embodiments, the isolated RNA molecule is treated with the protease at 50°C for approximately 30 minutes. In some embodiments, the protease is proteinase K. Alternatively or in some embodiments, the length of each bound residual peptide is no more than 10 amino acids, no more than 9 amino acids, no more than 8 amino acids, no more than 7 amino acids, no more than 6 amino acids, no more than 5 amino acids, no more than 4 amino acids, no more than 3 amino acids, no more than 2 amino acids, or a single amino acid.

[0135] Alternatively or additionally, in some embodiments, the method of this technology further includes removing genomic DNA after step (c) to obtain purified RNA molecules. In some embodiments, the method disclosed herein further includes contacting the purified RNA molecules with a protease at 50°C for about 30 minutes prior to step (d), optionally wherein the protease is proteinase K.

[0136] In any of the foregoing embodiments, the method of this technology further includes enriching the isolated or purified RNA molecule with a poly-dT oligonucleotide or a targeting decoy capture reagent after step (c) but before step (d). In other embodiments, the method of this technology further includes enriching the cDNA molecule with an adaptor tag with a targeting decoy capture reagent after step (j) but before step (k). In some embodiments, the targeting decoy capture reagent hybridizes with one or more disease-related genes, such as genes associated with cancer, neurodegenerative diseases, autoimmune diseases, metabolic diseases, etc. Examples of disease-related genes include, but are not limited to, AKT1, ALK, APC, AR, ARAF, ARID1A, ARID2, ATM, B2M, BCL2, BCOR, BRAF, BRCA1, BRCA2, CARD11, CBFB, CCND1, CDH1, CDK4, CDKN2A, CIC, CREBBP, CTCF, CTNNB1, DICER1, DIS3, DNMT3A, EGFR, EIF1AX, EP300, ERBB2, ER BB3, ERCC2, ESR1, ETV6, EZH2, FBXW7, FGFR1, FGFR2, FGFR3, FGFR4, FLT3, FOXA1, FOXL2, FOXO1, FUBP1, GATA3, GNA1 1. GNAQ, GNAS, H3F3A, HIST1H3B, HRAS, IDH1, IDH2, IKZF1, INPPL1, JAK1, KDM6A, KEAP1, KIT, KNSTRN, KRAS, MAP2K1 , MAPK1, MAX, MED12, MET, MLH1, MSH2, MSH3, MSH6, MTOR, MYC, MYCN, MYD88, MYOD1, NF1, NFE2L2, NOTCH1, NRAS, NTR K1, NTRK2, NTRK3, NUP93, PAK7, PDGFRA, PIK3CA, PIK3CB, PIK3R1, PIK3R2, PMS2, POLE, PPP2R1A, PPP6C, PRKCI, PTC H1, PTEN, PTPN11, RAC1, RAF1, RB1, RET, RHOA, RIT1, ROS1, RRAS2, RXRA, SETD2, SF3B1, SMAD3, SMAD4, SMARCA4, SMA RCB1, SOS1, SPOP, STAT3, STK11, STK19, TCF7L2, TGFBR1, TGFBR2, TP53, TP63, TSC1, TSC2, U2AF1, VHL, XPO1 and TERT.

[0137] In any of the above embodiments, the method of this technology further includes depleting the ribosomal RNA from the isolated or purified RNA molecule after step (c) but before step (d).

[0138] In any and all embodiments of the methods disclosed herein, labeling the RNA transcript bound by the residual peptide comprises thermally denaturing the isolated or purified RNA molecule and incubating the isolated or purified RNA molecule with the chemically conjugate containing the affinity ligand in a reaction buffer supplemented with about 5%–70% DMSO. In some embodiments, the reaction buffer is supplemented with about 5%, about 10%, about 15%, about 20%, about 25%, about 30%, about 35%, about 40%, about 45%, about 50%, about 55%, about 60%, about 65%, or about 70% DMSO. Examples of suitable reaction buffers include, but are not limited to, phosphate buffers (e.g., PBS buffer), carbonate buffers (e.g., carbonate-bicarbonate buffer), HEPES, borate, 2-[morpholino]ethanesulfonic acid (MES), or any buffer free of free amine or free carboxyl groups.

[0139] Alternatively or concurrently, in some embodiments, thermal denaturation includes heating the isolated or purified RNA molecule at about 50°C to 100°C for 1 to 5 minutes. In some embodiments, thermal denaturation includes heating the isolated or purified RNA molecule at about 50°C, about 55°C, about 60°C, about 65°C, about 70°C, about 75°C, about 80°C, about 85°C, about 90°C, about 95°C, or about 100°C for 1, 2, 3, 4, or 5 minutes.

[0140] In some embodiments of the methods disclosed herein, the chemically moietraconjugate is coupled to the α-amino group of each bound residual peptide. In some embodiments, the chemically moietraconjugate comprises a succinimide ester, a carboxylic acid ester, a tetrafluorophenyl ester, a sulfonyl dichlorophenol ester, a carbonyl azide, an aldehyde, or an isothiocyanate. In other embodiments of the methods disclosed herein, the chemically moietraconjugate is coupled to the carboxyl group of each bound residual peptide. In some embodiments, the chemically moietraconjugate comprises a primary amine, a carbodiimide, or an isocyanate. In any and all embodiments of the methods disclosed herein, the chemically moietraconjugate containing the affinity ligand is added to a final concentration of about 0.1 mM to about 10 mM and incubated for about 3 minutes to about 5 hours. In some embodiments, the chemical conjugate containing the affinity ligand is added to about 0.1 mM, about 0.2 mM, about 0.3 mM, about 0.4 mM, about 0.5 mM, about 0.6 mM, about 0.7 mM, about 0.8 mM, about 0.9 mM, about 1 mM, about 1.1 mM, about 1.2 mM, about 1.3 mM, about 1.4 mM, about 1.5 mM, about 1.6 mM, about 1.7 mM, about 1.8 mM, about 1.9 mM, about 2 mM, about 2.1 mM, about 2.2 mM, about 2.3 mM, about 2.4 mM, about 2.5 mM, about 2.6 mM, about 2.7 mM, about 2.8 mM, about 2.9 mM, about 3 mM, about 3.1 mM, about 3.2 mM, about 3.3 mM, about 3.4 mM. mM, approximately 3.5 mM, approximately 3.6 mM, approximately 3.7 mM, approximately 3.8 mM, approximately 3.9 mM, approximately 4 mM, approximately 4.1 mM, approximately 4.2 mM, approximately 4.3 mM, approximately 4.4 mM, approximately 4.5 mM, approximately 4.6 mM, approximately 4.7 mM, approximately 4.8 mM, approximately 4.9 mM, approximately 5 mM, approximately 5.1 mM, approximately 5.2 mM, approximately 5.3 mM, approximately 5.4 mM, approximately 5.5 mM, approximately 5.6 mM, approximately 5.7 mM, approximately 5.8 mM, approximately 5.9 mM, approximately 6 mM, approximately 6.1 mM, approximately 6.2 mM, approximately 6.3 mM, approximately 6.4 mM, approximately 6.5 mM, approximately 6.6 mM, approximately 6.7 mM, approximately 6.8 mM, approximately 6.9 mM, approximately 7 mM, approximately 7.1 mM, approximately 7.2 mM, approximately 7.3 mM, approximately 7.4 mM, approximately 7.5 mM, approximately 7.6 mM, approximately 7.7 mM, approximately 7.8 mM, approximately 7.9 mM, approximately 8 mM, approximately 8.1 mM, approximately 8.2 mM, approximately 8.3 mM, approximately 8.4 mM, approximately 8.5 mM, approximately 8.6 mM, approximately 8.7 mM, approximately 8.8 mM, approximately 8.9 mM, approximately 9 mM, approximately 9.1 mM, approximately 9.2 mM, approximately 9.3 mM, approximately 9.Approximately 4 mM, 9.5 mM, 9.6 mM, 9.7 mM, 9.8 mM, 9.9 mM, or 10 mM were incubated for approximately 3 minutes, 4 minutes, 5 minutes, 6 minutes, 7 minutes, 8 minutes, 9 minutes, 10 minutes, 15 minutes, 20 minutes, 25 minutes, 30 minutes, 35 minutes, 40 minutes, 45 minutes, 50 minutes, 55 minutes, 1 hour, 1.1 hours, 1.2 hours, 1.3 hours, 1.4 hours, 1.5 hours, 1.6 hours, 1.7 hours, 1.8 hours, 1.9 hours, 2 hours, 2.1 hours, and 2.2 hours, respectively. Approximately 2.3 hours, 2.4 hours, 2.5 hours, 2.6 hours, 2.7 hours, 2.8 hours, 2.9 hours, 3 hours, 3.1 hours, 3.2 hours, 3.3 hours, 3.4 hours, 3.5 hours, 3.6 hours, 3.7 hours, 3.8 hours, 3.9 hours, 4 hours, 4.1 hours, 4.2 hours, 4.3 hours, 4.4 hours, 4.5 hours, 4.6 hours, 4.7 hours, 4.8 hours, 4.9 hours, or 5 hours.

[0141] In any and all embodiments of the methods disclosed herein, the affinity ligand comprises biotin, a biotin derivative, digoxigenin, dinitrophenyl, a click chemistry reagent, a sugar, or a peptide. In some embodiments, the click chemistry reagent comprises an alkyne group, an azide group, a dibenzocyclooctyne group (DBCO), or a bicyclic nonyne (BCN) group.

[0142] Alternatively or alternatively, in some embodiments, fragmenting the RNA molecule includes using approximately 10 mM-30 mM of a divalent cation (e.g., Mg). 2+ Zn 2+ In the presence of [a specific substance], the RNA molecule is heated at 90°C-95°C for approximately 3-12 minutes. In some embodiments, fragmentation of the RNA molecule includes using approximately 10 mM, approximately 15 mM, approximately 20 mM, approximately 25 mM, or approximately 30 mM of divalent cations (e.g., Mg). 2+ Zn 2+ In the presence of [a specific substance], the RNA molecule is heated at 90°C, 91°C, 92°C, 93°C, 94°C, or 95°C for approximately 3 minutes, 4 minutes, 5 minutes, 6 minutes, 7 minutes, 8 minutes, 9 minutes, 10 minutes, 11 minutes, or 12 minutes, respectively. The length of the RNA fragment can be approximately 40-90 nucleotides or approximately 80-250 nucleotides.

[0143] In any and all embodiments of the methods disclosed herein, the biological sample is obtained from a subject diagnosed with a disease. In some embodiments, the disease is cancer, an autoimmune disease, a metabolic disease, or a neurodegenerative disease. Examples of cancer include, but are not limited to, adrenal cancer, bladder cancer, leukemia, bone cancer, brain cancer, breast cancer, carcinoma, cervical cancer, colon cancer, colorectal cancer, endometrial cancer, ear, nose, and throat (ENT) cancer, endometrial cancer, esophageal cancer, gastrointestinal cancer, head and neck cancer, Hodgkin's disease, colon cancer, kidney cancer, laryngeal cancer, leukemia, liver cancer, lymph node cancer, lymphoma, diffuse large B-cell lymphoma (DLBCL), lung cancer, melanoma, mesothelioma, myeloma, nasopharyngeal carcinoma, neuroblastoma, non-Hodgkin's lymphoma, oral cancer, ovarian cancer, pancreatic cancer, penile cancer, pharyngeal cancer, prostate cancer, rectal cancer, sarcoma, seminoma, skin cancer, gastric cancer, teratoma, testicular cancer, thyroid cancer, uterine cancer, vaginal cancer, vascular tumors and their metastases.

