General early cancer diagnosis

Abstract

The present invention relates to general early cancer diagnosis, and in particular, methods for quantifying DNA methylation are disclosed that can be used to screen for disease (e.g., cancer), diagnose disease (e.g., cancer type), monitor progression of disease, and monitor response to treatment.

Description

[0001] This application is a divisional application of the international application filed on May 14, 2018, with international application number PCT / US2018 / 032612, which entered the Chinese national phase on December 24, 2019, with application number 201880042384.X and invention title "Universal Early Cancer Diagnosis".

[0002] Related applications

[0003] This application claims the benefits of U.S. Provisional Application No. 62 / 560,660, filed September 19, 2017; U.S. Provisional Application No. 62 / 511,648, filed May 26, 2017; and U.S. Provisional Application No. 62 / 505,647, filed May 12, 2017. The entire teachings of the above applications are incorporated herein by reference.

[0004] Government support

[0005] This invention was made with government support under grants granted by the National Institutes of Health (NIH) Nos. CA193461, HG006193, GM099117, and DA036898. The government owns certain rights to this invention. Technical Field

[0006] This invention relates to universal early cancer diagnosis. Background Technology

[0007] DNA methylation has been used for cancer detection over the past two decades, but current methods rely on the average methylation of specific genes or gene panels, which lacks both sensitivity and specificity for non-invasive cancer diagnosis. In mammals, the canonical somatic DNA methylation panorama is established on embryonic specialization and subsequently disrupted in many cancer types. However, the underlying mechanisms guiding this genome-scale transformation remain elusive, and no clear model exists for its systematic acquisition or potential practical application. Summary of the Invention

[0008] This article discloses a method for quantifying DNA methylation, which can be used to screen for diseases (e.g., cancer), diagnose diseases (e.g., cancer type), monitor disease progression, and monitor response to treatment.

[0009] This article discloses a method for detecting circulating tumor DNA (ctDNA) in a sample, the method comprising using the proportion of uniformly methylated reads (PMR) of the sample to detect ctDNA in the sample. In one aspect, a methylated sequence of the sample is obtained, and at least one CpG island (CGI) is identified on the methylated sequence. The PMR of the identified CpG island is calculated and compared to a control background of normal tissue or epiblast. The presence of ctDNA in the sample is detected when the PMR of the sample is greater than the control background (e.g., when the signal is high by a bank sum test).

[0010] In some respects, the sample is selected from plasma, urine, stool, menstrual fluid, or lymph. The sample may contain cell-free DNA. In some respects, 0.01% to 1% ctDNA is detected in the sample, more specifically, 0.01% ctDNA. In some respects, the presence of ctDNA in the sample is detected with a sensitivity greater than 80%. In some respects, the presence of ctDNA in the sample is detected with a specificity greater than 75%. The presence of ctDNA in the sample can be detected with 100% sensitivity and 95% specificity.

[0011] In some respects, the presence of ctDNA indicates the presence of cancer. Samples can be obtained from individuals diagnosed with cancer, who have cancer, are at risk of developing cancer, or are suspected of having cancer. The cancers can be selected from urothelial carcinoma of the bladder, invasive breast cancer, colonic adenocarcinoma, colorectal adenocarcinoma, esophageal cancer, head and neck squamous cell carcinoma, clear cell renal carcinoma, papillary renal carcinoma, hepatocellular carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, prostate adenocarcinoma, stomach and esophageal cancer, thyroid cancer, endometrial cancer, and chronic lymphocytic leukemia. In some respects, the presence of ctDNA indicates the presence of a tumor.

[0012] This article also discloses a method for screening for cancer, which includes detecting ctDNA in a sample using the proportion of consistent methylated reads (PMR), wherein the presence of ctDNA indicates that the subject has cancer.

[0013] This article also discloses a method for treating subjects in need of cancer treatment, the method comprising: detecting ctDNA in a sample using the proportion of uniformly methylated reads (PMR) of the sample, wherein the presence of ctDNA indicates that the subject has cancer; and treating the subject's cancer.

[0014] This article also discloses a method for monitoring a subject's response to cancer treatment, the method comprising: detecting the amount of ctDNA in a first sample using the proportion of homologous methylated reads (PMR) obtained before the subject receives cancer treatment; detecting the amount of ctDNA in a second sample using the proportion of homologous methylated reads (PMR) obtained after the subject receives cancer treatment; and comparing the amount of ctDNA obtained from the first sample with the amount of ctDNA obtained from the second sample, wherein an increase in ctDNA indicates a negative response to cancer treatment and a decrease in ctDNA indicates a positive response to cancer treatment.

[0015] This article also discloses a method for monitoring the progression or improvement of cancer in subjects, the method comprising identifying ctDNA from the subject's cfDNA using a proportion of consistent methylation reads, wherein the presence of ctDNA indicates that the subject is at risk of developing cancer, and monitoring the amount of ctDNA in the cfDNA over time, wherein changes in the amount of ctDNA in the cfDNA indicate disease progression or improvement.

[0016] This article also discloses a method for assessing cancer in subjects, the method comprising identifying the presence of ctDNA from the subject's cfDNA using a proportion of consistent methylation reads, wherein if ctDNA is present, the subject has cancer or is at risk of developing cancer.

[0017] This paper also discloses a method for disrupting the methylation of CpG islands, the method comprising reducing PRC2 expression. Eed The expression of [certain substances] can be reduced through genome modification (e.g., CRISPR). Eed The expression of methylation. This paper further discloses a method for disrupting the methylation of CpG islands, the method comprising reducing methylation. Dnmtl , Dnmt3l or Dnmt3b The expression of [certain substances] can be reduced through genome modification (e.g., CRISPR). Dnmtl , Dnmt3l or Dnmt3b The expression.

[0018] This paper also discloses a method for disrupting the methylation of CpG islands, the method comprising mutating a member of the FGF pathway. This paper further discloses a method for disrupting the methylation of CpG islands, the method comprising mutating a member of the FGFR pathway. Attached Figure Description

[0019] This patent or application document contains at least one color-drawn drawing. Upon request and payment of the necessary fees, the Patent Office will provide a published copy of this patent or application with the color drawing.

[0020] Figure 1A-1E It presents a panoramic view of different post-implantation DNA methylation processes. Figure 1A The early developmental time series collected for this study are shown, including pre-compacted 8-cell stage embryos (2.25 days post-fertilization; E2.25), trophectoderm (TE) and inner cell mass (ICM) of E3.5 blastocysts, and extraembryonic ectoderm (ExE) and epiblast of E6.5 gestational bodies (n = 2 WGBS libraries / samples, see Methods section). Figure 1B The distribution of CpG methylation in 100 bp tiles is shown (top); the median methylation of the 100 bp tile varies with local CpG density (bottom), where the shaded areas represent the 25th and 75th percentiles. Figure 1C The study showed feature-level enrichment of differentially methylated CpGs compared to the genomic background. ExE-hypomethylated CpGs were mainly found in non-gene sequences, while ExE-hypermethylated CpGs were located in CpG islands (CGIs), transcription start sites (TSSs), and 5' exons. Here, TSS refers only to the upstream 1 kb of the annotated TSS, and 5' exons represent non-overlapping sets. Figure 1D The median methylation structure flanking the ExE-hypermethylated TSS in embryonic and extraembryonic tissues is shown, along with the relative methylation differences (Δ), which are highly differentiated at implantation. Shaded areas represent the 25th and 75th percentiles per 100 bp bin. Figure 1E This provides a genome browser track for WGBS, determination of transposase-accessible chromatin using high-throughput sequencing (ATAC-seq), and capture of RNA-seq data from three landmark loci. Density refers to the expected number of methylated CpGs per 100 bp of basic sequence and highlights the extensive epigenetic signals present in these regions within the ExE (Δ density refers to the difference compared to the epiblast). For Otx2 and Gata4 ExE-specific methylation and repression occur simultaneously, while HoxC clusters are expressed late in embryonic development. CGI is highlighted in green. TPM, per million transcripts.

[0021] Figures 2A-2F De novo methylation of promoters of genes that demonstrate early development can be regulated by external conditions. Figure 2A A schematic diagram illustrating the signaling pathway interactions between the epiblast (blue) and the ExE (red) is provided. Fibroblast growth factor (FGF) produced by the epiblast promotes ExE development, and the ExE expresses bone morphogenetic protein 4 (BMP4) to induce WNT protein in the epiblast. The epiblast-secreted precursor Nodal is processed by the ExE to establish a proximal-to-distal gradient and primary striations. 10 . Figure 2BDifferential expression and promoter methylation of key signaling components between the ExE and epiblast were shown. Many Fgf and associated receptors exhibited interactive expression and promoter methylation. Wnt3 induction was prominent in the epiblast, while Wnt6 and 7b were highly expressed in both the trophectoderm and ExE. Differential promoter methylation was defined as the TSS (±1 kb) of the annotation with the largest absolute difference. Figure 2C The ICM growth halo is shown after four days of incubation under conditions of entirely different growth factors or small molecules intended to stimulate or inhibit FGF and WNT activity. The outline highlights the purified components (methods section). Figure 2D Showing the Figure 2C The methylation box plots for the conditions described in the figure include all 100 bp tiles captured by RRBS and CGI targeted by ExE (ExE hyper CGI). The edges are at the 25th and 75th percentiles, respectively, and the whiskers are at the 2.5th and 97.5th percentiles, respectively. Figure 2E Significant CGI methylation was observed in ExE, the outer FGF / CHIR layer, and the FGF growth halo. The intersection of methylated CGIs increased by ≥0.1% compared to the epiblast (n = 3,420). The FGF4 condition exhibited the highest number of methylated CGIs, but intersected less with ExE than with the presence of CHIR: 25% of the ExE hyper CGIs overlapped with both conditions, while 51% overlapped with the outer FGF / CHIR growth halo (outer layer). Figure 2F Showing Figure 2E Clustering of differentially methylated CGIs, including methylation status in ExE, PRC2 regulation of the embryo, and TSS proximity (±2 kb). F / C inside and outside refer to the internal and external FGF / CHIR growth halo status, respectively.

[0022] Figures 3A-3D This study demonstrates the contribution of novel epigenetic regulators to the panorama of extraembryonic methylation. Figure 3A Box plots of the E6.5 epiblast from wild-type (WT) and CRISPR–Cas9-damaged samples are provided, including 100 bp tiles and ExE hyper CGI measurements by RRBS. Edges are at the 25th and 75th percentiles, respectively, and whisker lines are at the 2.5th and 97.5th percentiles, respectively. Figure 3B Provides a similar functionality to the sample matching EXE. Figure 3A Box plot. (and) Dnmt3a -and Dnmt3b - Compared to positive epiblasts Dnmt1 or Dnmt3b The disruption has a much greater impact on overall methylation levels and leads to high levels of genome depletion. Figure 3C A composite graph of the ExE hyper CGI obtained by knocking out states is provided. Eed - In invalid ExEs, CGI methylation is disrupted, particularly within +1 kb of TSS, without affecting the overall level. The black line represents the wild-type median. The composite plot maps the median from the RRBS data to 200 bp windows at 50 bp intervals. The gray boxes indicate ±1 kb of TSS. Figure 3D A heatmap of differential CGI methylation (≥0.1) between CRISPR–Cas9-targeted epiblasts or ExEs and their wild-type counterparts (n = 2,461) is provided. Differential ExE methylation status compared to epiblast and TSS proximity (±2 kb) is included for reference.

[0023] Figures 4A-4E This study demonstrates that extraembryonic targeted CpG islands are widely methylated in human cancer types. Figure 4A The results show that overall methylation disruption produced similar biases on CGI and promoters between ExE / epoderm or patient- or age-matched normal / cancer tissue comparisons. The heatmap shows the log z-scores for feature enrichment based on the above comparisons and a binomial test for trophoblastic ectoderm / ICM (TE). Of these 16 cancer types, only THCA did not show a significantly dysregulated methylome. The n-value refers to the number of matched cancer / normal tissue isolates in each type. The Cancer Genome Atlas (TCGA) samples included bladder urothelial carcinoma (BLCA), invasive breast carcinoma (BRCA), colonic adenocarcinoma (COAD), colorectal adenocarcinoma (READ), esophageal cancer (ESCA), head and neck squamous cell carcinoma (HNSC), clear cell renal carcinoma (KIRC), papillary renal carcinoma (KIRP), hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate adenocarcinoma (PRAD), gastric and esophageal cancer (STES), thyroid cancer (THCA), and endometrial cancer (UCEC). Here, the comparison of chronic lymphocytic leukemia (CLL) with B lymphocytes was performed between age-matched samples measured by WGBS. Figure 4B Box plots of 489 ExE hyper CGIs are provided, representing ExE hyper CGIs that remain in their state in humans, calculated for each of the 15 cancer types where CGI methylation is typically evident, based on the characteristics of each cancer or normal tissue. Asterisks: CLL samples measured by RRBS (n = 119) and represent comparisons between age-matched healthy B lymphocytes (n = 24). Margins are at the 25th and 75th percentiles, respectively, and whiskers are at the 2.5th and 97.5th percentiles, respectively. Figure 4CThis paper presents a heatmap of differential methylation for 8,942 orthologous CGIs, measured in TCGA or by RRBS and clustered by Euclidean distance. DMR bars include the cumulative number of cancer types in which a given CGI is referred to as hypermethylated, and the DMR status shared (conserved) between human placenta and human embryonic stem (ES) cells, mouse ExE cells and epiblast cells, or both comparisons. PRC2 indicates regulation by polycomb proteins in human ES cells. The numbers reflect the proportion of differentially methylated CGIs in each set across at least one cancer type. Figure 4D Intersection analysis of DMR status in TCGA and CLL samples is shown. Placental and ExE DMRs were similarly enriched in methylation in at least one human cancer type (86% and 84%, respectively, compared to 35% for all CGIs), and were methylated more frequently in them. Enrichment of conserved DMRs was greater than enrichment of extraembryonic DMRs for each individual species, and methylation was present in 94% of DMRs in at least one cancer type. Figure 4E Box plots of orthologous ExE hyper CGI from 107 ENCODE / Roadmap Epigenomics Project samples are provided, categorized by average methylation and highlighting cancer or cancer cell lines (in red). Samples classified alongside cancer and assigned a "normal" value, in descending order, include the trophoblast cell line HTR8svn, primary colon and colonic mucosa, placenta, and CD8+ T lymphocytes. Figure 13A Including other sample characteristics.

[0024] Figure 5A-5G This study demonstrates the traced differentiation within a panoramic view of DNA methylation during mouse implantation. Figures 5A-5F Sequencing metrics and coverage information for WGBS, RNA-seq, and ATAC-seq data are provided, including hierarchical clustering and Pearson correlations for CpG, genes, and gene promoters. WGBS data also include Euclidean distance, which is advantageous for examining sample similarity in overall low-methylation samples and for similarity scores on 100 bp tiles, locally incorporating the inherently higher variability of moderately methylated CpG to reduce noise. For RNA-seq and ATAC-seq data, biological duplicates are clustered together, as are 8-cell and post-implantation WGBS data, while tissues from E3.5 blastocysts are clustered together, rather than as separate inner cell mass (ICM) and trophectoderm (TE) compartments. Generally, there is minimal variation between the methylation status of the ICM and trophectoderm, with only slight deviations around the minimum total values ​​reached during this developmental period. Figure 5GThe epiblast and exoblast (ExE) are shown separated from the E6.5 post-implantation embryo. The gestational body was first removed from the maternal decidual tissue and then separated into epiblast and exoblast portions, with the apical ectoplacental cone (EPC) carefully removed. The visceral endoderm (VE) and trophoblast cells were then enzymatically digested and mechanically removed using a thin-glass capillary tube.

[0025] Figure 6A-6G It showcases the unique features of the extraembryonic methylation panorama. Figure 6A Box plots of CpG methylation are provided for all covered CpGs, as well as those that are significantly hypermethylated or hypomethylated within the ExE compared to the epiblast (ExE hyper or ExE hypo, respectively). Across the genome, ExE hypo CpGs largely reflect differential remethylation compared to the epiblast. Furthermore, ExE hyper CpGs are mostly unmethylated in the ICM and trophectoderm, and remain so in the epiblast, suggesting an ExE-specific mechanism. The margins are the 25th and 75th percentiles, respectively, and the whisker lines are the 2.5th and 97.5th percentiles, respectively. Figure 6B The differential methylation distribution of hyper or hypo CpG in ExE compared to the epiblast is shown. Hypomethylation appears to be a general feature of ExE and deviates from the default hypermethylated state in the epiblast. Alternatively, increased DNA methylation appears to be localized to unmethylated regions within the epiblast and originates therefrom, subsequently extending to embryonic and adult somatic cell tissues. Figure 6C The alternating CpG density distributions of ExE hypo and hyper CpGs are shown, indicating differential enrichment within different genomic features. However, ExE hypo CpGs are similar to the overall average, while ExE hyper CpGs appear in regions of higher CpG density. Figure 6D The percentage of dynamically methylated CpGs falling within annotated exons varies with distance from their assigned TSS. 44% of exon ExE hyper CpGs fall within 2 kb of their associated TSS. Figure 6E The percentage of dynamically methylated CpGs falling into annotated CpG islands (CGIs) is shown based on the proximity of the dynamically methylated CpG to the nearest TSS. ExE hyper CpGs are typically located proximal to the TSS and sloping downstream of the TSS, with 43% falling within ±2 kb. Figure 6FThe distribution of DNA methylation, including sequences of genes (TSS, exons, introns, and CGI) and repetitions (LINE, SINE, and LTR), is shown. For reference, black bars and arrows highlight the overall median and the 25th and 75th percentiles, respectively. Overall, all features exhibited the expected minimum DNA methylation values ​​within the ICM and trophectoderm of the E3.5 blastocyst prior to remethylation at implantation. Compared to its overall distribution, the ExE showed higher levels of de novo methylation within exons and introns, but lower levels in regions of retrotransposon origin, such as LINE and LTR. Depending on the genomic feature, the epiblast showed almost complete hypermethylation or hypomethylation and a bimodality at the TSS, which often contained CGI. The n-value refers to the number of annotated features of a given type. Figure 6G Violin plots providing 100 bp methylation data from early embryos, placenta, and fetal tissues illustrate the macroscopic epigenetic retention of somatic epiblastic or extraembryonic structures throughout subsequent development. White dots highlight the overall median, while blue and red reflect the median of 100 bp hypomethylated tiles and hyper CGIs in the ExE, respectively. Notably, the placenta largely retains the overall panorama of hypomethylation and targeted CGI methylation within the ExE, as established by E6.5. We separately display 100 bp tiles and CGIs specific to ExE for both hypomethylation and hypermethylation to restrict CpG as a salient feature where they change as a whole. WGBS data for somatic tissues and mid-pregnancy placenta are taken from reference 11.