[0144] Examples of neurodegenerative diseases include, but are not limited to, age-related memory impairment (AAMI), mild cognitive impairment (MCI), Alzheimer's disease, Down syndrome, boxer's dementia, cognitive dysfunction syndrome, multiple system atrophy, inclusion body myositis, Dutch hereditary cerebral hemorrhage with amyloidosis, Niemann-Pick disease type C, cerebral β-amyloid angiopathy, dementia associated with corticobasal degeneration, amyloidosis, Creutzfeldt-Jakob disease, Gerstmann-Strauss syndrome, kuru, pruritus, Huntington's disease, Parkinson's disease, ataxia, motor neuron disease, progressive supranuclear palsy, and amyotrophic lateral sclerosis (ALS).

[0145] Examples of autoimmune diseases include, but are not limited to, vasculitis such as antineutrophil cytoplasmic antibodies (ANCA), ANCA-associated vasculitis (AAV) or giant cell arteritis (GCA), Sjögren's syndrome, inflammatory bowel disease (IBD), pemphigus vulgaris, lupus nephritis, psoriasis, thyroiditis, type 1 diabetes mellitus, idiopathic thrombocytopenic purpura (ITP), ankylosing spondylitis, multiple sclerosis, systemic lupus erythematosus (SLE), rheumatoid arthritis, Crohn's disease, myasthenia gravis, neuromyelitis optica (NMO), IgG4-related disease, systemic sclerosis, insulin-dependent diabetes mellitus (IDDM), ankylosing spondylitis, atopic dermatitis, uveitis, and graft-versus-host disease (GVHD).

[0146] Examples of metabolic disorders include, but are not limited to, familial hypercholesterolemia, Gaucher disease, Hunter syndrome, Krab disease, maple syrup diabetes, metachromatic leukodystrophy, mitochondrial encephalopathy, lactic acidosis-like stroke (MELAS), Niemann-Pick disease, phenylketonuria (PKU), porphyria, Ty-Sachs disease, and Wilson's disease.

[0147] In any and all embodiments of the methods disclosed herein, one or more of the enriched RBP binding sites detected in the individual transcript are allele-specific and / or associated with diseases (e.g., cancer, neurodegenerative diseases, autoimmune diseases, metabolic diseases, etc.). In any of the foregoing embodiments, the method of this technology further includes identifying the RBP that binds to one or more of the enriched RBP binding sites detected in the individual transcript. In some embodiments, the identity of the RBP that binds to one or more of the detected enriched RBP binding sites in the individual transcript is determined by identifying RBP-specific crosslinking patterns within sequence motifs in the individual transcript (e.g., via computational analysis). In some embodiments, the RBP crosslinking patterns are identified using shared motifs defined by eCLIP, mCross crosslinking patterns, position-enriched k-mer analysis (PEKA), or sequence motifs defined by RBNS.

[0148] In one aspect, this disclosure provides a method for evaluating the efficacy of a drug in inhibiting the interaction between a specific RNA-binding protein (RBP) and an RNA transcript, the method comprising (a) contacting a first cell population with at least one repressive oligonucleotide that specifically inhibits the expression of RBP, wherein the first cell population is obtained from a biological sample; (b) contacting a second cell population with a test drug, wherein the second cell population is obtained from the biological sample and has not been treated with the at least one repressive oligonucleotide; (c) performing the ARORA method described herein to identify one or more RBP-RNA interaction (PRI) sites in at least one transcript derived from the first cell population, the second cell population, and a control cell population, wherein the control cell population is obtained from the biological sample and has not been treated with the at least one repressive oligonucleotide and the test drug; (d) defining an RBP-specific PRI by identifying a crosslinking pattern within one or more sequence motifs in the individual transcript, the crosslinking pattern being (i) present in the control cell population and (ii) absent in the first cell population; and determining that the test drug is effective when the RBP-specific PRI identified in step (d) is absent in the second cell population. In another aspect, this disclosure provides a method for evaluating the specificity of a drug in inhibiting the interaction between a specific RNA-binding protein (RBP) and RNA transcripts, the method comprising: (a) contacting a first cell population with at least one repressive oligonucleotide that specifically inhibits the expression of RBP, wherein the first cell population is obtained from a biological sample; (b) contacting a second cell population with a test drug, wherein the second cell population is obtained from the biological sample and has not been treated with the at least one repressive oligonucleotide; (c) performing the ARORA method described herein to identify one or more RBP-RNA interaction (PRI) sites in at least one transcript derived from the first cell population, the second cell population, and a control cell population, wherein the control cell population is obtained from the biological sample and has not been treated with the at least one repressive oligonucleotide and the test drug; and (d) identifying the nonspecific effect of the test drug by identifying a crosslinking pattern within one or more sequence motifs in the individual transcript, the crosslinking pattern being (a) present in the control cell population and the first cell population and (b) absent in the second cell population. The at least one repressive oligonucleotide may be siRNA, shRNA, antisense oligonucleotide, or sgRNA. In some embodiments, the test drug is a small molecule, nucleic acid, or peptide.

[0149] Example

[0150] The following examples further illustrate the technology of the present invention, and these examples should not be construed as limiting it in any way.

[0151] Example 1: Materials and Methods

[0152] Cell lines and treatment

[0153] HepG2, K562, 293T, and BJ fibroblasts were obtained from ATCC and grown in recommended basal medium containing 10% FBS. Actinomycin and anisoxin were obtained from Sigma Aldrich and used at a concentration of 100 µM for 1 hour to induce translational stress.

[0154] RNA preparation

[0155] Mammalian cells HEK293T, HepG2, or K562 were grown in 15 cm dishes until they reached a maximum confluence of 80%-90%. The cells were washed twice with PBS and then incubated in a UV incubator (Spectrolinker #XL-1000) at 254 nm with 100-400 mJ / cm². 2 Crosslinking at the specified dosage. Cells were scraped and precipitated immediately after crosslinking. In the non-crosslinked control, cells were directly precipitated and stored at -80°C before use.

[0156] RNA was extracted using the Monarch Total RNA Mini-Prep Kit (NEB #T2010) with some modifications. Briefly, 600 µL of 1X protection reagent, 60 µL of proteinase K reaction buffer, and 30 µL of proteinase K were added to the sample, and the sample was incubated at 50 °C for 30 min. The sample was then processed according to the protocol to remove genomic DNA with a DNA removal column, followed by RNA purification and elution. The eluted RNA was then treated again with proteinase K at 50 °C for 30 min, followed by RNA purification using the Zymo RNA Purification and Concentration-5 Kit, retaining transcripts >200 nt in length. For transcriptome experiments, poly-A tail RNA was enriched using the poly-A purification kit (Thermo Fisher #AM1922) through two sequential enrichments, followed by ribosomal RNA depletion using the ribosomal depletion kit (Galen Molecular #dp-K024-000054).

[0157] RNA labeled with N-hydroxysuccinimide (NHS) conjugate

[0158] Heat 1 µg of total RNA or ribose-depleted RNA at 70 °C for 2 min and then rapidly cool on ice. Dilute the RNA in a labeling buffer consisting of 1X PBS pH 7.4 supplemented with 50% DMSO to limit RNA secondary structure formation, which may impair labeling at some crosslinking sites. Add NHS-dPEG-Biotin (Sigma-Aldrich #QBD10200) to a final concentration of 1 mM and incubate the reaction at room temperature for 30 min. Terminate the reaction with Tris and purify the RNA using the RNeasy Mini Kit (Qiagen #74104).

[0159] RNA dot blot

[0160] 50 ng RNA was spotted onto a Hybond-N+ membrane (Thermo Fisher #45-000-763). Then, it was heated at 120 mJ / cm² at 254 nm. 2 The dose was passed through a UV crosslinked membrane and spotted for 1 hour using an IRDye 800CW streptavidin, followed by imaging on a LI-COR Odyssey CLx.

[0161] RNA fragmentation, size selection, and dephosphorylation

[0162] 1 µg of biotinylated RNA per sample was incubated in 20 µL of 2X 10X Fast AP buffer (ThermoFisher) at 94 °C for 9 min on a thermal cycler block to obtain most RNA fragments in the 40–100 nucleotide length range. The RNA fragments were then subjected to a two-step size selection using AMPure RNAClean XP beads. In the first step, 22 µL of RNAClean XP beads and 12 µL of isopropanol were added. 25 µL of RNAClean XP beads and 41 µL of isopropanol were added to the flow-through buffer (enriching fragments <approximately 100 nucleotides) to bind fragments longer than 40 nucleotides to the beads. The eluted, size-selected RNA fragments were then treated with FastAP alkaline phosphatase and T4 PNK to dephosphorylate the fragments. The dephosphorylated fragments were purified again with RNAClean XP beads and eluted in 10 µL of H2O. RNA content was quantified by Qubit, and 1 µL of Superasin was added before storage at -80 °C.

[0163] RNA 3' ligation and streptavidin pull-down

[0164] Two biological replicates were used for each condition: two uncrosslinked samples and two biological replicates for crosslinked samples at 400 J / m². 200 ng of RNA and 100 pmol of each of two different indexed 3' oligonucleotides from each sample were denatured on a heating block at 70 °C for 2 min, followed by rapid cooling on ice. A mixture of T4 RNA ligase 1 was added, and the ligation reaction was incubated at room temperature for 90 min. The ligation reaction was terminated by adding RLT buffer and EDTA. Individual, uniquely indexed samples from all four samples were then pooled, purified with RNAClean XP beads, and eluted in 20 µL H₂O. 2 µL of each sample was retained as input and stored on ice until cDNA synthesis.

[0165] MyOne-C1 streptavidin beads (ThermoFisher, 5 µL beads / 200 ng RNA) were washed multiple times with binding and washing (B&W) buffer and resuspended on ice in 600 µL of B&W buffer containing 3 µL of superasin. 18 µL of 3'-ligated RNA was denatured on a heating block at 70 °C for 2 min, followed by rapid cooling on ice. The denatured RNA was added to the MyOne-C1 beads and incubated at 4 °C with inverted rotation for 1 h. The RNA-bound paramagnetic beads were precipitated in a magnetic field, and the supernatant was discarded. The beads were washed three times with ice-cold B&W buffer, followed by one wash with 5 mM Tris 7.5.

[0166] cDNA synthesis

[0167] Reverse transcription (RT) was performed using Maxima H-reverse transcriptase (ThermoFisher) or AffinityScript (Agilent). RNA-bound paramagnetic beads were resuspended in 20 µL of Maxima H-reverse transcriptase reaction buffer or AffinityScript buffer, AR2 primers, and 0.3 µL of mouse RNase inhibitor. The Maxima H-reverse transcriptase reaction mixture was separated and added to the input sample to a final volume of 20 µL. The beads were vigorously mixed to ensure suspension, and then incubated at 55 °C on a ThermoMixer block (Eppendorf) for 40 min, pulsed at 1200 rpm for 10 seconds every 10 min to maintain bead suspension. After 40 min of reverse transcription, in a pull-down sample with RNA bound to the beads, the beads were placed in a magnet, the RT reaction mixture was removed, and the beads were washed once with 200 µL of 5 mM Tris 7.5, then resuspended in 20 µL of 5 mM Tris 7.5. Next, 1N NaOH was added to all samples (drop-down and input), and the samples were incubated on a ThermoMixer at 70°C for 12 minutes, mixed at a constant 1200 rpm to degrade RNA. The paramagnetic beads were then removed in a magnet. The cDNA samples were purified using AMPure beads.

[0168] The second sequencing adaptor is ligated to the 3' end of the cDNA.

[0169] Add 40 pmol of 3Tr3-10N adaptor to each purified cDNA sample and denature the mixture at 75°C for 3 minutes on a heating block, followed by rapid cooling. Add T4 RNA ligase 1 mixture and incubate the ligation reaction overnight at room temperature. The next day, purify the cDNA twice sequentially with AMPure beads and then elute in 20 µL of 5 mM Tris 7.5.