[0026] Figures 7A-7H This demonstrates that transcriptional differences between the epiblast and ExE are partly guided by DNA methylation. Figure 7A Enrichment analysis of a selected set of TSS genes exhibiting ExE hypermethylation, including gene ontology, canonical pathways, and genetic and chemical perturbations, revealed a high enrichment of transcription factors and signaling pathways involved in early embryonic patterning. Furthermore, these genes containing promoter CGIs are canonical targets of PRC2, which coordinates the selective expression of key developmental regulators during gastrulation. Figure 7B The tumor suppressor p16 was shown. Ink4a p19 Arf (Both are from) Cdkn2a (encoding) and p15 Ink4b (Depend on Cdkn2b DNA methylation and open chromatin dynamics (encoding). Although these loci are basal transcribed or non-transcribed during early development, three regions are dynamically methylated in the ExE (highlighted in gray), including a >10-kb region that covers the entire p16Ink4a The locus is either completely unmethylated in the epiblast or extensively methylated in the ExE. The CGI is highlighted in green, and the location of the contained TSS is highlighted in red. Figure 7C A scatter plot of log2 expression kinetics against differential CGI methylation between the epiblast and ExE is provided. Although most genes containing dynamically methylated CGI promoters are functional and not yet highly expressed during later embryonic development, de novo methylation in ExEs is generally associated with transcriptional repression. ExE hyper CGIs are highlighted in pink. Promoter CGIs are assigned to the nearest genes within ±2 kb boundaries. Figure 7D Box plots of promoter methylation and expression are provided in extraembryonic and embryonic compartmental constraints. Promoters are defined as ±1 kb of the annotated TSS and are scored as dynamically methylated in the ExE if the difference from the epiblast is ≥0.1. Expression changes between the dynamically methylated promoter set and the basal promoter set are presented at increasing thresholds, based on their expression in the epiblast. Although many genes containing CGI promoters are not dynamically expressed in the epiblast or ExE and are associated with downstream developmental functions, transcriptional repression is a consistent feature of promoter methylation, even at these low thresholds. Figure 7E The median open chromatin signal for ExE hyper CGI-related TSS, measured by ATAC–seq, is shown during the pre-implantation to post-implantation transition. ExE hyper CGI-related genes are heavily enriched for their roles in embryonic patterning and are not majorly expressed until the onset of gastrulation. During the transition from the blastocyst to the germ layer, these promoters acquire open chromatin signals, indicating transcription initiation or activation, which is not observed in ExEs where these promoters are de novo methylated. Shaded areas reflect the 25th and 75th percentiles per fixed 100 bp bin. Figure 7F The expression of key epigenetic and master transcriptional regulators and differential promoter methylation during early embryonic and extraembryonic development were revealed. Most epigenetic regulators showed minimal expression differences between the epiblast and ExE, with Dnmts being a notable exception. Key isoforms of Dnmt3a and Dnmt3b, along with overall remethylation, were upregulated in the epiblast, while repression of Dnmt3a in the ExE corresponded to de novo promoter methylation. Alternatively, maintenance methyltransferase Dnmt1 and the non-catalytic cofactor Dnmt31 were induced in the blastocyst and maintained at high levels in the ExE, while in the epiblast... Dnmt3lPromoter reversible methylation occurs. The H3K36 demethylase Kdm2b shows differential expression of catalytically active and inactive isoforms in the epiblast and ExE, respectively, with isoform switching appearing to be performed de novo methylation around the CGI promoter used by somatic cells. The ExE is characterized by persistent expression of the major regulators Cdx2, Eomes, and Elf5 (references 48-51), while the still pluripotent epiblast remains positive for Pou5f1 (also known as Oct4). Many other regulators in subsequent developmental stages are predominantly expressed in the epiblast, and their promoters are de novo methylated in the ExE. Differences in promoter methylation are indicated by the TSS (transfer signature) of the annotations showing the greatest absolute difference between the ExE and the epiblast. TPM (transfer signature per million transcripts) is also mentioned. Further high-resolution genome browser tracks for selected transcriptional and epigenetic regulators are shown in Figures 8 and 11, respectively. Figure 7G Unsupervised hierarchical clustering of 11,780 genes in late preimplantation and early postimplantation development is shown, divided into 20 distinct dynamics (“clusters”). Cluster 10 includes genes specifically induced within the epiblast rather than within the ExE. Heatmap intensity reflects row-normalized z-scores. Figure 7H yes Figure 7F The enrichment of 20 significant gene ontologies representing gene expression dynamics, including those regulated by ExE-methylated CGI promoters, was calculated using a binomial test. Cluster 10 was enriched for both developmental function and ExE promoter methylation.

[0027] Figures 8A-8D This section showcases the unique bifurcation and epigenetic enhancement of transcriptional regulators during post-implantation development. Genome browser tracks of WGBS, ATAC-seq, and RNA-seq data for transcriptional regulators associated with embryonic or extraembryonic development are displayed. CGI is highlighted in green, and the location of contained TSSs is highlighted in red. Embryonic regulators include Pou5f1, Nanog, and Pdrm14, which are progressively expressed pre-implantation, with Pou5f1 and Nanog remaining expressed in the epiblast. For these genes, repression in the ExE is accompanied by differential methylation of their TSSs, which acts as... Pou5f1 The localized hypermethylation "peak" within the ~5 kb region at the gene locus, which was originally hypomethylated in the epiblast, is clearly visible. NanogAt the locus, the upstream region remains hypomethylated in both tissues. Finally, de novo methylation of the Prdm14 promoter is representative of ExE-specific CGI promoter methylation occurring at hundreds of genes with downstream developmental functions. Density refers to the predicted number of methylated CpGs per 100 bp of basic sequence and particularly highlights the extensive epigenetic signals present in these regions within the ExE (Δ density refers to the difference compared to the epiblast). Figure 8B This showed that extraembryonic development was partially guided by the master regulator Elf5, which was not induced until implantation and was reversibly methylated at its TSS in the epiblast. Interestingly, many transcriptional regulators associated with pluripotency and germline development persisted within the ExE, including Zfp42 and its paralogs Dppa2 and Dppa4. Elf5 Similarly, the promoters of these genes are differentially methylated in the epiblast and are often characterized by extensive kilobase-level hypomethylation around their TSS in the ExE. Figure 8C A scatter plot of log2 TPM versus promoter methylation is provided, revealing a higher sensitivity to low methylation levels in ExEs compared to the epiblast. Median, 25th, and 75th percentiles of expression within the 0.1 methylation bin are included for reference. The percentages of unmethylated promoters are very similar across tissues and exhibit comparable expression values. Promoters are calculated as ±1 kb of the annotated TSS. The vertical dashed line indicates the median methylation value for ExE hyper CGI. Figure 8D This study shows read-level methylation in ExE hyper CGIs from both the ExE and epiblast. The methylation status of each sequencing read within a given CGI was ranked and binned by percentile. The median, 25th, and 75th percentiles of these ranks were plotted for both ExE and epiblast hyper CGIs. Typically, about 80% of reads falling into these regions are methylated in the ExE, with a median methylation value of 0.25. This value is very close to the mean undetermined phase measurement for the overall CGI, indicating that de novo methylation occurs in ExEs at high fractions and to a similar degree.

[0028] Figures 9A-9C This study demonstrates the epigenetic limitations of FGF production and its sensing of embryonic or extraembryonic compartments. Figure 9A Genome browser tracks show WGBS, ATAC-seq, and RNA-seq data of selected growth factors, receptors, and enhancers that are dynamically regulated during early post-implantation development. Fgf Loci such as ICM expression Fgf4 and epiblast expression Fgf5 and Fgf8All are regulated by CGI-containing promoters that are de novo methylated in the ExE. Alternatively, FGF-sensing genes, for example... Fgfr2 and enhancing protein Fgfbp1 The expression of Fgfr2 becomes specific to ExE and is characterized by extensive kilobase-level hypomethylated domains surrounding their respective TSSs in this tissue. Furthermore, the asymmetric distribution of Fgfr2-expressing cells during ICM specialization suggests that the tissue remains sensitive to these growth factors before epigenetic restriction is imposed by DNA methylation during implantation. 52,53 CGI is highlighted in green, and the location of contained TSS is highlighted in red. Density refers to the expected number of methylated CpG per 100 bp of basic sequence, and particularly highlights the extensive epigenetic signals present in these regions within the ExE (Δ density refers to the difference compared to the epiblast). Figure 9B Bright-field images of ICM growth halos after two to four days under entirely different growth factor or small molecule conditions are provided. All ICMs were cultured on irradiated feeder cells in basal N2 / B27 medium supplemented with leukemia inhibitory factor (LIF). 2i refers to standard FGF inhibition and WNT activity conditions, including the MEK inhibitor PD0325901 and the GSK3β inhibitor CHIR99021, the latter acting as a WNT agonist. 37 PD refers to FGF signaling suppressed when cultured alone with PD0325901 without additional WNT input. 54 FGF4 / CHIR exhibited dual FGF and WNT activity upon culturing with recombinant FGF4 and CHIR99021, and included distinct internal and external tissue structures that appeared and independently separated and patterned during culture. Finally, the ICM was cultured in separate FGF4 cells. Outlines highlight specific portions of each growth halo, which were subsequently purified for dual RNA-seq and RRBS analysis (see Methods section). Scale bars are shown in the lower right. Figure 9C The results show that, based on developmental trajectories, differential methylation of CGIs during in vitro culture differs from that of ExE. Specific TSS-related CGIs are shown, with methylation occurring in ExE, under both ExE and FGF / CHIR conditions, or only under ExE, and the corresponding log2 fold changes in average regulation of gene expression. Common targets include early developmental genes that are repressed in each condition, such as… Prdm14 Although they are often highly expressed within FGF / CHIR, it is noteworthy that some of these genes, particularly those related to germline, can undergo de novo methylation late in embryonic development. 55The difference between FGF and ExE and FGF / CHIR conditions lies in the methylation of CGIs associated with the epiblast or neuroectoderm, including genes expressed under FGF conditions, such as... Otx2 , Igfbp2 and Sfrp2 Although this set includes other neuroectodermal master regulators that have not yet been expressed, such as Pax6 Finally, ExE and FGF / CHIR differentiate at the promoter methylation of endoderm master regulators, which are highly expressed, for example, in transitions from the interior to the exterior of FG / CHIR. Foxa2 , Hnf1b , Gata4 and Sox17 It is worth noting that the branch of CGI methylation corresponds to Fgfr2 Expression and Fgf4 Inhibitions were observed in vivo as follows: Fgf4 was highly expressed internally but inhibited externally (32.0 to 3.5 TPM), while Fgfr2 (2.3 to 13.5 TPM) was induced. PD and FGF / CHIR conditions were also uniquely positive for Dnmt3b and 3l expression, but no ExE hyper CGI methylation was observed in the presence of PD0325901 (TPMs of Dnmt3b and Dnmt3l were 30.2 and 60.9 externally to FGF / CHIR, compared to 61.0 and 41.3 for PD), suggesting a need for additional cofactors or post-translational modifications to redirect these enzymes to this characteristic set.

[0029] Figures 10A-10C This demonstrates the generation of dual expression and methylation libraries from growth halo and embryo knockout data. Figures 10A-10B Sequencing metrics and coverage information for dual RNA-seq and RRBS libraries generated to evaluate ICM growth halo and CRISPR–Cas9-disrupted E6.5 embryos are provided, including similarity metrics between duplicate samples (Euclidean distance and Pearson correlation for RRBS and Pearson correlation for RNA-seq). RRBS samples also include the mean and median methylation values ​​for 100 bp tiles. Figure 10C The study demonstrates that CRISPR-Cas9-disrupted embryos are generated via zygote injection of three single-stranded guide RNA (sgRNA) sequences, which are specific for early exons common to different isoforms. Genomic coordinates and preinterstitial sequences are provided (see Methods section).

[0030] Figure 11A-11HThe dynamic behavior of key epigenetic regulators during early implantation is shown. Genome browser tracks for WGBS, ATAC-seq, and RNA-seq data are displayed (log2 TPM is shown for selected isotypes). CGI is highlighted in green, and the location of the contained TSS is highlighted in red. Figure 11A The results showed that Dnmt1 was not significantly expressed in early division, partly due to transient maternal imprinting exceeding the TSS used by somatic cells. Dnmt1s ) 33,56 However, it showed moderate induction within the ICM. Then, at implantation, it was induced in both the epiblast and the ExE. Dnmt1 was expressed at higher levels in the ExE and in the maternal-specific TSS (…). Dnmt1o Persistent focal hypomethylation was observed around the blastocyst, which was not observed in the epiblast, turning preimplantation-specific hypomethylated regions into hypermethylated genomic averages. Figure 11B The short Dnmt3a2 isoform was shown to be induced to high levels in the epiblast and also expressed in embryonic stem (ES) cells. Alternatively, Dnmt3a2 The CGI-containing promoter is methylated in ExE and its transcription is inhibited. Figure 11C The results show that, like Dnmt1, the Dnmt3b promoter contains CGIs imprinted by the maternal body during the preimplantation period. 33,56 It is clearly induced in the blastocyst, but its abundance becomes asymmetrical in the epiblast after implantation. Figure 11D DNMT3L is shown to be a non-catalytic cofactor that enhances the de novo activity of DNMT3A and B, and has specific functions in early embryos and germlines. 57 During implantation, Dnmt3l is initially expressed in both the ICM and trophectoderm, but it remains expressed in the ExE and is silenced in the epiblast by de novo promoter methylation. Figure 11E The results show that the H3K36 demethylase KDM2B plays a specific role in establishing the boundary between the promoter and the active transcriptional genome, as well as in PRC2 recruitment and the establishment of facultative heterochromatin. 58–61 Non-catalytically active isoforms Kdm2b2 The process begins with another TSS downstream of the exon encoding the demethylated Jumonji domain of catalytically active Kdm2b1 (Reference 17). Kdm2b2 is the most prevalent isoform during preimplantation development and is still expressed in the ExE. Alternatively, Kdm2b1 may be induced only within the epiblast during implantation, while its CGI-containing promoter is methylated in the ExE. Like Dnmt1s and Dnmt3b, the CGI promoter of Kdm2b1 is a maternal methylation imprint that becomes hypomethylated during implantation. 33,56 . Figure 11FExtraembryonic genomic remethylation was shown to be highly dependent on DNMT3B and DNMT1. Pairwise comparisons were made between wild-type epiblast and ExE (y-axis) plots (x-axis) measured by RRBS against matched 100 bp tiles of CRISPR-Cas9 disrupted tissue. Dnmt1 , Dnmt3b and Dnmt31 When disrupted, extraembryonic methylation levels are weakened at the whole-genome level. The epiblast only affects... Dnmt1 and Dnmt3b Both are less sensitive to damage than ExE, presumably due to compensation from DNMT3A. It is worth noting that when... Dnmt1 When absent, the overall methylation level of ExE decreased more than that of the epiblast, indicating a higher dependence on maintenance and lower efficiency of de novo methyltransferase activity in this tissue. Gray marker lines are included, along with red LOESS regression best fits. The upper left of each figure includes the number of 100 bp tiles used in each comparison and... r2 value. Figure 11G Composite diagrams of CRISPR-Cas9-targeted epiblasts and ExE containing the ExE hyper CGI promoter are provided. Generally, in samples with ineffective Eeds, limited effects are observed only in the epiblast, aside from a slight increase in peripheral methylation. Alternatively, in Dnmt1 -、 Dnmt3b -and Dnmt3l - In null exoesters (ExEs), both proximal and peripheral methylation of the TSS are reduced. Null exoesters exhibit unique specificity in weakening methylation at the TSS, particularly downstream of the first kilobase. Both the epiblast and the exoester contain black wild-type medians for comparison. Lines represent the median, and shaded areas represent the 25th and 75th percentiles, respectively. For RRBS data, the composite plot shows the median within a 200 bp window with 50 bp intervals. Figure 11H Statistical tests for the suppression of hyper CGI-related genes in ExE were provided, demonstrating comparable demand for Eed in both the epiblast and ExE. Gene expression in knockout samples was compared to matched wild-type samples using DESeq2 with raw counts as input. Enrichment of hyper CGI-related genes in ExE was assessed using the Wilcoxon rank-sum test and expressed as z-scores, converted to p-values ​​assuming a normal distribution. Bonferroni correction for multiple tests was used to derive the free-response ratio (FDR).

[0031] Figure 12A-12F The general characteristics of cancer methylome and CGI DMR are shown. Figure 12AThis study shows a comparison of median methylation of differentially regulated CGI-containing promoters in primary colon tumor isolates and CLLs, as measured by WGBS, with those in colon and B lymphocytes. Hyper CGIs of ExEs, as identified in this study and shown in Figure 1, are included for reference. Differences in median methylation between extraembryonic or cancerous tissues and epiblastic or normal tissues are also included. The general characteristics of these two cancer methylation groups are similar to those of ExEs, with a maximum increase in DNA methylation centered on the TSS that steadily diminishes in the periphery. Alternatively, hypomethylated CGIs in extraembryonic or tumorigenic settings show maximum variation at boundaries or “CpG islands” at a certain distance from the TSS, as previously reported in cancer studies. 62 The shaded areas represent the 25th and 75th percentiles per 100 bp bin. Figure 12B The methylation levels of hypermethylated CGIs in ExE for epiblasts, colon tumors for colon, and CLLs for B lymphocytes are shown, including subsets of those with differential methylation states between the cancers and extraembryonic developments. The methylation state of each sequencing read within a given hypermethylated CGI was ranked and binned by percentile. The median and 25th and 75th percentiles of these ranks, referred to as hypermethylated CGIs in each pairwise comparison, are plotted. The comparisons of ExE / epidermis and CLL / B lymphocytes show very similar distributions indicating overall inconsistency, meaning that clustered methylation in these features is similar to that observed in isomorphism, likely obtained through dispersive de novo methylation across most alleles within the population. Colon tumors exhibit significantly higher read-level methylation, with a median of ~0.7. However, the per-read methylation levels in non-tumor-matched colon tissues are also quite high, with >50% of reads showing some degree of methylation. This may indicate that epigenetic shifts at these loci within colonic tissue precede tumorigenesis, as has been observed in several other tissues, as shown in Figure 13. The distribution of read-level methylation is identical for all cancer type-specific CGIs, regardless of whether they are also ExE hyper CGIs. Therefore, targeting ExE hyper CGIs is a conserved feature across human cancer types, but the degree to which they are methylated may be system-specific. Figure 12CData from ENCODE samples are provided, reflecting the embryonic and extraembryonic concordance of humans compared to the well-characterized human cancer cell line HCT116. The human ES cell line HUES64, representative of the pluripotent epiblast, shows significant enrichment of both repressive PRC2 deposition H3K27me3 and activating H3K4me3 modifications at orthologous ExE hyper CGIs. Conversely, the human placenta shows attenuated enrichment of both modifications at these regions, as does HCT116. As shown in Figures 4 and 13, both systems exhibit significant methylation beyond that of the ExE hyper CGI. As a control, the 'ExE hypo' CGI uniformly displays high H3K4me3 levels. Enrichment density heatmaps of the entire ExE hyper CGI set are provided, ranked on the graph according to their H3K27me3 enrichment in HUES64. Normalized enrichment represents the fold enrichment of chromatin immunoprecipitation against sample-matched whole-cell extracts (WCE). Figure 12D Mean methylation boxplots of 489 ExE methylation orthologous CGIs (ExE hyper CGIs) in 14 tissue-matched TCGA cancer types showing a panoramic dysregulation of DNA methylation are provided. Asterisks: CLL samples measured by RRBS (n = 119) and indicate comparisons between age-matched healthy B lymphocytes (n = 24). Edges are at the 25th and 75th percentiles, respectively, and whiskers are at the 2.5th and 97.5th percentiles, respectively. Figure 12E Absolute methylation values ​​are provided for CGIs of all orthologous mappings for the TCGA dataset and CLL boxplots, CGIs methylated across all cancer types, and CGIs specifically methylated in mouse ExEs. ExE hyper CGIs are particularly enriched in all 15 malignancies exhibiting general overall hypomethylation and CGI methylation as part of their somatic origin, exceeding the enrichment observed for hypermethylated CGIs in any given cancer type. Figure 12F Box plots are provided for the same data on cancer type-specific CGI DMR and CGI DMR that are also methylated in mouse ExEs. Notably, the degree of hyper CGI methylation in mouse ExEs reflects the cancer type, with some exhibiting higher absolute methylation values ​​than others. However, in 14 out of 15 cases, the absolute methylation status of cancer type-specific CGI DMR and CGI DMR that are also methylated in ExEs were almost identical, and often slightly higher. Therefore, the absolute methylation value appears to be determined by a specific cancer or cancer type, while targeting extraembryonic methylated CGIs is a general characteristic.

[0032] Figures 13A-13BIt demonstrates the broad conservation of extraembryonic methylation patterns across cancer types and cell lines. Figure 13A Box plots of orthologous ExE hyperCGI from 107 ENCODE / Roadmap epigenomics project samples are provided in Figure 4, with noteworthy additional features for each sample highlighted below. Human extraembryonic tissues, including trophoblastic cell lines and the proplacenta, also share conserved CGI methylation with mice. Normal tissues exhibiting high mean methylation of ExE hyperCGI include many endoderm lineages, such as the colonic mucosa, stomach, and liver (mean methylation values ​​of 0.225, 0.185, and 0.179, respectively), and mature cell types of the adaptive immune system, such as CD8+ and CD4+ T lymphocytes and B lymphocytes (mean methylation values ​​of 0.199, 0.173, and 0.173, respectively). In contrast, ectoderm and epidermal cells show lower methylation levels than other somatic tissues, although cancer cell lines and primary tumors derived from these tissues remain sensitive to hypermethylation. Figure 13B The genome browser tracks of orthologous loci initially presented for mouse development are shown in Figure 1 for three germ layers: human fetal tissue, human primary B lymphocytes, and CLL samples, representing each germ layer (brain, ectoderm; heart, mesoderm; stomach, endoderm). The CGIs surrounding these loci are preserved in a hypomethylated state during embryonic development, where the bimodal structure of the DNA methylation panorama is clearly maintained. In B lymphocytes, some low-level invasive methylation has significantly extended beyond developing hypomethylated regions, such as... Figure 13A Observed in the Roadmap samples. However, in the transition to CLL, extensive methylation was observed on these CGIs, although the methylation values ​​in the surrounding areas decreased. The red lines and shaded areas reflect the local mean and standard deviation calculated by local regression (LOESS) to compensate for the larger number of CpGs in human orthologs compared to mice, which complicates the visual estimation of local methylation at these scales. CGIs are highlighted in green.