[0170] Library quantification, amplification, and size selection

[0171] Each 1 µL library was diluted 1:10 with nuclease-free water, and qPCR was performed in triplicate with 1 µL of diluted cDNA in each well to obtain the cycle threshold (C). T This refers to the estimated value of the obtained library fragments. Compared to the enrichment of non-crosslinked samples, we typically obtain enrichment of crosslinked samples with values ​​exceeding >20X on pull-down. Then, a 10 µL cDNA library was amplified to a total C1 using the barcode Illumine adaptor. T- 3 cycles. The library is purified using AMPure to remove excess primers, and then size-selected on a 4% agarose gel to collect fragments of 200-350 bp. Before sequencing, the library's molar concentration and fragment length are analyzed using a bioanalyzer.

[0172] RNA sequencing

[0173] ARORA libraries were sequenced using an Illumina PE100, with each library containing a total of 5 million reads for ribosomal RNA interaction analysis. ARORA libraries used for whole transcriptome analysis of RBP-RNA interactions were also sequenced, with each library containing a total of 400 million reads.

[0174] Computational analysis

[0175] The paired-end sequencing reads contain two types of molecular barcodes. First, read1 contains a specific 9 bp nucleotide sequence for multiple purposes. Second, both read1 and read2 contain unique molecular identifiers (UMIs) (random oligonucleotide sequences) to uniquely identify the input RNA molecule. We used a UMI tool to extract the 9 bp multiple barcode and 4 bp (read1) and 1 bp (read2) deduplicated barcode sequences, and appended them to the read names. Additionally, adaptor and polyG tail trimming was performed using FASTP. After extraction and trimming, the paired-end reads were aligned to the human reference genome (hg19) using the STAR alignment tool.

[0176] Then, using the bamtools filter, each of the initial four experiments (two cell lines, each with dropdown and input) was divided into eight different crosslinking conditions (four conditions, two replicates of each), resulting in a total of 32 BAM files, one for each cell line and condition. The UMI tool was then used to remove duplicates from each of the 32 BAM files based on a 14bp UMI (read1 + read2).

[0177] Finally, based on the CIGAR information contained in the BAM file, we identified single nucleotide deletions, SNVs, and reverse transcriptase (RT) termination sites generated from PRI isolated by affinity purification.

[0178] Example 2: Cross-linked oligopeptides as specific attachment chemicals for affinity purification of RBP-RNA interaction sites.

[0179] The UVC cross-linking of the RBP-RNA complex produces irreversible cross-linking at precise nucleotide sites that directly contact the amino acids. 6Although intact protein-RNA complexes are recovered via immunoprecipitation in methods such as CLIP, we infer that orthogonal chemistry of amino acids could be used to identify PRIs that are both general and specific to amino acids, regardless of the identity of the specific amino acid interacting with RNA. Figure 1A Here, the UVC-crosslinked protein-RNA complex is subjected to extensive proteolysis to remove the vast majority of the amino acid content of the crosslinked RNA, leaving only minimal residual peptides (or single amino acids), but exposing an α-amino group at the N-terminus of each peptide and a carboxyl group at the C-terminus, regardless of amino acid identity. Endogenous RNA, including those with known post-transcriptional chemical modifications, does not contain reactive free amino or carboxyl groups. Therefore, this method generates orthogonal chemical moieties on RNA at precise sites of UVC crosslinking of PRI. Furthermore, since the vast majority of the protein content is released from the RNA via proteolysis, the resulting RNA has the same properties as standard uncrosslinked RNA. We have observed that UVC-crosslinked NHS-biotinylated RNA can undergo standard treatments for RNA-seq library preparation, including separation and purification with paramagnetic beads, automated liquid handling, consistent and tunable fragmentation with divalent cations and heat, enrichment with poly-dT oligonucleotides or targeted hybridization capture, or ribosomal RNA depletion. Therefore, this method generates orthogonal chemical moieties at any site of PRI and is expected to be compatible with any commercial process used for RNA-seq library preparation.

[0180] ARORA utilizes orthogonal chemistry generated from free α-amino or carboxyl groups on RNA-crosslinked oligopeptides for site-specific attachment of chemical labels or affinity ligands. N-hydroxysuccinimide (NHS) esters are commonly used in orthogonal linking chemistry for proteins and react selectively with free α- or amino groups at the N-terminus or ε-amino groups in lysine side chains, but react poorly with RNA or DNA. Following RNA thermal denaturation, RNA is incubated with NHS-biotin in PBS buffer supplemented with 50% DMSO at room temperature to reduce RNA secondary structure during labeling, and optimal labeling is achieved by enhancing the nucleophilic attack of NH2 on amino acids or peptides by NHS-biotin. NHS-biotin broadly labels RNA isolated from UVC-crosslinked cells but not RNA isolated from cells subjected to simulated irradiation. Figure 1BImportantly, we observed that RNA cross-linked with UVC in vitro was not labeled in the absence of protein, confirming that NHS-biotin labeling of free amino groups on UVC-crosslinked RNA requires protein-derived amino groups and is not generated due to UVC-induced RNA nucleotide damage. RNA from cells irradiated with NHS-biotin-labeled UVC showed no degradation compared to unirradiated RNA and unlabeled RNA (RIN score 9-10). RNA gel electrophoresis showed that biotin-labeled RNA was present across a wide molecular weight range similar to the typical pattern of total RNA, with high signal intensity corresponding to abundant 28S and 18S ribosomal subunits and lower labeling intensity corresponding to lower mRNA molecular weights (…). Figure 1C The extent to which NHS adapter marker RNA is present is directly proportional to the delivered UV-C dose. Figure 1D Streptavidin paramagnetic beads recovered biotinylated RNA from various types of cellular RNA, including the ribosomal (28S) large subunit, GAPDH mRNA, and the long non-coding RNA MALAT1, in a UVC dose-dependent manner, with increasing UVC cross-linking dose leading to a sequential increase in the yield of each RNA, consistent with the increasing PRI cross-linking efficiency with increasing UVC dose. Figure 1E Therefore, NHS attachment chemicals effectively and specifically label amino acid moieties covalently cross-linked with RNA via UVC.

[0181] We also found that attachment chemistry targeting the C-terminal carboxyl group also allows for highly specific labeling of peptides cross-linked with RNA. Figure 4A The chemical substance 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) can activate free carboxyl groups and promote the attachment of amino-labeled chemicals. In reactions containing EDC and amino-PEG-biotin, RNA isolated from cells after cross-linking with UVC is readily labeled with biotin, while RNA isolated from uncross-linked cells or RNA cross-linked in vitro by UVC in the absence of proteins is rarely labeled. Therefore, these data demonstrate that both the primary amine at the N-terminus of a peptide and the carboxyl group at the C-terminus of a peptide provide unique orthogonal chemical motifs for labeling amino acids cross-linked with RNA.

[0182] Example 3: ARORA transtranscriptome for identifying protein-RNA interactions at single nucleotide resolution

[0183] Enriching NHS-biotin-tagged RNA with streptavidin beads can identify cellular RNA regions with high-density RBP interactions; however, identifying PRI at single-nucleotide resolution would provide more information about the location, identity, and RNA motif that determines RBP-RNA interactions. The CLIP method has previously demonstrated that different reverse transcriptases exhibit characteristic behaviors around PRI sites, allowing for the differentiation of PRI at single-nucleotide resolution. 6,16 Some RTs exhibit a high propensity for cross-link-induced truncation, where the final nucleotide incorporated by the RT enzyme typically corresponds to the nucleotide cross-linked in the PRI. For other RT enzymes, cross-link-induced mutations (such as deletions or substitutions) occur at a high frequency near the PRI. Because ARORA utilizes extensive protease treatment followed by the covalent addition of bulk adducts, PRI labeled with ARORA possesses a chemically distinct amino acid-nucleic acid structure compared to RNA fragments typically sequenced in CLIP. Therefore, we investigated the effect of ARORA labeling of nucleic acids on cross-link-induced effects on various RTs. We generated in vitro synthesized oligonucleotide templates with a single amino-C2-deoxythymidine residue (amino-C2dT), which is chemically similar to a single lysine amino acid cross-linked with thymidine or uracil, providing a single internal site for NHS labeling of the oligonucleotide. Primer extension assays performed with four distinct RT enzymes after ARORA labeling and streptavidin purification of the oligonucleotide revealed two distinct RT behavior patterns around the amino-C2dT residue. Each of the three RT enzymes (Maxima H-, SuperScript IV, and TGIRT) produces approximately 50% of the RT-generated cDNA, with a molecular weight of about 70 nucleotides, consistent with the truncation of cDNA synthesis at the amino-C2dT residue. 10%-20% of the fragments have a molecular weight of approximately 95-98 nt, consistent with the deletion of 2-5 nucleotides, while 30% of the cDNA is full-length (100 nt). Figure 1F Importantly, efficient truncation and mutation were only observed when cDNA was synthesized from bead-bound RNA fragments, while fragments in free solution exhibited significantly lower rates of both types of cross-linking-induced effects. Without being bound by theory, it is believed that immobilization of cross-linked nucleic acids increases steric hindrance for RT enzymes and prevents sustained synthesis through the cross-linking sites. In contrast to the three enzymes with efficient RT termination and deletion, one enzyme, AffinityScript, showed no significant truncation or deletion of cDNA, whether synthesized in solution or bound to beads, consistent with the reverse transcription readthrough pattern.

[0184] The ARORA library generated by Maxima H-RT was analyzed using total cellular RNA. Figure 4BSequencing of RNA from cells cross-linked at 400 mJ / cm² and enriched with streptavidin beads confirmed that RNA isolated from these cells resulted in an average of 0.08 deletions and 0.0025 substitutions per read, significantly higher than RNA from non-cross-linked cells and RNA from cross-linked cells not enriched with streptavidin beads. Figure 1G No significant deletions or substitutions were observed in the libraries generated using AffinityScript RT. Therefore, ARORA libraries generated using Maxima H-RT are suitable for analyzing PRIs identified by crosslink-induced truncation and mutations, while ARORA libraries generated using AffinityScript can be used in situations requiring readthrough of crosslink sites (e.g., in the identification of allele-specific PRIs).

[0185] UVC crosslinks almost all amino acids with RNA, but favors pyrimidines over purines, with the order U>C>G / A. Consistent with this, we observed enrichment of U>C>G>A at significantly enriched RT termination, deletion, or substitution sites. Figure 4C Interestingly, we observed a slight enrichment of RT termination at the A residue in both the NCL_PD and x400_PD samples compared to their inputs, suggesting that NHS tagging under these conditions may be moderately reactive to the C6 amino group of A in the absence of PRI. However, this enrichment was compensated for when calculating the per-nucleotide enrichment of the X400_PD sample compared to the NCL_PD sample. Figures 4D to 4E ).

[0186] Optimal fragmentation of RNA is crucial for elucidating PRI at single nucleotide resolution, which was observed in RNA fragments of 40–90 nucleotides in length. 17 cDNA generated from larger RNA fragments exhibits significantly higher cDNA truncation rates at non-crosslinking sites, thus significantly reducing the ability to achieve single-nucleotide resolution (PRI). However, whole-transcriptome methods such as CLIP utilize enzymatic fragmentation of RNA to maintain protein integrity, which is crucial for effective antibody IP. Unfortunately, this approach introduces significant biases due to RNase bias towards single- or double-stranded secondary structures, selectivity in cleavage to specific bases, and limited accessibility of certain regions of secondary and tertiary RNA structures at biological temperatures. 14In contrast, ARORA-generated RNA consists of minimal amino acid adducts on the RNA, which are essentially irreversible and highly stable to heat and denaturing agents. The ARORA label on the RNA remains intact after liquefaction, high salt concentrations, or boiling for >20 minutes. Therefore, ARORA-labeled RNA can be fragmented using heat and divalent cations, a method without structural or base bias, and can generate RNA fragment size distributions in a predictable and reproducible manner. Figure 5A We performed ARORA analysis on total RNA from HepG2 cells using AffinityScript RT or MaximaH-RT under conditions of long RNA fragment (4 min fragmentation) and short RNA fragment (9 min fragmentation). Figure 5B Even in libraries generated from short fragments, libraries generated after synthesizing cDNA on beads in a pull-down sample were significantly truncated compared to libraries freely generated from the same material in solution, confirming the significant enrichment of cross-linked truncated cDNA in pull-down libraries. Figure 5C ).

[0187] As expected, in libraries generated from total cellular RNA, >95% of all reads mapped to ribosomal RNA under all conditions. Analyzing the distribution of RT truncation sites (RT terminations), we observed significant differences between libraries generated using truncated RTs and those generated using pass-through RTs, as well as a significant impact of fragmentation time on the ability to distinguish library conditions. Figure 6A A library was generated using RNA isolated from UVC cross-linked cells and enriched via streptavidin bead pull-down (x400_PD) under the following conditions (R). 2 The correlation within approximately 0.9 indicates that ARORA is highly reproducible under all conditions. Figure 6B However, the three different library conditions exhibited significantly different abilities to distinguish x400_PD samples from the three control libraries: non-crosslinked, streptavidin-enriched (NCL_PD), and input (NCL_IN and x400_IN). Figure 6CCompared to libraries generated using AffinityScript, libraries generated using Maxima H- significantly reduced the correlation of RT termination between x400_PD and control libraries. These results are consistent with in vitro observations of cDNA generated using AffinityScript that was not significantly truncated at crosslinking sites. Furthermore, libraries generated using shorter RNA fragments (40-90 nt after 9 minutes of fragmentation) significantly reduced the correlation between x400_PD and control libraries compared to libraries generated from larger RNA fragments (80-250 nt after 4 minutes of fragmentation). Therefore, specific RT conditions and small, consistent RNA fragment sizes optimize ARORA's ability to distinguish conditions under which crosslinked PRIs can be identified by enriching RNA from crosslinked cells (x400_PD conditions) with streptavidin.

[0188] Next, we attempted to identify the ARORA of PRI (see...). Figure 1H The model was compared to existing methods. RNP-MaP was previously developed in the Weeks lab to identify PRI sites on single RNAs. 8 RNP-MaP utilizes a heterobifunctional cross-linking agent in which the amino-reactive linker covalently binds to the protein, while the diaziridine moiety cross-links with the nucleic acid upon exposure to UVA light. In RNP-MaP, PRI is identified by adduct-induced mutations, which are present at approximately 1 per 10 in the control sample. -3 The background rate occurs and increases to 2-5 per 10^10 protein-RNA cross-links in vivo. -3 The rate of occurrence. We compared the PRIs identified by ARORA and RNP-MaP in the RNA component of the ribonucleoprotein complex RNase MRP (RMRP). ARORA showed significant agreement at the precise nucleotides predicted as RNA-protein interaction sites defined by RNP-MaP ( Figure 1I , Figures 7A to 7B (P=0.0035, Fisher exact). We further analyzed these data by relaxing the predicted binding sites by one nucleotide in either direction to account for the fact that each method has a different nucleotide preference for crosslinking, and that RNP-MaP can crosslink nucleotides further away due to the 7 nm spacer arm in the heterobifunctional crosslinking agent. After relaxing the PRI annotation by a single nucleotide, we obtained a significantly increased consistency between the results (P=10). -5This confirms that ARORA largely reproduces the data obtained through RNP-MaP. Almost all PRIs identified by RNP-MaP were also identified by ARORA; however, ARORA identified PRIs other than those annotated by RNP-MaP. RNP-MaP requires PRIs to be solvent-accessible or within 7 nm of lysine residues for cross-linking to occur. We hypothesize that the sites uniquely identified by ARORA represent PRIs that either occur deep within the ribonucleoprotein complex, or at solvent-inaccessible sites that are still cross-linkable via UVC, or PRIs that do not contain lysine amino acids.

[0189] RNP-MaP requires 10 nucleotides per transcript to be analyzed. 4 ARORA achieves a statistically significant increase in the number of unique reads required to identify adduct-induced mutations that identify PRI. For example, approximately 2,500,000 unique reads are needed in RNP-MaP to identify PRI in a 268-nucleotide-long RMRP transcript. In contrast, ARORA utilizes cross-link-induced truncation, which occurs in approximately 50% of the reads, and deletions in approximately 7-8% of the reads, resulting in a much higher information density than RNP-MaP. Thus, ARORA achieves highly reproducible annotation of PRI in libraries with fewer than approximately 1200 total reads in RMRP. Figure 7C This translates to an efficiency gain of approximately 2,000-fold compared to RNP-MaP, a key gain that makes whole transcriptome annotation of PRI feasible.

[0190] We performed low-depth whole transcriptome sequencing on ARORA libraries in HepG2 and K562 cells to benchmark ARORA against other reference datasets, including eCLIP annotations from ENCODE and targeted PRI annotations identified by SHAPE footprint analysis. 10,13,14 We observed at least one region of significant ARORA enrichment (>30X enrichment) in most mRNAs, indicating that a subset of sequences within the mRNA (typically in the 3' or 5' UTR) is highly enriched for regulatory PRI. The length of the significantly enriched individual regions was typically 80–150 nucleotides, consistent with the 1–2X size of the read length of most reads. More interestingly, most of the significantly enriched regions were associated with very narrow regions truncated by reverse transcriptase (RT termination peaks), which can be clearly visualized in transcripts even with moderately low read depths if the RT termination peak appears in a narrow window. Figures 8A to 8BCompared to the input and NCL_PD samples, the RT termination peaks in the x400_PD samples were highly enriched and were generally discrete and well-defined, ranging in length from 3 to 15 nucleotides. This is strikingly similar to the size of the RNA motif that determines sequencing specificity PRI, which is typically in the range of 5–7 nucleotides in length. 11 We are currently conducting large-scale computational analysis on these motifs, but an investigation of thousands of individual RT termination peaks shows that each peak conforms to a known RBP motif.

[0191] Due to several technical limitations, eCLIP cannot achieve single-nucleotide resolution of RBP binding sites based on RT termination. Therefore, eCLIP analysis only identifies and reports significantly enriched sequences, i.e., RNA regions heavily enriched by eCLIP through pull-down of specific proteins. Consequently, they defined a window within 100 bases of the 5' end of the read to search for RNA motifs. Binding motifs were defined as significantly enriched motifs within a 100bp window of the RT termination site for each RBP. We compared the significantly enriched sequences defined by eCLIP with data obtained from ARORA. ARORA largely corresponds to the enriched domains defined by eCLIP. Here, dense RT termination peaks in ARORA can be observed at the 5' end of most RNA sequences highly enriched by eCLIP (>16X). Most mRNAs have a moderate number of RT termination peaks (<15) in their 3' UTR, while a significantly minority of mRNAs have significant RT termination peaks in their 5' UTR. ACTB mRNA has been well characterized by eCLIP, which identified eight highly enriched RBPs in the 3' UTR region, appearing in four distinct 100-150 nucleotide regions. ARORA identified 12 RT termination peaks in the 3' UTR of ACTB mRNA; all but one of these peaks were contained within or at the 5' edge of the significantly enriched region in eCLIP. Figure 8C Typically, significant RT termination peaks are not common in the 5' UTR of transcripts. A notable exception is selected mRNAs (i.e., VEGFA mRNAs) with known IRES elements within the 5' UTR. 18 And mRNAs with a 5' TOP domain (i.e., RPL30 mRNA). 19 It is particularly common in the mRNA of most proteins required for translation, including almost all RP and elongation factors. Figure 8DFinally, the transferrin receptor is a moderately abundant mRNA, extensively regulated in its 3' UTR by specialized RNA-binding proteins (eCLIP profiling) and iron-reactive proteins (not included in the eCLIP dataset but detected by SHAPE footprinting). ARORA identified numerous RT termination peaks within the 3' UTR of the TFRC, most of which corresponded to the typical RBP motif at the 5' end of the highly enriched eCLIP region. Figure 8E Additionally, ARORA reproduced two IRP footprints in the TFRC 3'UTR identified by SHAPE footprint analysis. In summary, a single experiment using ARORA extensively reproduced previous PRI footprint analyses annotated using three orthogonal techniques: RNP-MaP, eCLIP, and SHAPE footprint analysis.

[0192] Example 4: ARORA reveals constitutive and dynamic protein-RNA interactions, elucidating unique biological states.

[0193] Ribosomes are the largest single ribonucleoproteins in cells, central to protein synthesis, and involved in the most energy-intensive processes, accounting for approximately one-third of the energy consumption in mammalian cells. Ribosomes are generally considered homogeneous macromolecular machines, showing little variation across biological states. However, recent years have revealed a more complex picture, uncovering significant context-dependent differences in ribosome composition, regulation, and dynamics, which have implications for the pathogenesis of benign diseases, including ribosomal disorders, neurodegenerative diseases, and cancers. 20-23 The structure of ribosomes has been extensively studied using cryo-electron microscopy (cryo-EM), revealing the dynamic structural changes of ribosomes at baseline and during protein translation. 24 Identification of ribosome determinants of disease can reveal new insights into disease pathogenesis and may lead to new diagnostic and therapeutic approaches. However, the technical complexity and cost of cryo-EM have significantly limited its application and prevented extensive investigations of ribosome structure across biological conditions.

[0194] Our preliminary development of ARORA using total cellular RNA revealed robust, reproducible PRIs across three ribosomal RNAs encoded in the pre-45S rRNA transcripts (18S, 5.8S, and 28S). We compared the PRI locations identified by ARORA with those identified by cryo-EM. We observed a strong similarity between the PRI locations identified by ARORA and cryo-EM annotations between ribosomal proteins (RPs) and ribosomal RNAs (rRNAs). Figure 2A , Figure 9 Fisher's exact p < 10 -6 ) 25PRIs identified by cryo-EM are largely limited to core ribosome components, such as RPs. However, cryo-EM cannot identify PRIs of regulatory proteins that dynamically interact with the ribosome, due to transient PRIs and the heterogeneity of these PRIs within the detected ribosome particles. In contrast, the CLIP method has identified numerous regulatory protein interactions with rRNA that were not elucidated by cryo-EM. 26 For example, ribosome collisions induced by protein synthesis inhibitors (such as actinomycin, anisoxin, or homoharringtonine) induce stress signaling in p38 cells. 27 This signaling pathway is activated by the binding of ZAKα to the helix 14 of the colliding ribosome. We performed ARORA on normal diploid human fibroblasts, which identified PRI at the RP-rRNA interaction site defined by cryo-EM, but no evidence of PRI at the ZAKα binding site on the 18S helix 14. However, ARORA performed on the same cells 1 hour after exposure to actinomycin or anisoxin recruited significant PRI at the precise ZAKα binding site identified by CLIP, consistent with ARORA detection of a dynamic, stress-induced interaction between ZAKα and the 18S helix 14. Figure 2A ).

[0195] To evaluate the ability of ARORA to identify variability in ribosome structure and regulation across different biological conditions, we performed ARORA on a cohort of 14 widely studied and diverse breast cancer cell lines, including multiple representative cell lines from each of the three major breast cancer types: ER+, HER2+, and triple-negative breast cancer. At the sites of RP-rRNA interaction defined by cryo-EM, ARORA demonstrated high correlation and low variability across all samples. Figure 2B The low variability in ARORA reactivity is closely associated with known sites of RP-rRNA interaction (Fisher exact test, P < 10). -5 This indicates that PRI at RP-rRNA interaction regions exhibit less variability across cell lines than at non-RP-rRNA interaction sites, as expected. Furthermore, PRI within specific rRNA domains show very high correlations with each other, similar to topologically correlated domains identified using 3C, Hi-C, and correlation techniques for identifying functional units within chromatin. 28 These highly correlated rRNA domains tend to correspond to some secondary structures of rRNA, particularly to the extended region, which is the most evolutionarily unique rRNA sequence in human cells and the site of many regulatory PRIs that tune translation.