[0033] Figure 14A-14F This study reveals the genetic characteristics of ExE CGI methylation in cancer. Figure 14A The results showed low-methylation CGI in 14 TCGA cancer types and CLL, such as Figure 4D The cross-crossing analysis shown indicates that the cancer-related hypomethylated CGIs exhibited an overall loss of methylation in tandem with hypermethylated CGIs. Generally, CGI hypomethylation is more specific, resulting in an exponential decay of cross-cancer cross-crossings. Notably, even for hypomethylated CGIs, those CGIs that are also hypomethylated in mouse ExE, human placenta, or both (conserved) still showed higher cross-cancer cross-crossings. Figure 14BCross-analysis of cancer-dysregulated genes across TCGA cancer types is presented. Among genes significantly dysregulated in at least n (0–14) TCGA cancer types, the fractions of functionally related genes associated with ExE hyperCGI were predicted via GRAIL using a total gene network established through text mining. The FDR was set at 5% as the cutoff value. As the number of TCGA cancer types increased, the fractions of downregulated ExE hyperCGI-related genes within the set generally increased, while the fractions of upregulated genes decreased significantly. Figure 14C Box plots of mean methylation of 489 orthologous ExE hyper CGI genes from 10,629 existing cancers in TCGA are provided, with matched mutation and methylation data, separated according to the mutation status of genes as part of the FGF signaling pathway. Overall, the median mean ExE hyper CGI methylation level was 0.328 in cancers with FGF pathway mutations, compared to 0.275 in cancers without FGF pathway mutations (P < 10⁻¹⁶, rank-sum test). Margins are at the 25th and 75th percentiles, and whiskers are at the 2.5th and 97.5th percentiles, respectively. Figure 14D As predicted by GRAIL, among the 539 genes present in the top 10 recurrent mutation pathways in cancer, 68 are functionally associated with ExE hyper CGI-related genes (FDR < 5%). Genes in the FGF signaling pathway are highlighted in red. Typically, FGF signaling pathway genes have high connectivity scores with ExE hyper CGI-related genes (enrichment z-score of FGF pathway members = 3.88 within the p-value distribution of all 539 genes). Figure 14E Statistical enrichment of FGF pathway genes with a marked amplification or deletion bias in the TCGA database is provided, indicating the typically carcinogenic nature of this pathway in tumorigenesis. Figure 14FThe methylation status of ExE hyper CGI is shown in colonic and hematopoietic mouse cancer models with disrupted de novo methyltransferase activity. All samples were measured by RRBS. Datasets included: primary colonic tissue overexpressing Dnmt3b (promoter methylation status reported, reference 63); genetic models of acute myeloid leukemia (AML), including models transformed by MLL-AF9 fusion (reference 64), cMyc and BCL2 overexpression (reference 64), and FLT3 intratandem repeat (FLT3-IDT, reference 65); and acute and chronic lymphocytic leukemia models driven solely by Dnmt3a gene knockout (references 66 and 67). Methylation of ExE hyper CGI was observed in both colonic Dnmt3b overexpression and hematopoietic Dnmt3a knockout. Other oncogenic driver genes appear to induce de novo methylation of these regions in the presence or absence of DNMT3 expression, suggesting numerous potential pathways to achieve the same molecular phenotype. Wild-type hematopoietic tissue is included for reference and is taken from references 66 and 67. Margins are at the 25th and 75th percentiles, and whisker lines are at the 2.5th and 97.5th percentiles, respectively.

[0034] Figures 15A-15C This study demonstrates that haplotype-level methylation increases the signal-to-noise ratio in the detection of circulating tumor DNA (ctDNA). Figure 15A Methylated reads can be classified into consistent unmethylated reads, inconsistent reads, and consistent methylated reads. Figure 15B Hypermethylated CpG islands were shown in the extraembryonic ectoderm (ExE), and significant methylation (5%) was still observed in the epiblast (Epi) when average methylation was used. Conversely, the background noise was significantly reduced when the proportion of fully methylated reads (PMR) was applied. Figure 15C ).

[0035] Figures 16A-16D This demonstrates the detection of low-frequency tumors (0.01%) from cell-free DNA (cfDNA) through simulation. Simulated sequencing reads were obtained from tumor-like ExE tissue and normal Epi tissue. To detect ctDNA from early-stage cancer patients, the fractionation of tumor-like reads covered 1% (…). Figure 16A ), 0.1% Figure 16B ) and 0.01% Figure 16C It also includes a negative control, which does not simulate tumor-like reads. Figure 16D The p-values ​​for detection based on the rank-sum test are shown on the Y-axis. Each figure shows simulated coverage, guaranteeing 5x coverage of tumor-like DNA.

[0036] Figures 17A-17B It provides a genetic model of cancer. Figure 17AIt shows the evolution of cancer from normal tissue to metastatic cancer. Figure 17B Many genes that may mutate in lung adenocarcinoma were identified.

[0037] Figures 18A-18D This paper presents a framework for developing a universal cancer diagnostic tool using methylation signatures. Cancer cells can differentiate from a single genotype and become more unique. Figure 18A For example, p53 mutations can be detected in tumors of varying fractions across many cancer types. Figure 18B Cancer cells will gradually converge towards a common state. Figure 18C Methylation signatures can be detected in a large proportion of tumors across various cancer types. Figure 18D ).

[0038] Figure 19 This study demonstrates precise molecular detection using novel methylation signatures to predict tissue of cancer origin. The search space was identified (left panel), and the sensitivity / specificity of detection across tissue systems using novel methylation signatures is presented (right panel).

[0039] Figures 20A-20E A method for targeting common molecular pathways was demonstrated. Figure 20A and Figure 20B It shows the involvement of cancer-like states ( Figure 20A ) and somatic cell state ( Figure 20B The pathways and factors involved. A common problem with targeting common molecular pathways is that most molecular targets are shared between cancer states and somatic states, and depend on differential sensitivity (…). Figure 20C Solutions targeting molecular pathways include identifying target-specific coordinators, inducers, or proliferators, as well as cancer-specific interactions. Figure 20D ). Figure 20E A method is provided for identifying targets of molecular pathways involved in cancer-like states, including single-animal screening, perturb-seq, identification of candidates, and docking sites of inhibitory / targeting misdirection co-regulators. Detailed Implementation

[0040] This paper discloses methods for quantifying DNA methylation, which can be used to screen for diseases (e.g., cancer), diagnose diseases (e.g., cancer type), monitor disease progression, and monitor response to treatment regimens. It also discloses a platform for developing early, non-invasive diagnostics that informs novel treatment modalities. Furthermore, it discloses methods for early detection of highly predictive epigenomic alterations, including genome-wide dysregulations of developmental gene promoters, and optimized diagnostic methods for precise detection at ppm resolution. Finally, it discloses the development logic for novel molecular therapies, including expanding the therapeutic window by targeting the unique characteristics of pan-cancer “cellular states.”

[0041] When used in this article, “CpG” and “CpG dinucleotide” are used interchangeably, referring to a dinucleotide sequence containing adjacent guanine and cytosine, wherein cytosine is located at the 5' of guanine.

[0042] When used in this paper, "CpG island" or "CGI" refers to a region with a high frequency of CpG sites. The region is at least 200 bp in size, has a GC percentage greater than 50%, and the observed-to-expected CpG ratio is greater than 60%.

[0043] When used in this article, "haplotype" refers to a combination of CpG sites found on the same chromosome. Similarly, "DNA methylation haplotype" refers to the DNA methylation status of CpG sites on the same chromosome.

[0044] In some embodiments, a screening sample (e.g., a fluid sample) is used. The sample may be screened using whole-genome bisulfite sequencing (WGBS), TCGA Illumina Infinium Human Methylation 450K BeadChip sequencing (TCGA), and / or degenerate representative bisulfite sequencing (RRBS), or by other suitable methylation assays known in the art. The identified methylated sequences may be analyzed to identify differentially methylated loci and / or regions (e.g., CpG islands). In some aspects described herein, for WGBS data, a CGI is considered differentially methylated if it is covered by at least five CpGs and 80% of them are significantly hypermethylated / hypomethylated. In some aspects described herein, for TCGA data, a CGI is considered differentially methylated if 80% of the covered CpGs are significantly hypermethylated or hypomethylated. In some aspects described herein, for RRBS data, a cutoff value of 10% difference in CGI level methylation is used to identify differential methylation. In some areas, RNA sequencing (RNA-seq) and / or transposase-accessible chromatin assays (ATAC-seq) using sequencing are also used to screen samples.

[0045] In some implementations, DNA methylation haplotypes corresponding to the methylation patterns of CpG are identified from the screening. DNA methylation haplotypes can be classified into three categories: uniformly unmethylated haplotypes, disordered haplotypes, and uniformly methylated haplotypes. Haplotypes are also referred to herein as sequencing reads. In some aspects, the proportion of uniformly unmethylated reads (PUR), the proportion of disordered reads (PDR), and the proportion of uniformly methylated reads (PMR) are calculated. As described herein, PMR can be used to quantify DNA methylation (e.g., for diagnostic purposes).

[0046] In some embodiments, the invention disclosed herein relates to a method for detecting circulating tumor DNA (ctDNA) in a sample using the proportion (PMR) of uniformly methylated reads (i.e., fully methylated haplotypes). In some aspects, a methylated sequence of the sample is obtained, and at least one CpG island (CGI) is identified on said methylated sequence. The PMR of the identified CpG island is calculated and then compared to a control background of normal tissue or epiblast. When the PMR of the sample is greater than the control background (e.g., a high signal via a bank sum test), the presence of ctDNA is detected in the sample.