[0196] Regions of rRNA with low RP contact density (especially those in the 18S and 28S rRNA extension regions) exhibited high variability among breast cancer cell lines. This suggests that RP in these regions is highly variable among cell lines, particularly among regulatory or dynamic rRNA-binding proteins. Enrichment of helices 14, 16, and 18 on 18S rRNA regulates function (…). Figure 2B Helices 16 and 18 form at the gates of mRNA entry channels that regulate translation initiation, elongation, and termination, and are binding sites for many regulatory factors (including eIF, eEF, and eRF) and mRNA-interacting proteins (such as DDX3). Helix 14 is involved in signal transduction; ribosomal stress is the binding site for ZAKα in response to ribosomal collisions. 27 At each of these sites, highly variable PRI was observed among breast cancer cell lines. We found it particularly interesting that a subset of BRCA cell lines exhibited variability in PRI within the ribosomal stress response site on helix 14, which binds to ZAKα. The Cancer Cell Line Encyclopedia examined the sensitivity of cancer cell lines to a variety of clinical and preclinical small molecules, including sensitivity to the drug omaxitazine (the drug name for homoharringtonine), which inhibits translation via a mechanism similar to actinomycin or anisoxin and induces ribosomal collisions and ribosomal stress. 29 We observed that a subset of BRCA cell lines with high PRI detected in the ZAKα binding region of ARORA on helix 14 of the 18S axis were resistant to omaxitazine, while BRCA cell lines with low PRI in this region were significantly more sensitive to the drug. Figures 2C to 2D We hypothesize that high basal PRI levels within the ZAKα binding region on helix 14 identify cell lines with high basal ribosomal stress that have adapted to and overcome this stress, leading to resistance to drugs that induce ribosomal stress. Therefore, ARORA identifies biologically relevant differences in ribosomal regulation and structure, with clear potential for direct clinical relevance in identifying therapeutic vulnerabilities in cancer cells.

[0197] Example 5: ARORA identifies cancer driver mutations in non-coding regulatory RNAs

[0198] Non-protein coding genomes undergo frequent mutations and alterations in human cancers, contributing to the pathogenesis of the disease. 30-32However, we currently lack tools to readily determine whether non-coding mutations are drivers of cancer phenotypes or merely passenger mutations. eCLIP has previously demonstrated the ability to identify allele-specific PRIs for individual RBPs, although conditions are not well-suited to single-nucleotide resolution of PRIs. We examined whether ARORA could identify mutations in regulatory RNAs that alter RBP interactions. When considering how sequencing an ARORA library at single-nucleotide resolution of PRIs could identify mutations altering binding, it is clear that the position of the PRI relative to the mutation site will have a considerable impact on the interpretation of the sequencing data. Figure 3A When using RT-enzymes that effectively truncate at the crosslinking site and identify PRIs at single-nucleotide resolution (such as Maxima H-), if the major crosslinking nucleotide within the PRI is at the 5' of the RNA sequence altering the binding single-nucleotide variant, the variant nucleotide can be easily quantified, and allele enrichment or depletion under x400_PD conditions can be clearly calculated to determine whether the allele is favored or unfavored in the PRI, respectively. However, if the major crosslinking nucleotide within the PRI is at the 3' of the altering binding single-nucleotide variant, the RT-enzyme will not read through the variant nucleotide because the cDNA is truncated before reaching the variant site. In this case, the representation of the preferred variant in binding is depleted in the ARORA sequencing library generated with the truncated RT-enzyme. In contrast, ARORA sequencing libraries generated with a readthrough RT-enzyme (such as AffinityScript) lose the single-nucleotide resolution of the PRI, but the ability to read through the crosslinking site can fully capture the variant sequence and quantify allele enrichment or depletion at the PRI.

[0199] We examined low-depth whole transcriptome ARORA datasets from HepG2 and K562 to determine whether ARORA identifies allele-specific PRI. Various non-pathogenic normal variants, such as GAPDH rs1065691, resulted in no change in allele fractions across the four conditions, confirming that ARORA does not select and is consistent with the fact that these normal variants have no effect on RBP-mRNA interaction. Figure 3B In contrast, a HepG2 cell-specific single nucleotide variant was observed in the 5'UTR of the oncogene MYC. This variant transcript (with a C184T residue at residue 184) is different from the wild-type transcript. The frequency of T transitions was significantly altered by cross-linking and pull-down, suggesting potentially allele-specific PRI (T transitions). Figure 3BIn UVC crosslinked and pull-down samples, the C184T variant was reduced in total reads. Although the sequencing depth in the MYC 5'UTR was insufficient to achieve single nucleotide resolution, ARORA sequencing still demonstrated significant enrichment (approximately 7X) of the RT termination only at the 3' nucleotide immediately adjacent to C184T, a feature observed in HepG2 samples but not in K562 samples. We performed ARORA quantification using qPCR of this region to precisely quantify its enrichment, confirming that HepG2 cells possess a unique and significant PRI not present in K562 or the immortalized, highly proliferating 293T cell line. Figure 3C In summary, these data indicate that HepG2 cells contain a unique MYC 5'UTR variant associated with allele-specific binding and the acquisition of PRI, which is not present in other cells.

[0200] Although enrichment at the 3' RT termination of C184T and the deletion of C184T in the total read indicate that C184T has acquired RBP binding compared to the wild-type transcript, we performed ARORA in HepG2 using read-through RT AffinityScript to quantify the relative proportion of MYC alleles recovered by ARORA. Under these conditions, ARORA enrichment in HepG2 of sequences overlapping with the C184T mutation further confirms significant PRI occurring in this region. In the ENCODE eCLIP database, of the 103 RBPs investigated in HepG2 cells, only two RBPs (DDX3X and DDX6, both involved in translation initiation) were strongly enriched at the precise location identified by ARORA at the MYC 5' UTR ( Figure 3D This suggests that ARORA may be detecting the binding of translational regulators. Furthermore, ARORA with crosslink site readout revealed that PRI in the MYC 5'UTR of these cells is indeed allele-specific, as 85% of the MYC 5'UTR fragments recovered by ARORA contained C184T, although this represented 40% of all MYC transcripts in the input sample. This indicates that potential cis-regulators in the MYC 5'UTR primarily interact with MYC transcripts carrying 5'UTR C184T. Figures 3E to 3F In summary, ARORA discovered a variant in the 5'UTR of the oncogene MYC that alters the interaction between the MYC 5'UTR and RBP (possibly DDX3X or DDX6), a translation regulator that enhances the translation efficiency of a subset of mRNAs through interaction with the 5'UTR.

[0201] Finally, these data demonstrate ARORA's outstanding potential for clinical diagnostics. We have observed that ARORA performance is preserved in samples derived from frozen cell line "blocks" (simulated frozen tissue), with the ability to annotate PRIs that are highly correlated with results obtained with ARORA in the same cell lines in cell culture. Figure 10 Therefore, ARORA can be performed on frozen clinical samples that are standard for molecular and genetic diagnostic testing.

[0202] Example 6: Identification of RBPs that bind to specific sites on RNA via ARORA

[0203] As described in this paper, ARORA comprehensively identifies specific sites on RNA molecules that bind to RBPs at single-nucleotide resolution. However, the specific identity of the RBP binding at any given site was previously unknown. While eCLIP, RNA bind-n-seq, and other techniques have mapped nucleotide sequence motifs preferentially bound to specific RBPs, many sequence motifs bind to different RBPs, limiting the ability of individual sequence motifs to predict the precise RBP binding at a given site on RNA. Analytical methods, including mCross and PEKA, have recently provided single-nucleotide resolution for RBP crosslinking patterns within sequence motifs derived from eCLIP. These analyses have identified many individual RBPs with typical crosslinking patterns to preferred sequence motifs. mCross analysis demonstrates reproducible, unique crosslinking patterns of individual RBPs within a given sequence motif, many of which are RBP-specific and distinct from other RBPs binding to the same or similar sequence motifs. Therefore, even without prior knowledge of the protein / peptide binding to the RNA site, crosslinking patterns within sequence motifs can identify the exact RBP binding to a specific site on RNA.

[0204] Applying the cross-linking patterns identified by mCross analysis to sequencing data from ARORA, we found that cross-linking patterns within sequencing motifs in ARORA allow for the identification of RBPs that bind to specific sites on RNA. Analyzing the mCross cross-linking patterns of three RBPs (PUM2, TARDBP, and YBX3), we found that subsets of RNA transcripts containing sequence motifs specific to each RBP exhibited typical cross-linking patterns within the sequence motifs. Figures 16A to 16CSites exhibiting RBP-specific crosslinking patterns within sequence motifs correspond to sites highly enriched with RBP in eCLIP data, distinguishing these sites from others containing RBP-preferred motifs but not effectively bound by RBPs. mCross has identified 39 high-confidence crosslinking patterns for RBPs, as well as many more low-confidence crosslinking patterns. Therefore, ARORA's annotation of crosslinking patterns based on sequence motifs and the application of experimentally defined RBP-specific crosslinking patterns allows for the simultaneous annotation of sites specifically bound by numerous RBPs in a single assay. Furthermore, by identifying crosslinking patterns within sequence motifs lost after RBP-induced loss in cell lines, ARORA can de novo discover crosslinking patterns for RBPs not defined by mCross. Since this does not require the availability of high-quality immunoprecipitation antibodies, it presents a significant advantage over CLIP techniques, which are limited by antibody specificity and quality.

[0205] The ability to precisely identify RBPs that bind to specific sites on RNA has further facilitated new insights into RNA regulation. Current methods, such as CLIP, perform profiling on individual RBPs, thus preventing direct comparisons between the strength or stoichiometry of binding between different RBPs interacting with RNA. In contrast, ARORA identifies the relative strength and stoichiometry of RBP-RNA interactions on individual transcripts. For example, eCLIP identified dozens of RBPs binding to GAPDH mRNA at different locations throughout the transcript. ARORA reproduces these binding sites defined by eCLIP while also annotating numerous additional binding sites ( Figure 16B Furthermore, ARORA demonstrated that the RBP binding sites on GAPDH mRNA are overwhelmingly enriched for YBX3 interactions, while the enrichment of other RBP interactions is several orders of magnitude lower than that of YBX3 sites. Figure 16BAs cold shock proteins, the YBX family of proteins plays a crucial role in maintaining glucose metabolism during cellular stress, and the loss of a single YBX family member leads to the downregulation of glucose metabolism-related mRNAs, such as GAPDH and other mRNAs of glycolysis pathway components. The highly enriched YBX3 site is the strongest RBP binding site on the mRNAs of most proteins involved in glycolysis, demonstrating that YBX is a major regulator of this key pathway and identifying potential sites relevant to therapeutic targeting in cancer and other diseases. Compared to glycolysis genes, most ribosomal protein mRNAs have very few YBX3 binding sites enriched. Furthermore, ARORA found that almost all ribosomal protein mRNAs exhibit dominant enrichment of RBP binding at the 5' terminal oligopyrimidine bundle (5'TOP) site, a conserved feature in the 5' UTR of these mRNAs, which is essential for regulating ribosomal protein translation. Therefore, ARORA's identification of the strongest binding sites in RNA annotates the most critical RBP-RNA interactions for regulating mRNA function, and thus identifies sites most relevant to cellular mechanisms, sites with information for disease diagnosis and disease mechanism discovery, and sites for therapeutic target discovery.

[0206] We developed ARORA to address a major unmet need for identifying novel druggable targets for cancer therapies. We developed a targeted, rapid, and cost-effective technique for the comprehensive identification of protein-protein interaction sites on RNA. This method utilizes a biochemical process to convert specific nucleotides that bind to RBPs into orthogonal chemical moieties conjugated to ligands for affinity purification. This produces sequencing libraries in which the precise nucleotides contacting the RBP are precisely encoded in each fragment of the library and resolved via next-generation sequencing. Crucially, this method allows purified RNA to have the same solubility and processing properties as unmodified cellular RNA, making the process highly suitable for automation and high-throughput procedures.

[0207] ARORA represents a significant technological advancement over existing methods for identifying RBP-RNA interaction sites in several respects: (1) the use of a universal chemical motif (α-amino or carboxyl) in amino acids makes it possible to crosslink any protein to RNA with UVC, regardless of amino acid identity, allowing for whole transcriptome analysis of all RNA-binding proteins in a single assay; (2) the chemistry is fully compatible with standard high-throughput and automated library platforms, and RNA in ARORA can be processed via hybridization capture to selectively deplete or selectively enrich transcripts for specific transcripts analyzed by ARORA; (3) it allows for comprehensive, whole transcriptome investigations to identify dynamic, stress-responsive, and allele-specific PRIs that contribute to disease, increasing the potential to identify RBP-RNA interactions that can be used for diagnostic or therapeutic development; (4) ARORA provides direct, quantitative, chemometric comparisons of PRIs on transcripts and between biological conditions, compared to binary (bound vs. unbound) results provided by current methods; and (5) PRI annotation efficiency is improved by several orders of magnitude compared to existing methods.

[0208] Example 7: ARORA-based drug screening method

[0209] Several small molecule and oligonucleotide drugs are currently being developed to specifically inhibit single RBPs or RBP families involved in specific disease processes. However, previous reports have shown that drugs thought to specifically inhibit single RBPs actually have broad effects on RBP-RNA interactions and exhibit numerous off-target effects (Walters et al., RNA. Oct 2023; 29(10): 1458–1470). ARORA, as a assay that comprehensively annotates the interactions between all RBPs and all RNA transcripts, could in principle identify the efficacy and specificity of specific RBP inhibitor drugs. We will use ARORA to determine the efficacy and specificity of RBP inhibitors (e.g., oligonucleotide inhibitors and small molecule inhibitors).

[0210] Selected individual RBPs will be inhibited by an antisense oligonucleotide (ASO) or a small molecule inhibitor that supposedly inhibits the activity of the same RBP, which reduces the mRNA targeting the RBP and thus aims to reduce RBP abundance in cells. Following treatment with an inhibitory ASO, ARORA will be performed in cells and compared to an untreated control. ARORA will also be performed under additional conditions where the efficacy of a small molecule drug with a alleged mechanism of action that inhibits the same RBP will be detected. Efficacy will be defined by the fraction of RBP-specific PRIs lost upon treatment with the small molecule inhibitor, as identified by crosslinking patterns and sequence motifs. Specificity will be determined by defining the fraction of PRIs that are not specific to the RBP target, which are altered upon treatment with the same small molecule inhibitor. Therefore, the effects of RBP inhibitory drugs will be comprehensively investigated in a single ARORA assay.

[0211] Example 8: Identification of RBPs binding to specific sites on RNAs associated with aplastic aplasia (AD)

[0212] AD is a congenital osteochondrodysplasia. Individuals affected by this disease have skeletal and cartilage dysplasia, leading to characteristics such as dwarfism and hypermobility of the joints. 38 Mutations in POP1, a gene encoding essential RNA-binding protein components of two different but related human RNases (RNase MRP and RNase P), are associated with AD. 39 POP1 directly binds to RMRP (the RNA component of the RNase MRP ribonucleoprotein complex) to form the functional RNase MRP. 40 Individually, POP1 binds to RPPH1 RNA to form a functional ribonucleoprotein complex of RNase P. ARORA identified strong, indistinct protein-binding sites on both RMRP and RPPH1 RNA that correspond to well-described binding sites of POP1 on the RNA components of RNases MRP and P. Genetic deletion of functional POP1 prevents its incorporation into these ribonucleoprotein complexes. AD cell models with POP1 deficiency are expected to show a lack of protein-RNA interactions at POP1 binding sites on RMRP and RPPH1 RNA, as detected by ARORA.

[0213] ARORA will be performed in two cell lines as described in Example 1: (1) wild-type human cells with POP1 and (2) POP1 mutant cells serving as an AD model. In (1) wild-type human cells with POP1 and (2) POP1 mutant cells serving as an AD model, the stoichiometry of protein-RNA interactions at POP1 binding sites on RMRP and RPPH1 RNA will be quantified and compared with non-POP1 protein-RNA interactions on these RNAs. It is expected that the stoichiometry of protein-RNA interactions at POP1 binding sites will be reduced in the POP1 mutant cell line compared to wild-type cells, while protein-RNA interactions at non-POP1 protein-RNA interaction sites will be preserved. Therefore, ARORA can serve as a sensitive and specific measure of POP1 protein deficiency in the assembly of RNases MRP and P, which are associated with AD pathogenesis.

[0214] Example 9: RBPs binding to specific sites on RNA associated with congenital pure red cell aplasia (DBA) Identification

[0215] DBA is a form of ribosomalopathy, a rare autosomal dominant inherited disorder caused by pathogenic loss-of-function mutations in one of the many ribosomal proteins (RPs) that are key structural components of the ribosome. 41 DBA is commonly associated with bone marrow dysfunction, severely disrupting erythrocyte production and leading to potentially fatal anemia. While mutations in several genes containing ribosomal proteins are associated with DBA, mutations in RPL5 and RPL11 are among the most common causes. Loss-of-function mutations in RP genes lead to stoichiometry imbalances in RP, resulting in impaired or incomplete ribosome assembly in ribosomal biomolecules, lacking RP components. Consequently, stress-induced cellular stress responses occur in ribosomal biomolecules, particularly p53 signaling, triggering cell cycle arrest or inducing programmed cell death. These cellular responses lead to hematopoietic dysfunction and other tissue disorders. ARORA demonstrated the ability to identify RP binding sites on ribosomal RNA (rRNA), identifying conserved protein binding sites at the contact sites of RPL5 and RPL11 with the large subunit (28S) of rRNA. ARORA is expected to identify the loss of protein-rRNA contacts at the RPL5-28S rRNA contact or RPL11-28S rRNA contact sites in DBA cell models, where RPL5 or RPL11 is a gene defect.

[0216] ARORA was performed in three cell lines as described in Example 1: (1) wild-type human cells, (2) RPL5 mutant cells as a DBA model, and (3) RPL11 mutant cells as a DBA model. The stoichiometry of protein-RNA interactions at the RPL5 or RPL11 binding sites on 28S rRNA was quantified and compared with protein-RNA interactions of other non-RPL5 / RPL11 proteins binding to 28S rRNA in the three cell lines: (1) wild-type human cells, (2) RPL5 mutant cells as a DBA model, and (3) RPL11 mutant cells as a DBA model. It was expected that, compared to wild-type cells, the stoichiometry of protein-RNA interactions at the RPL5 or RPL11 binding sites could be reduced in the RPL5 or RPL11 mutant cell lines, while protein-RNA interactions and other non-RPL5 / RPL11 binding site protein-RNA interactions could be preserved. Therefore, ARORA can serve as a sensitive and specific measure of the loss of RPL5 or RPL11 proteins in the assembly of the large ribosomal subunit (a pathogenic mechanism of DBA and other ribosomal diseases).

[0217] Example 10: Using ARORA to evaluate the efficacy of small molecules

[0218] Translation initiation in eukaryotic cells requires the assembly of the elongation initiation factor complex 4F (eIF4F), which recruits small ribosomal subunits to the 5' cap of the mRNA before the formation of the pre-assembled initiation complex. 42 eIF4F consists of three subunits: eIF4E, an eIF4G scaffold protein that recognizes the 5' cap, and the RNA helicase eIF4A. eIF4A helicase activity is essential for removing secondary structures from the 5'UTR, a step that facilitates the translocation of the pre-initiation complex along the 5'UTR to the start codon. Many oncogenes contain 5'UTRs with highly structured secondary structures and therefore depend on eIF4A for efficient translation. Consequently, cancer cell growth often relies on eIF4A helicase activity to maintain oncogenic signaling, making eIF4A an attractive therapeutic target for oncology drug development. Rocaglates (such as silvestrol), natural compounds that inhibit translation by modulating the binding of eIF4A to mRNA, represent a class of molecules of high interest as therapeutic agents for oncology and viral infections. 43, 44 Zotatifil, a rocargage derivative, has recently shown promising results in an initial phase I / II clinical trial for breast cancer. 45Some rocargages (such as zotatifil) stabilize the interaction of eIF4A with purine-rich (AG) sequences in the 5'UTR of selected mRNAs, resulting in effective steric blocking, preventing the translocation of the pre-initiation complex along the 5'UTR, and inhibiting translation. 46 ARORA is expected to be able to determine the efficacy and selectivity of certain rocalamine esters (such as zotatifil) in stabilizing and blocking translation of eIF4A-mRNA in the 5'UTR.

[0219] We will perform ARORA under two cellular conditions as described in Example 1: (1) wild-type human cells treated with DMSO medulant only and (2) the same human cell lines treated with rocargagerone (zotatifil or related molecules) in DMSO for 2–6 hours. Under both conditions, the stoichiometry of protein-RNA interactions at the eIF4A binding site in the 5'UTR of mRNA will be quantified and compared with protein-RNA interactions at sites in the 3'UTR that are largely independent of eIF4A. It is expected that the stoichiometry of protein-RNA interactions at the eIF4A binding site in the purine-rich (AG) sequence of the 5'UTR in cells may increase 2–6 hours after treatment with rocargagerone (zotatifil or related molecules) compared to cells treated with DMSO only. In contrast, protein-RNA interactions in the 3'UTR of mRNA may be preserved, which are largely independent of eIF4A. Therefore, ARORA can serve as a sensitive quantitative method for stabilizing eIF4A-mRNA interactions and translational repression in selected mRNAs with highly structured 5'UTRs.

[0220] equivalent

[0221] The present invention is not limited to the specific embodiments described herein, which are intended to be shown as individual aspects of the invention. Many modifications and variations can be made to the invention without departing from its spirit and scope, as will be apparent to those skilled in the art. Based on the foregoing description, functionally equivalent methods and apparatuses within the scope of the invention will be apparent to those skilled in the art, in addition to those methods and apparatuses listed herein. Such modifications and variations are intended to fall within the scope of the invention. It should be understood that the present invention is not limited to specific methods, reagents, compounds, compositions, or biological systems, although such specific methods, reagents, compounds, compositions, or biological systems can vary. It should also be understood that the terminology used herein is for the purpose of describing specific embodiments only and is not intended to be limiting.

[0222] Furthermore, when features or aspects of this disclosure are described in accordance with the Markush Group, those skilled in the art will recognize that this disclosure is also described in accordance with any individual member or subgroup of the Markush Group.

[0223] As those skilled in the art will understand, for any and all purposes, particularly for the purpose of providing a written description, all scopes disclosed herein also encompass any and all possible subscopes and combinations thereof. Any listed scope can be readily considered sufficiently descriptive and such that the same scope can be decomposed into at least equal halves, thirds, quarters, fifths, tenths, etc. As a non-limiting example, each scope discussed herein can be readily divided into a lower third, a middle third, and an upper third, etc. As those skilled in the art will also understand, all language, such as “up to,” “at least,” “greater than,” “less than,” etc., includes the listed numbers and refers to a scope that can subsequently be decomposed into subscopes as described above. Finally, as those skilled in the art will understand, a scope includes each individual member. Thus, for example, a group having 1-3 cells means a group having 1, 2, or 3 cells. Similarly, a group having 1-5 cells means a group having 1, 2, 3, 4, or 5 cells, and so on.

[0224] All patents, patent applications, provisional applications and publications mentioned or cited herein are incorporated herein in their entirety by reference to the extent that they are not inconsistent with the explicit teachings of this specification, including all figures and tables.

[0225] References

[0226] 1Investigators, GPP et al. 100,000 Genomes Pilot on Rare-DiseaseDiagnosis in Health Care - Preliminary Report. The New England journal of medicine 385, 1868-1880, doi:10.1056 / NEJMoa2035790 (2021).

[0227] 2Anastasiadou, E., Jacob, LS & Slack, FJ Non-coding RNAnetworks in cancer. Nature reviews. Cancer 18, 5-18, doi:10.1038 / nrc.2017.99(2018).