[0047] In some cases, samples are selected from plasma, urine, stool, menstrual fluid, lymph, or any other bodily fluid where ctDNA may be present. Samples may contain DNA (e.g., cell-free DNA (cfDNA)). In some cases, samples are derived from tumors. It is generally understood that the fraction of ctDNA (e.g., cfDNA) in a sample is typically low. In some cases, the background noise for detecting ctDNA in a sample can be reduced by using fully methylated haplotypes.

[0048] The presence of ctDNA can be detected in cfDNA with higher sensitivity and specificity than methods previously known to those skilled in the art. For example, PMR can be used to detect ctDNA in a sample with a sensitivity greater than 75%, 80%, 85%, 90%, 95%, or 99%. In some aspects, PMR is used to detect ctDNA in a sample with 100% sensitivity. PMR can be used to detect ctDNA in a sample with a specificity greater than 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95%. In some aspects, PMR is used to detect ctDNA in a sample with 95% specificity. In some aspects, PMR is used to detect ctDNA in a sample with at least 90% sensitivity and at least 90% specificity. In some aspects, PMR is used to detect ctDNA in a sample with at least 100% sensitivity and at least 95% specificity.

[0049] The amount of ctDNA detected in a sample can be measured and quantified. In some embodiments, the sample contains 0.005% to 1.5% ctDNA, 0.01% to 1% ctDNA, 0.05% to 0.5% ctDNA, or 0.1% to 0.3% ctDNA. In some implementations, the sample contains 0.01% ctDNA. In some aspects, the presence of 0.01% ctDNA in cfDNA is detected using PMR with approximately 100% sensitivity and approximately 95% specificity, with a p-value cutoff of 10. -4 .

[0050] In some implementations, the presence of ctDNA in a sample indicates the presence of cancer. In some aspects, the presence of ctDNA indicates the presence of a tumor. Alternatively, the sample is obtained from an individual without a tumor. For example, the sample may be obtained from an individual in the early stages of cancer but without a tumor, or from an individual with blood cancer (e.g., leukemia). In some aspects, the sample may be obtained from an individual diagnosed with cancer, having cancer, at risk of developing cancer, or suspected of having cancer.

[0051] When used in this article, the phrase "cancer" is intended to be used broadly to refer to any cancerous disease. In some respects, cancer is selected from glioblastoma, colon cancer, lung cancer, breast cancer, and prostate cancer. In other respects, cancer is selected from urothelial carcinoma of the bladder, invasive breast cancer, colon adenocarcinoma, colorectal adenocarcinoma, esophageal cancer, head and neck squamous cell carcinoma, clear cell renal carcinoma, papillary renal carcinoma, hepatocellular carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, prostate adenocarcinoma, stomach and esophageal cancer, thyroid cancer, endometrial cancer, and chronic lymphocytic leukemia.

[0052] In some embodiments, the invention disclosed herein relates to a method for screening for cancer by detecting ctDNA in a sample using PMR as described herein, wherein the presence of ctDNA in the sample indicates that the subject has cancer.

[0053] The methods described herein can be applied to subjects at risk of developing or recurring cancer. Subjects at risk of developing cancer can be, for example, those who have not been diagnosed with cancer but have an increased risk of developing it. Determining whether a subject is considered to be at “increased risk” for cancer is within the skill range of a general practitioner. Any appropriate tests and / or criteria may be used. For example, a subject may be considered to have an “increased risk” of developing cancer if any one or more of the following apply: (i) the subject has a genetic mutation or polymorphism that is associated with an increased risk of developing or suffering from cancer relative to other members of the general population who do not have such a mutation or polymorphism (e.g., genetic mutations in certain TSGs are known to be associated with an increased risk of cancer); (ii) the subject has a certain gene or protein expression profile, and / or a specific substance is present in a sample (e.g., blood) obtained from the subject that is associated with an increased risk of developing or suffering from cancer relative to the general population; (iii) the subject has one or more risk factors, such as a family history of cancer, exposure to tumor promoters or carcinogens (e.g., physical carcinogens, such as ultraviolet or ionizing radiation; chemical carcinogens, such as asbestos, tobacco or smoke components, aflatoxin, arsenic; biological carcinogens, such as certain viruses or parasites); (iv) the subject is over a specified age, such as being over 60 years of age. Subjects suspected of having cancer can be those who have one or more symptoms of cancer or who have undergone recommended diagnostic procedures or procedures consistent with possible cancer. Subjects at risk of cancer recurrence can be those who have received cancer treatment and appear to be cancer-free (e.g., as assessed by appropriate methods).

[0054] In other embodiments, the present invention provides a method for treating a subject in need of cancer treatment. As described herein, ctDNA in a sample is detected using PMR, wherein the presence of ctDNA indicates that the subject has cancer. The individual is then treated for cancer using any treatment method (e.g., a therapeutic agent or procedure) generally known to those skilled in the art.

[0055] For example, therapies or anticancer agents that can be used to treat patients include anticancer agents, chemotherapy drugs, surgery, radiotherapy (e.g., gamma rays, neutron beam radiotherapy, electron beam radiotherapy, proton therapy, brachytherapy, and whole-body radioisotopes), endocrine therapy, biological response modifiers (e.g., interferon, interleukin), hyperthermia, cryotherapy, drugs that reduce any adverse effects, or combinations thereof. Non-limiting examples of cancer chemotherapy agents that can be used include, for example, alkylating and alkyl-like agents, such as nitrogen mustard gas (e.g., chlorambucil, nitrogen mustard, cyclophosphamide, ifosfamide, and melphalan), nitrosoureas (e.g., carmustine, formustine, lomustine, streptozocin); platinum preparations (e.g., alkyl-like agents such as carboplatin, cisplatin, oxaliplatin, BBR3464, saplatin), busulfan, dacarbazine, procarbazine, temozolomide, thiotemozolomide, etc. Piperazine, tromethorphan, and uracil nitrogen mustard; antimetabolites such as folic acid derivatives (e.g., aminopterin, methotrexate, pemetrexed, raltitrexed); purine derivatives such as cladribine, clofarabine, fludarabine, mercaptopurine, pentostatin, thioguanine; pyrimidine derivatives such as capecitabine, cytarabine, fluorouracil, fluorouracil, gemcitabine; spindle toxins / mitotic inhibitors such as taxanes (e.g., docetaxel, paclitaxel), periwinkle derivatives (e.g., periwinkle) Bases, vincristine, vinorelbine, and vinorelbine); epothilones; cytotoxic / antitumor antibiotics, such as anthracyclines (e.g., daunorubicin, doxorubicin, epirubicin, edabricin, mitoxantrone, pixantrone, and pentorubicin); various naturally occurring streptomycin compounds (e.g., actinomycin, bleomycin, mitomycin, purcamycin); and hydroxyurea; topoisomerase inhibitors, such as camptothecin derivatives (e.g., camptothecin, vinorelbine ... Topotecan, irinotecan) and podophyllotoxins (e.g., etoposide, teniposide); monoclonal antibodies used in cancer therapy, such as anti-receptor tyrosine kinases (e.g., cetuximab, panitumumab, trastuzumab), anti-CD20 (e.g., rituximab and tosimomab), and others such as alemtuzumab, aevacizumab, gemtuzumab; photosensitizers, such as aminolevulinic acid, methyl aminolevulinate, porphyrin sodium, and verteporfen;Tyrosine and / or serine / threonine kinase inhibitors, such as Abl, Kit, insulin receptor family members, VEGF receptor family members, EGF receptor family members, PDGF receptor family members, FGF receptor family members, mTOR, Raf kinase family, phosphatidylinositol (PI) kinases such as PI3 kinase, PI kinase-like kinase family members, cyclin-dependent kinase (CDK) family members, and Aurora kinase family members (e.g., marketed inhibitors or inhibitors that have shown efficacy in at least one phase III clinical trial for oncology, such as cidinibub, crizotinib, dasatinib, erlotinib, gefitinib, imatinib, lapatinib, nilotinib, sorafenib, sunitinib). The list includes, but is not limited to, nifedipine (Vandetanib), growth factor receptor antagonists; and others such as retinoids (e.g., avitacic acid and retinoic acid), hexamethylmelamine, acridine, anagrelide, arsenic trioxide, asparaginases (e.g., pegaspargase), besalodin, bortezomib, denitroglycerin-interleukin-toxin conjugates, estmustine, ixapirone, masorofol, mitotane and testrolide, Hsp90 inhibitors, proteasome inhibitors (e.g., bortezomib), angiogenesis inhibitors (e.g., anti-vascular endothelial growth factor agents such as bevacizumab (Avastin) or VEGF receptor antagonists), matrix metalloproteinase inhibitors, various pro-apoptotic agents (e.g., apoptosis inducers), Ras inhibitors, anti-inflammatory agents, cancer vaccines or other immunomodulatory therapies, etc. It should be understood that the foregoing classification is non-limiting.

[0056] The present invention also provides a method for monitoring a subject's response to cancer treatment, the method comprising: detecting the amount of ctDNA in a first sample obtained before the subject receives cancer treatment using a PMR (Prognostic Method of Analysis) of the first sample; detecting the amount of ctDNA in a second sample obtained after the subject receives cancer treatment using a PMR of the second sample; and comparing the amount of ctDNA detected in the first sample with the amount of ctDNA detected in the second sample. In some aspects, the amount of ctDNA detected in the second sample will be less than the amount of ctDNA detected in the first sample, indicating a positive response to cancer treatment (e.g., treatment effectiveness). Alternatively, the amount of ctDNA detected in the second sample will be greater than or equal to the amount of ctDNA detected in the first sample, indicating a negative or neutral response to cancer treatment (e.g., treatment ineffectiveness).

[0057] In other embodiments, the present invention provides a method for monitoring the progression or improvement of cancer in a subject. The method includes identifying ctDNA from the subject's cfDNA using PMR as described herein, wherein the presence of ctDNA indicates a risk of developing cancer, and monitoring the amount of ctDNA in the cfDNA over time, wherein changes in the amount of ctDNA in the cfDNA indicate cancer progression or improvement.

[0058] In other embodiments, the present invention provides a method for assessing cancer in a subject, the method comprising identifying the presence of ctDNA from the subject's cfDNA using PMR as described herein, wherein if ctDNA is present, the subject has cancer or is at risk of developing cancer.