[0228] 3Gebauer, F., Schwarzl, T., Valcarcel, J. & Hentze, M. W. RNA-bindingproteins in human genetic disease. Nature reviews. Genetics 22, 185-198, doi:10.1038 / s41576-020-00302-y (2021).

[0229] 4Mitschka, S. & Mayr, C. Context-specific regulation and function ofmRNA alternative polyadenylation. Nature reviews. Molecular cell biology 23,779-796, doi:10.1038 / s41580-022-00507-5 (2022).

[0230] 5Ramanathan, M., Porter, D. F. & Khavari, P. A. Methods to study RNA-protein interactions. Nature methods 16, 225-234, doi:10.1038 / s41592-019-0330-1 (2019).

[0231] 6Hafner, M. et al. CLIP and complementary methods. Nat Rev MethodPrime 1, doi:ARTN 20

[0232] 10.1038 / s43586-021-00018-1 (2021).

[0233] 7Queiroz, R. M. L. et al. Comprehensive identification of RNA-proteininteractions in any organism using orthogonal organic phase separation(OOPS). Nature biotechnology 37, 169-178, doi:10.1038 / s41587-018-0001-2(2019).

[0234] 8Weidmann, C. A., Mustoe, A. M., Jariwala, P. B., Calabrese, J. M. &Weeks, K. M. Analysis of RNA-protein networks with RNP-MaP defines functionalhubs on RNA. Nature biotechnology 39, 347-356, doi:10.1038 / s41587-020-0709-7(2021).

[0235] 9Grawe, C., Stelloo, S., van Hout, F. A. H. & Vermeulen, M. RNA-Centric Methods: Toward the Interactome of Specific RNA Transcripts. TrendsBiotechnol 39, 890-900, doi:10.1016 / j.tibtech.2020.11.011 (2021).

[0236] 10Corley, M. et al. Footprinting SHAPE-eCLIP Reveals Transcriptome-wide Hydrogen Bonds at RNA-Protein Interfaces. Molecular cell 80, 903-914e908, doi:10.1016 / j.molcel.2020.11.014 (2020).

[0237] 11Dominguez, D. et al. Sequence, Structure, and Context Preferencesof Human RNA Binding Proteins. Molecular cell 70, 854-867 e859, doi:10.1016 / j.molcel.2018.05.001 (2018).

[0238] 12Ramanathan, M. et al. RNA-protein interaction detection in livingcells. Nature methods 15, 207-212, doi:10.1038 / nmeth.4601 (2018).

[0239] 13Van Nostrand, E. L. et al. A large-scale binding and functional mapof human RNA-binding proteins. Nature 583, 711-719, doi:10.1038 / s41586-020-2077-3 (2020).

[0240] 14Van Nostrand, E. L. et al. Robust transcriptome-wide discovery ofRNA-binding protein binding sites with enhanced CLIP (eCLIP). Nature methods13, 508-514, doi:10.1038 / nmeth.3810 (2016).

[0241] 15Urdaneta, E. C. et al. Purification of cross-linked RNA-proteincomplexes by phenol-toluol extraction. Nature communications 10, 990, doi:10.1038 / s41467-019-08942-3 (2019).

[0242] 16Van Nostrand, E. L., Shishkin, A. A., Pratt, G. A., Nguyen, T. B. &Yeo, G. W. Variation in single-nucleotide sensitivity of eCLIP derived fromreverse transcription conditions. Methods 126, 29-37, doi:10.1016 / j.ymeth.2017.08.002 (2017).

[0243] 17Zarnegar, B. J. et al. irCLIP platform for efficientcharacterization of protein-RNA interactions. Nature methods 13, 489-492,doi:10.1038 / nmeth.3840 (2016).

[0244] 18Leppek, K., Das, R. & Barna, M. Functional 5' UTR mRNA structuresin eukaryotic translation regulation and how to find them. Nature reviews.Molecular cell biology 19, 158-174, doi:10.1038 / nrm.2017.103 (2018).

[0245] 19Hsieh, A. C. et al. The translational landscape of mTOR signallingsteers cancer initiation and metastasis. Nature 485, 55-61, doi:10.1038 / nature10912 (2012).

[0246] 20Leppek, K. et al. Gene- and Species-Specific Hox mRNA Translationby Ribosome Expansion Segments. Molecular cell 80, 980-995 e913, doi:10.1016 / j.molcel.2020.10.023 (2020).

[0247] 21Simsek, D. et al. The Mammalian Ribo-interactome Reveals RibosomeFunctional Diversity and Heterogeneity. Cell 169, 1051-1065 e1018, doi:10.1016 / j.cell.2017.05.022 (2017).

[0248] 22Ebright, R. Y. et al. Deregulation of ribosomal protein expressionand translation promotes breast cancer metastasis. Science 367, 1468-1473,doi:10.1126 / science.aay0939 (2020).

[0249] 23Parks, M. M. et al. Variant ribosomal RNA alleles are conserved andexhibit tissue-specific expression. Sci Adv 4, eaao0665, doi:10.1126 / sciadv.aao0665 (2018).

[0250] 24Behrmann, E. et al. Structural snapshots of actively translatinghuman ribosomes. Cell 161, 845-857, doi:10.1016 / j.cell.2015.03.052 (2015).

[0251] 25Bernier, C. R. et al. RiboVision suite for visualization andanalysis of ribosomes. Faraday Discuss 169, 195-207, doi:10.1039 / c3fd00126a(2014).

[0252] 26Van Nostrand, E. L. et al. Principles of RNA processing fromanalysis of enhanced CLIP maps for 150 RNA binding proteins. Genome biology21, 90, doi:10.1186 / s13059-020-01982-9 (2020).

[0253] 27Vind, A. C. et al. ZAKalpha Recognizes Stalled Ribosomes throughPartially Redundant Sensor Domains. Molecular cell 78, 700-713 e707, doi:10.1016 / j.molcel.2020.03.021 (2020).

[0254] 28Akgol Oksuz, B. et al. Systematic evaluation of chromosomeconformation capture assays. Nature methods 18, 1046-1055, doi:10.1038 / s41592-021-01248-7 (2021).

[0255] 29Seashore-Ludlow, B. et al. Harnessing Connectivity in a Large-ScaleSmall-Molecule Sensitivity Dataset. Cancer discovery 5, 1210-1223, doi:10.1158 / 2159-8290.CD-15-0235 (2015).

[0256] 30Weinhold, N., Jacobsen, A., Schultz, N., Sander, C. & Lee, W.Genome-wide analysis of noncoding regulatory mutations in cancer. Naturegenetics 46, 1160-1165, doi:10.1038 / ng.3101 (2014).

[0257] 31Li, K. et al. Noncoding Variants Connect Enhancer Dysregulationwith Nuclear Receptor Signaling in Hematopoietic Malignancies. Cancerdiscovery 10, 724-745, doi:10.1158 / 2159-8290.CD-19-1128 (2020).

[0258] 32Schmitt, A. M. & Chang, H. Y. Long Noncoding RNAs in CancerPathways. Cancer cell 29, 452-463, doi:10.1016 / j.ccell.2016.03.010 (2016).

[0259] 33Guo, Z. et al. A Functional 5'-UTR Polymorphism of MYC Contributesto Nasopharyngeal Carcinoma Susceptibility and Chemoradiotherapy InducedToxicities. J Cancer 10, 147-155, doi:10.7150 / jca.28534 (2019).

[0260] 34Xu-Monette, Z. Y. et al. Clinical and Biologic Significance of MYCGenetic Mutations in De Novo Diffuse Large B-cell Lymphoma. Clin Cancer Res22, 3593-3605, doi:10.1158 / 1078-0432.CCR-15-2296 (2016).

[0261] 35Chappell, S. A. et al. A mutation in the c-myc-IRES leads toenhanced internal ribosome entry in multiple myeloma: a novel mechanism ofoncogene de-regulation. Oncogene 19, 4437-4440, doi:10.1038 / sj.onc.1203791(2000).

[0262] 36Cobbold, L. C. et al. Upregulated c-myc expression in multiplemyeloma by internal ribosome entry results from increased interactions withand expression of PTB-1 and YB-1. Oncogene 29, 2884-2891, doi:10.1038 / onc.2010.31 (2010).

[0263] 37.Makitie O, Vakkilainen S. Cartilage-Hair Hypoplasia - AnauxeticDysplasia Spectrum Disorders. In: Adam MP, Feldman J, Mirzaa GM, Pagon RA,Wallace SE, Amemiya A, editors. GeneReviews((R)). Seattle (WA)1993.

[0264] 38.Barraza-Garcia J, Rivera-Pedroza CI, Hisado-Oliva A, Belinchon-Martinez A, Sentchordi-Montane L, Duncan EL, Clark GR, Del Pozo A, Ibanez-Garikano K, Offiah A, Prieto-Matos P, Cormier-Daire V, Heath KE. Broadeningthe phenotypic spectrum of POP1-skeletal dysplasias: identification of POP1mutations in a mild and severe skeletal dysplasia. Clin Genet. 2017;92(1):91-8. Epub 20170222. doi: 10.1111 / cge.12964. PubMed PMID: 28067412.

[0265] 39.Weidmann CA, Mustoe AM, Jariwala PB, Calabrese JM, Weeks KM.Analysis of RNA-protein networks with RNP-MaP defines functional hubs on RNA.Nature biotechnology. 2021;39(3):347-56. Epub 2020 / 10 / 21. doi: 10.1038 / s41587-020-0709-7. PubMed PMID: 33077962; PMCID: PMC7956044.

[0266] 40.Kampen KR, Sulima SO, Vereecke S, De Keersmaecker K. Hallmarks ofribosomopathies. Nucleic acids research. 2020;48(3):1013-28. doi: 10.1093 / nar / gkz637. PubMed PMID: 31350888; PMCID: PMC7026650.

[0267] 41.Brito Querido J, Diaz-Lopez I, Ramakrishnan V. The molecular basisof translation initiation and its regulation in eukaryotes. Nature reviewsMolecular cell biology. 2024;25(3):168-86. Epub 20231205. doi: 10.1038 / s41580-023-00624-9. PubMed PMID: 38052923.

[0268] 42.Chu J, Zhang W, Cencic R, O'Connor PBF, Robert F, Devine WG,Selznick A, Henkel T, Merrick WC, Brown LE, Baranov PV, Porco JA, Jr.,Pelletier J. Rocaglates Induce Gain-of-Function Alterations to eIF4A andeIF4F. Cell reports. 2020;30(8):2481-8 e5. doi: 10.1016 / j.celrep.2020.02.002.PubMed PMID: 32101697; PMCID: PMC7077502.

[0269] 43.Schmidt T, Dabrowska A, Waldron JA, Hodge K, Koulouras G,Gabrielsen M, Munro J, Tack DC, Harris G, McGhee E, Scott D, Carlin LM, HuangD, Le Quesne J, Zanivan S, Wilczynska A, Bushell M. eIF4A1-dependent mRNAsemploy purine-rich 5'UTR sequences to activate localised eIF4A1-unwindingthrough eIF4A1-multimerisation to facilitate translation. Nucleic acidsresearch. 2023;51(4):1859-79. doi: 10.1093 / nar / gkad030. PubMed PMID:36727461; PMCID: PMC9976904.

[0270] 44.Ezra Rosen MS, David Berz, Jennifer Lee Caswell-Jin, Alexander I.Spira, Georgina A. Fulgar, Mark Densel, Nawaid Rana, Samuel Sperry, DouglasWarner, and Funda Meric-Bernstam ASCO Authors' Group. Phase 1 / 2 doseexpansion study evaluating first-in-class eIF4A inhibitor zotatifin inpatients with ER+ metastatic breast cancer 2023.

[0271] 45.Gerson-Gurwitz A, Young NP, Goel VK, Eam B, Stumpf CR, Chen J,Fish S, Barrera M, Sung E, Staunton J, Chiang GG, Webster KR, Thompson PA.Zotatifin, an eIF4A-Selective Inhibitor, Blocks Tumor Growth in ReceptorTyrosine Kinase Driven Tumors. Front Oncol. 2021;11:766298. Epub 20211124.doi: 10.3389 / fonc.2021.766298. PubMed PMID: 34900714; PMCID: PMC8663026.

Claims

1. A method for identifying one or more RNA-binding protein (RBP)-RNA interaction (PRI) sites in at least one transcript derived from a biological sample comprising multiple cells, the method comprising: (a) Crosslinking the plurality of cells present in the biological sample with short-wavelength ultraviolet light to stabilize a plurality of RBP-RNA complexes containing RNA transcripts, wherein each RBP-RNA complex corresponds to PRI; (b) Isolating RNA molecules from a plurality of cross-linked cells, wherein the RNA molecules comprise unbound RNA molecules and the RNA transcripts of the plurality of RBP-RNA complexes; (c) Under conditions of cleaving the RBPs of the plurality of RBP-RNA complexes, the isolated RNA molecules are contacted with a protease to produce RNA transcripts bound by residual peptides, wherein each bound residual peptide (i) corresponds to a PRI site and (ii) contains an α-amino group at its N-terminus and a carboxyl group at its C-terminus. (d) The RNA transcript bound by the residual peptide is labeled by coupling the α-amino group or the carboxyl group of each bound residual peptide to a chemically conjugated compound containing an affinity ligand; (e) The RNA molecule is fragmented in the presence of heat and divalent cations to generate RNA fragments having 5' ends and 3' ends, wherein the RNA fragments comprise unlabeled RNA fragments and RNA fragments labeled with the chemical conjugate containing the affinity ligand; (f) The sequencing adaptor is ligated to the 3' end of the RNA fragment to generate an RNA fragment with the adaptor tag; (g) Using affinity purification to capture RNA fragments tagged with adaptor tags and labeled with the chemical conjugate containing the affinity ligand; (h) Reverse transcription of the labeled RNA fragment with the adaptor tag to generate multiple cDNA molecules, wherein the labeled RNA fragment with the adaptor tag is optionally attached to a solid surface or in solution; (i) The sequencing adaptor is ligated to the 3' end of the cDNA molecule to generate multiple cDNA molecules with the adaptor tag; (j) Amplify the plurality of cDNA molecules with adaptor tags to generate a nucleic acid library of transcripts with RNA-RBP interactions; (k) Sequencing and aligning the plurality of cDNA molecules with adaptor tags in the nucleic acid library to identify individual transcripts; (l) Identify the frequency of the cDNA 3' end at each nucleotide in the individual transcripts; as well as (m) When the frequency of the cDNA 3' end at a specific nucleotide in the individual transcript meets a predetermined threshold, the enriched RBP binding site in the individual transcript is detected.

2. The method according to claim 1, wherein the UV box for crosslinking operates at 50-600 mJ / cm 2 The dose is administered over approximately 30 seconds to 10 minutes.

3. The method according to claim 2, wherein the UV box for crosslinking operates at 300-500 mJ / cm 2 The dose is administered over approximately 3 minutes or at a rate of 100-400 mJ / cm. 2 The dose is administered over approximately 10 minutes.

4. The method according to claim 1, wherein the crosslinking is performed with a UV laser for about 10s-20s.

5. The method according to any one of claims 1 to 4, wherein the isolated RNA molecules are treated with the protease at 50°C for about 30 minutes, optionally wherein the protease is proteinase K.

6. The method according to any one of claims 1 to 5, wherein each bound residual peptide is a single amino acid or has a length not exceeding 10 amino acids.

7. The method according to any one of claims 1 to 6, further comprising removing genomic DNA after step (c) to obtain purified RNA molecules.

8. The method of claim 7, further comprising contacting the purified RNA molecules with a protease at 50°C for about 30 minutes prior to step (d), optionally wherein the protease is proteinase K.

9. The method according to any one of claims 1 to 8, further comprising enriching the isolated or purified RNA molecules with poly-dT oligonucleotides or a targeting decoy capture reagent after step (c) but before step (d).

10. The method according to any one of claims 1 to 9, further comprising depleting the ribosomal RNA from the isolated or purified RNA molecule after step (c) but before step (d).

11. The method according to any one of claims 1 to 10, wherein labeling the RNA transcript bound by the residual peptide comprises thermally denaturing the isolated or purified RNA molecule and incubating the isolated or purified RNA molecule with the chemically conjugate containing the affinity ligand in a reaction buffer supplemented with 5%-70% DMSO.

12. The method of claim 11, wherein thermal denaturation comprises heating the isolated or purified RNA molecules at 50°C-100°C for 1-5 minutes.

13. The method according to any one of claims 1 to 12, wherein the chemically derived conjugate comprises succinimide ester, carboxylic acid ester, tetrafluorophenyl ester, sulfonyl dichlorophenol ester, carbonyl azide, aldehyde, or isothiocyanate.

14. The method of claim 13, wherein the chemical conjugate is coupled to the α-amino group of each bound residual peptide.

15. The method according to any one of claims 1 to 12, wherein the chemically derived conjugate comprises a primary amine, carbodiimide, or isocyanate.

16. The method of claim 15, wherein the chemically conjugate is coupled to the carboxyl group of each bound residual peptide.

17. The method according to any one of claims 1 to 16, wherein the affinity ligand comprises biotin, a biotin derivative, digoxigenin, dinitrophenyl, a click chemistry reagent, a sugar, or a peptide.

18. The method of claim 17, wherein the click chemical reagent comprises an alkyne group, an azide group, a dibenzocyclooctynyl group (DBCO), or a bicyclic nonyneyl (BCN) group.

19. The method according to any one of claims 1 to 18, wherein fragmenting the RNA molecule comprises heating the RNA molecule at 90°C to 95°C for 3 to 12 minutes in the presence of 10 mM to 30 mM divalent cations.

20. The method according to any one of claims 1 to 19, wherein the length of said RNA fragment is about 40-90 nucleotides or about 80-250 nucleotides.

21. The method according to any one of claims 1 to 20, further comprising identifying in the individual transcript (i) a 1-3 base pair insertion or deletion generated by PRI isolated by affinity purification, (ii) a single nucleotide variant (SNV) generated by PRI isolated by affinity purification, or (iii) a reverse transcriptase (RT) termination site generated by PRI isolated by affinity purification.

22. The method according to any one of claims 1 to 21, wherein the biological sample is obtained from a subject diagnosed with a disease.

23. The method of claim 22, wherein the disease is cancer, an autoimmune disease, a metabolic disease, or a neurodegenerative disease.

24. The method according to claim 23, wherein the cancer is selected from adrenal cancer, bladder cancer, leukemia, bone cancer, brain cancer, breast cancer, carcinoma, cervical cancer, colon cancer, colorectal cancer, endometrial cancer, ear, nose and throat (ENT) cancer, endometrial cancer, esophageal cancer, gastrointestinal cancer, head and neck cancer, Hodgkin's disease, intestinal cancer, kidney cancer, laryngeal cancer, leukemia, liver cancer, lymph node cancer, lymphoma, diffuse large B-cell lymphoma (DLBCL), lung cancer, melanoma, mesothelioma, myeloma, nasopharyngeal carcinoma, neuroblastoma, non-Hodgkin's lymphoma, oral cancer, ovarian cancer, pancreatic cancer, penile cancer, pharyngeal cancer, prostate cancer, rectal cancer, sarcoma, seminoma, skin cancer, gastric cancer, teratoma, testicular cancer, thyroid cancer, uterine cancer, vaginal cancer, vascular tumors and their metastases.

25. The method according to any one of claims 1 to 24, wherein the biological sample is derived from breast tissue, kidney tissue, cervical tissue, endometrial tissue, head or neck tissue, gallbladder tissue, parotid gland tissue, prostate tissue, brain tissue, pituitary gland tissue, kidney tissue, muscle tissue, esophageal tissue, stomach tissue, small intestine tissue, colon tissue, liver tissue, spleen tissue, pancreas tissue, thyroid tissue, heart tissue, lung tissue, bladder tissue, adipose tissue, lymph node tissue, uterine tissue, ovarian tissue, adrenal gland tissue, testicular tissue, tonsils, thymus, blood, hair, oral cavity, skin, serum, plasma, CSF, semen, prostatic fluid, seminal plasma, urine, feces, sweat, saliva, sputum, mucus, bone marrow, lymph, or tears.

26. The method according to any one of claims 1 to 25, wherein the biological sample is a fresh or frozen sample.

27. The method according to any one of claims 1 to 26, wherein the plurality of cDNA molecules are reverse transcribed using a truncated reverse transcriptase or a readthrough reverse transcriptase.

28. The method according to any one of claims 9 to 27, wherein the targeted decoy trapping agent hybridizes with one or more disease-related genes, optionally wherein the disease is cancer, an autoimmune disease, a neurodegenerative disease, or a metabolic disease.

29. The method according to any one of claims 1 to 28, wherein one or more of the enriched RBP binding sites detected in the individual transcripts are allele-specific and / or disease-related.

30. The method according to any one of claims 1 to 29, further comprising identifying an RBP that binds to one or more of the enriched RBP binding sites detected in the individual transcript.

31. The method of claim 30, further comprising identifying RBP-specific crosslinking patterns within sequence motifs in the individual transcripts.

32. The method of claim 31, wherein the RBP-specific crosslinking pattern is identified using a shared motif defined by eCLIP, an mCross crosslinking pattern, a position-enriched k-mer analysis (PEKA), or a sequence motif defined by RBNS.

33. A method for evaluating the efficacy of a drug in inhibiting the interaction between a specific RNA-binding protein (RBP) and RNA transcripts, the method comprising: a. Contacting a first cell population with at least one inhibitory oligonucleotide that specifically inhibits the expression of RBP, wherein the first cell population is obtained from a biological sample; b. Contact the second cell population with the test drug, wherein the second cell population is obtained from the biological sample and has not been treated with the at least one inhibitory oligonucleotide; c. Performing the method according to claim 1 to identify one or more RBP-RNA interaction (PRI) sites in at least one transcript derived from the first cell population, the second cell population, and the control cell population, wherein the control cell population is obtained from the biological sample and has not been treated with the at least one inhibitory oligonucleotide and the test drug; d. Define RBP-specific PRI by identifying crosslinking patterns within one or more sequence motifs in the individual transcripts, wherein the crosslinking patterns (i) are present in the control cell population and (ii) are not present in the first cell population; as well as e. When the RBP-specific PRI identified in step (d) is not present in the second cell population, the test drug is determined to be effective.

34. A method for evaluating the specificity of a drug in inhibiting the interaction between a specific RNA-binding protein (RBP) and RNA transcripts, the method comprising: a. Contacting a first cell population with at least one inhibitory oligonucleotide that specifically inhibits the expression of RBP, wherein the first cell population is obtained from a biological sample; b. Contact the second cell population with the test drug, wherein the second cell population is obtained from the biological sample and has not been treated with the at least one inhibitory oligonucleotide; c. Performing the method according to claim 1 to identify one or more RBP-RNA interaction (PRI) sites in at least one transcript derived from the first cell population, the second cell population, and the control cell population, wherein the control cell population is obtained from the biological sample and has not been treated with the at least one inhibitory oligonucleotide and the test drug; d. The nonspecific effect of the test drug is identified by identifying crosslinking patterns within one or more sequence motifs in the individual transcripts, wherein the crosslinking patterns (a) are present in the control cell population and the first cell population and (b) are not present in the second cell population.

35. The method according to claim 33 or 34, wherein the at least one inhibitory oligonucleotide is siRNA, shRNA, antisense oligonucleotide, or sgRNA.

36. The method according to any one of claims 33 to 35, wherein the test drug is a small molecule, nucleic acid, or peptide.

Images