[0059] In a further embodiment, the present invention provides a method for identifying DNA methylation signatures of an individual cancer type. For example, for a specific cancer type, a CpG methylation pattern representing a single DNA methylation haplotype can be quantified. Examples of cancer types include, but are not limited to, colon cancer, lung cancer, lung (squamous cell) carcinoma, breast cancer, prostate cancer, glioblastoma, bladder cancer, esophageal cancer, head and neck cancer, kidney (clear cell) carcinoma, kidney (papillary cell) carcinoma, liver cancer, and uterine (body) cancer.

[0060] In some aspects, the present invention uses DNA methylation signatures to provide predictions of the tissue of origin for cancer. For example, methylation markers can be detected with sensitivity and specificity in a variety of tissue systems. Examples of such tissues include, but are not limited to, adrenal glands, B cells, bladder, bone / soft tissue, brain, breast, cervix, colon, eye, germ cells, head and neck, kidney, liver, lung, bone marrow, mesothelial cell line, neuroendocrine system, pancreas, prostate, skin, stomach, thymus, and uterus.

[0061] The present invention also provides methods for identifying therapeutic targets within one or more molecular pathways. In some aspects, the molecular pathways are common between somatic and cancer-like states. In some aspects, identifying therapeutic targets includes single-animal screening, perturb-seq, candidate identification, and docking sites for inhibiting and / or targeting misdirection co-regulators.

[0062] This invention also provides a method for disrupting the methylation of CpG islands, the method comprising reducing PRC2 expression. Eed The present invention further provides a method for disrupting the methylation of CpG islands, the method comprising reducing... Dnmtl , Dnmt3l or Dnmt3b Expression. In some respects, expression can be reduced through genome modification (e.g., using CRISPR / Cas or TALEN systems).

[0063] The CRISPR / Cas system can utilize various Cas proteins (Haft et al., PLoS Comput Biol. 2005; 1(6)e60). In some embodiments, the CRISPR / Cas system is a CRISPR type I system. In some embodiments, the CRISPR / Cas system is a CRISPR type II system. In some embodiments, the CRISPR / Cas system is a CRISPR type V system. It should be understood that although examples of methods utilizing CRISPR / Cas (e.g., Cas9 and Cpf1) and TALEN are described in detail herein, the invention is not limited to the use of these methods / systems. Other methods known to those skilled in the art for targeting polynucleotide sequences to reduce or eliminate expression in target cells may be utilized herein.

[0064] This invention is contemplated for any purpose, but particularly for reducing or eliminating the expression or activity of the encoded product, by altering, for example, modifying or cleaving a target polynucleotide sequence in cells. In some embodiments, the alteration results in a reduction in the expression of the target polynucleotide sequence. The terms “reduction” and “lowering” are generally used herein to mean a reduction of a statistically significant amount. However, to avoid doubt, “reduction” or “lowering” includes a reduction of at least 10% compared to a reference level, such as a reduction of at least about 20%, or at least about 30%, or at least about 40%, or at least about 50%, or at least about 60%, or at least about 70%, or at least about 80%, or at least about 90%, or up to and including a reduction of 100% (i.e., no level compared to the reference sample), or any reduction between 10% and 100% compared to the reference level.

[0065] It should be understood that the CRISPR / Cas system can cleave the target polynucleotide sequence in a variety of ways. In some embodiments, cleavage of the target polynucleotide sequence results in a double-strand break. In some embodiments, cleavage of the target polynucleotide sequence results in a single-strand break.

[0066] In some embodiments, the CRISPR / Cas system includes a Cas protein or a nucleic acid sequence encoding a Cas protein, and at least one or two ribonucleic acids (e.g., gRNA) capable of guiding and hybridizing the Cas protein to and from a target motif of a target polynucleotide sequence. In some embodiments, the CRISPR / Cas system includes a Cas protein or a nucleic acid sequence encoding a Cas protein, and a single ribonucleic acid or at least one pair of ribonucleic acids (e.g., gRNA) capable of guiding and hybridizing the Cas protein to and from a target motif of a target polynucleotide sequence. As used herein, the terms "protein" and "peptide" are used interchangeably to refer to a series of amino acid residues linked by peptide bonds (i.e., a polymer of amino acids) and include modified amino acids (e.g., phosphorylated, glycosylated, etc.) and amino acid analogs. Exemplary peptides or proteins include gene products, naturally occurring proteins, homologues, parahomologues, fragments and other equivalents, variants, and analogs thereof.

[0067] In some embodiments, the Cas protein comprises one or more amino acid substitutions or modifications. In some embodiments, the one or more amino acid substitutions include conserved amino acid substitutions. In some cases, substitution and / or modification can prevent or reduce protein hydrolytic degradation and / or prolong the half-life of the peptide in the cell. In some embodiments, the Cas protein may comprise peptide bond substitutes (e.g., urea, thiourea, carbamate, sulfonylurea, etc.). In some embodiments, the Cas protein may comprise naturally occurring amino acids. In some embodiments, the Cas protein may comprise alternative amino acids (e.g., D-amino acids, β-amino acids, homocysteine, phosphoserine, etc.). In some embodiments, the Cas protein may comprise modifications to include a portion (e.g., PEGylation, glycosylation, lipidation, acetylation, end-capping, etc.).

[0068] In some embodiments, the Cas protein comprises a core Cas protein. Exemplary Cas core proteins include, but are not limited to, Cas1, Cas2, Cas3, Cas4, Cas5, Cas6, Cas7, Cas8, and Cas9. In some embodiments, the Cas protein comprises an E. coli subtype Cas protein (also referred to as CASS2). Exemplary E. coli subtype Cas proteins include, but are not limited to, Cse1, Cse2, Cse3, Cse4, and Cas5e. In some embodiments, the Cas protein comprises a Ypest subtype Cas protein (also referred to as CASS3). Exemplary Ypest subtype Cas proteins include, but are not limited to, Csy1, Csy2, Csy3, and Csy4. In some embodiments, the Cas protein comprises an Nmeni subtype Cas protein (also referred to as CASS4). Exemplary Nmeni subtype Cas proteins include, but are not limited to, Csn1 and Csn2. In some embodiments, the Cas protein comprises a Dvulg subtype Cas protein (also referred to as CASS1). Exemplary Dvulg subtype Cas proteins include Csd1, Csd2, and Cas5d. In some embodiments, the Cas protein comprises a Tneap isotype Cas protein (also known as CASS7). Exemplary Tneap isotype Cas proteins include, but are not limited to, Cst1, Cst2, and Cas5t. In some embodiments, the Cas protein comprises a Hmari isotype Cas protein. Exemplary Hmari isotype Cas proteins include, but are not limited to, Csh1, Csh2, and Cas5h. In some embodiments, the Cas protein comprises an Apern isotype Cas protein (also known as CASS5). Exemplary Apern isotype Cas proteins include, but are not limited to, Csa1, Csa2, Csa3, Csa4, Csa5, and Cas5a. In some embodiments, the Cas protein comprises an Mtube isotype Cas protein (also known as CASS6). Exemplary Mtube isotype Cas proteins include, but are not limited to, Csm1, Csm2, Csm3, Csm4, and Csm5. In some embodiments, the Cas protein comprises a RAMP module Cas protein. Exemplary RAMP module Cas proteins include, but are not limited to, Cmr1, Cmr2, Cmr3, Cmr4, Cmr5, and Cmr6.

[0069] In some embodiments, the Cas protein is a Streptococcus pyogenes (Streptococcus pyogenes). Streptococcus pyogenes Cas9 protein or its functional portion. In some embodiments, the Cas protein is a Staphylococcus aureus (Staphylococcus aureus) Staphylococcus aureus Cas9 protein or its functional portion. In some embodiments, the Cas protein is a *Streptococcus thermophilus* (Cas9 protein). Streptococcus thermophilusCas9 protein or its functional portion. In some embodiments, the Cas protein is a *Neisseria meningitidis* (*Neisseria meningitidis*) protein. Neisseria meningitides Cas9 protein or its functional portion. In some embodiments, the Cas protein is *Treponema pallidum* (a type of cauda turcica). Treponema denticola The Cas9 protein or its functional portion thereof. In some embodiments, the Cas9 protein is a Cas9 protein or its functional portion derived from any bacterial species. The Cas9 protein is a member of the type II CRISPR system, which typically includes a trans-coding small RNA (tracrRNA), an endogenous ribonuclease 3 (rnc), and the Cas protein. The Cas9 protein (also known as the CRISPR-associated endonuclease Cas9 / Csn1) is a polypeptide containing 1368 amino acids. Cas9 contains two endonuclease domains, including a RuvC-like domain (residues 7-22, 759-766, and 982-989), which cleaves target DNA that is not complementary to the crRNA; and an HNH nuclease domain (residues 810-872), which cleaves target DNA that is complementary to the crRNA.

[0070] In some embodiments, the Cas protein is the Cpf1 protein or a functional portion thereof. In some embodiments, the Cas protein is Cpf1 or a functional portion thereof from any bacterial species. In some aspects, Cpf1 is *Francisella catarrhalis* (a new culprit). Francisella novicida Cpf1 is a protein of the U112 genus or a functional part thereof. In some respects, Cpf1 is an amino acid of the genus *C.* (cocci). Acidaminococcus sp. Cpf1 is a protein of the BV3L6 family or a functional portion thereof. In some respects, Cpf1 is a component of bacteria in the family Spirulinaceae (…). Lachnospiraceae bacterium ND2006 protein or its functional portion. Cpf1 protein is a member of the type V CRISPR system. Cpf1 protein is a polypeptide containing approximately 1300 amino acids. Cpf1 contains a RuvC-like endonuclease domain. Cpf1 uses a single ribonuclease domain to cleave target DNA in an interleaved pattern. The interleaved DNA double-strand breaks result in 4- or 5-nt 5' overhangs.

[0071] As used herein, a "functional portion" refers to a portion of a peptide that retains the ability to complex with at least one ribonucleic acid (e.g., guide RNA (gRNA)) and cleave a target polynucleotide sequence. In some embodiments, the functional portion comprises a combination of operably linked functional domains of the Cas9 protein, selected from a DNA-binding domain, at least one RNA-binding domain, a helicase domain, and a nuclease domain. In some embodiments, the functional portion comprises a combination of operably linked functional domains of the Cpf1 protein, selected from a DNA-binding domain, at least one RNA-binding domain, a helicase domain, and a nuclease domain. In some embodiments, the functional domains form a complex. In some embodiments, the functional portion of the Cas9 protein comprises a functional portion of a RuvC-like domain. In some embodiments, the functional portion of the Cas9 protein comprises a functional portion of an HNH nuclease domain. In some embodiments, the functional portion of the Cpf1 protein comprises a functional portion of a RuvC-like domain.

[0072] It should be understood that the present invention considers various methods of contacting target polynucleotide sequences with Cas proteins (e.g., Cas9). In some embodiments, exogenous Cas proteins can be introduced into cells in the form of peptides. In some embodiments, Cas proteins can be conjugated or fused to cell-penetrating peptides or cell-penetrating peptides. As used herein, "cell-penetrating peptide" and "cell-penetrating peptide" refer to peptides or peptides that promote molecular uptake into cells, respectively. Cell-penetrating peptides may contain detectable markers.

[0073] In some embodiments, the Cas protein may be conjugated or fused to a charged protein (e.g., a protein with a positive, negative, or overall neutral charge). Such a connection may be covalent. In some embodiments, the Cas protein may be fused to a superpositively charged GFP to significantly increase the Cas protein's ability to penetrate cells (Cronican et al. ACS ChemBiol. 2010; 5(8):747-52). In some embodiments, the Cas protein may be fused to a protein switching domain (PTD) to facilitate its entry into cells. Exemplary PTDs include Tat, oligoarginine, and penetratin. In some embodiments, the Cas9 protein comprises a Cas9 polypeptide fused to a cell-penetrating peptide. In some embodiments, the Cas9 protein comprises a Cas9 polypeptide fused to a PTD. In some embodiments, the Cas9 protein comprises a Cas9 polypeptide fused to a tat domain. In some embodiments, the Cas9 protein comprises a Cas9 polypeptide fused to an oligoarginine domain. In some embodiments, the Cas9 protein comprises a Cas9 polypeptide fused to a penetratin domain. In some embodiments, the Cas9 protein comprises a Cas9 polypeptide fused to a superpositively charged GFP. In some embodiments, the Cpf1 protein comprises a Cpf1 polypeptide fused to a cell-penetrating peptide. In some embodiments, the Cpf1 protein comprises a Cpf1 polypeptide fused to a PTD. In some embodiments, the Cpf1 protein comprises a Cpf1 polypeptide fused to a tat domain. In some embodiments, the Cpf1 protein comprises a Cpf1 polypeptide fused to an oligoarginine domain. In some embodiments, the Cpf1 protein comprises a Cpf1 polypeptide fused to a membrane-penetrating peptide domain. In some embodiments, the Cpf1 protein comprises a Cpf1 polypeptide fused to a superpositively charged GFP.

[0074] In some embodiments, the Cas protein can be introduced into cells containing a target polynucleotide sequence in the form of a nucleic acid encoding a Cas protein (e.g., Cas9 or Cpf1). The process of introducing the nucleic acid into cells can be achieved using any suitable technique. Suitable techniques include calcium phosphate or lipid-mediated transfection, electroporation, and transduction or infection using viral vectors. In some embodiments, the nucleic acid comprises DNA. In some embodiments, the nucleic acid comprises modified DNA as described herein. In some embodiments, the nucleic acid comprises mRNA. In some embodiments, the nucleic acid comprises modified mRNA as described herein (e.g., synthetic modified mRNA).

[0075] In some embodiments, nucleic acids encoding the Cas protein and nucleic acids encoding at least one or both of the aforementioned ribonucleic acids are introduced into cells via viral transduction (e.g., lentivirus transduction).

[0076] In some embodiments, the Cas protein is complexed with one or two ribonucleic acids. In some embodiments, the Cas protein is complexed with two ribonucleic acids. In some embodiments, the Cas protein is complexed with one ribonucleic acid. In some embodiments, the Cas protein is encoded by a modified nucleic acid (e.g., a synthetic modified mRNA) as described herein.

[0077] The method of the present invention considers the use of any ribonucleic acid capable of guiding and hybridizing the Cas protein to a target motif of a target polynucleotide sequence. In some embodiments, at least one of the ribonucleic acids comprises tracrRNA. In some embodiments, at least one of the ribonucleic acids comprises CRISPR RNA (crRNA). In some embodiments, a single ribonucleic acid comprises a guide RNA that guides the Cas protein to and hybridizes with a target motif of a target polynucleotide sequence in the cell. In some embodiments, at least one of the ribonucleic acids comprises a guide RNA that guides the Cas protein to and hybridizes with a target motif of a target polynucleotide sequence in the cell. In some embodiments, one or two ribonucleic acids each comprise a guide RNA that guides the Cas protein to and hybridizes with a target motif of a target polynucleotide sequence in the cell. As those skilled in the art will appreciate, the ribonucleic acids of the present invention can be selected to hybridize with a variety of different target motifs depending on the specific CRISPR / Cas system employed and the sequence of the target polynucleotide. The one or two ribonucleic acids can also be selected to minimize hybridization with nucleic acid sequences other than the target polynucleotide sequence. In some embodiments, the one or two ribonucleic acids hybridize to a target motif that contains at least two mismatches when compared with all other genomic nucleotide sequences in the cell. In some embodiments, the one or two ribonucleic acids hybridize to a target motif that contains at least one mismatch when compared with all other genomic nucleotide sequences in the cell. In some embodiments, the one or two ribonucleic acids are designed to hybridize to a target motif immediately adjacent to a deoxyribonucleic acid motif recognized by the Cas protein. In some embodiments, each of the one or two ribonucleic acids is designed to hybridize to a target motif immediately adjacent to a deoxyribonucleic acid motif recognized by the Cas protein, the deoxyribonucleic acid motif flanked by a mutant allele located between the target motifs.

[0078] It will be readily understood by those skilled in the art that the present invention is highly suitable for achieving the stated objectives and obtaining the stated results and advantages, as well as those inherent therein. The details of the description and embodiments herein are representative of certain implementations and are exemplary, and are not intended to be limiting of the scope of the invention. Modifications and other uses will be apparent to those skilled in the art. These modifications are encompassed within the spirit of the invention. Various substitutions and modifications may be made to the invention disclosed herein without departing from the scope and spirit of the invention, as will be readily apparent to those skilled in the art.

[0079] Unless otherwise expressly indicated, references not specifically indicated in quantity in this specification and claims shall be construed as including a plurality of references. A claim or description including "or" among one or more members of the group is considered to satisfy the requirement that one or more members of the group are present, employed, or otherwise associated in a given product or process, unless otherwise indicated or otherwise apparent from the context. This invention includes embodiments in which exactly one member of the group is present, employed, or otherwise associated in a given product or process. This invention also includes embodiments in which more than one or all members of the group are present, employed, or otherwise associated in a given product or process. Furthermore, it should be understood that this invention provides for all variations, combinations, and substitutions in which one or more limitations, elements, clauses, descriptions, etc., from one or more of the listed claims are incorporated into another claim subordinate to the same basic claim (or, any other related claim), unless otherwise stated or unless it would be obvious to a person skilled in the art that such a contradiction or inconsistency would result. All embodiments described herein are considered applicable to all different aspects of the invention where appropriate. It is also considered that any of the described embodiments or aspects can be freely combined with one or more other such embodiments or aspects where appropriate. When elements are presented as a list, such as in Markush groups or a similar format, it should be understood that each subgroup of elements is also disclosed, and any element can be removed from the group. It should be understood that, generally, when an aspect of the invention is referred to as containing a specific element, feature, etc., certain embodiments or aspects of the invention consist of, or are substantially composed of, such elements, features, etc. For simplicity, not all examples are explicitly and specifically described herein. It should also be understood that any embodiment or aspect of the invention may be explicitly excluded from the claims, regardless of whether such a specific exclusion is described in the specification. For example, any one or more active agents, additives, ingredients, optional pharmaceuticals, organism types, diseases, objects, or combinations thereof may be excluded.

[0080] Where a range is given herein, the invention includes embodiments that include an endpoint, embodiments that exclude two endpoints, and embodiments that include one endpoint and exclude the other. It should be contemplated that both endpoints are included, unless otherwise stated. Furthermore, it should be understood that, unless otherwise stated or clearly apparent from the context to those skilled in the art, values ​​expressed as ranges may take any specific value or subrange within the range, up to one-tenth of the lower limit unit of the range, in different embodiments of the invention, unless the context explicitly specifies otherwise. It should also be understood that where a series of numerical values ​​is stated herein, the invention includes embodiments similarly associated with any intermediate value or range defined by any two values ​​in the series, and the lowest value may be considered the minimum and the highest value may be considered the maximum. Numerical values ​​used herein include values ​​expressed as percentages. For any embodiment of the invention where the numerical value is prefixed with "about" or "approximately," the invention includes embodiments describing precise values. For any embodiment of the invention where the numerical value is not prefixed with "about" or "approximately," the invention includes embodiments where the numerical value is prefixed with "about" or "approximately."

[0081] The terms “approximately” or “about” generally include falling within 1% of the number in any direction (greater or less than the number), or in some embodiments within 5%, or in some embodiments within 10%, unless otherwise stated or otherwise apparent from the context (unless the value will not be allowed to exceed 100% of the possible value). It should be understood that, unless explicitly stated otherwise, in any method claimed herein that includes more than one action, the order of actions of the methods is not necessarily limited to the order of the actions of the described methods, but the invention includes embodiments that limit the order. It should also be understood that, unless otherwise stated or apparent from the context, any product or composition described herein may be considered “isolated.”

[0082] When used herein, the term "comprising" is used to indicate components, methods and their respective elements that are essential to the invention, but is open to including unspecified elements, whether necessary or not.

[0083] When used herein, the term "consistent essentially of..." refers to those elements required for a given embodiment. This term allows for the presence of other elements that do not materially affect the fundamental, novel, or functional features of that embodiment of the invention.

[0084] The term "composed of" refers to the composition, method and its corresponding components as described herein, excluding any elements not described in the description of this embodiment.

[0085] It should be understood that the invention disclosed herein is not limited in its application to the details set forth in the specification or by way of example. The invention includes other embodiments and can be practiced and implemented in various ways. Furthermore, it should be understood that the terms and expressions used herein are for descriptive purposes and should not be considered limiting.

[0086] Although certain components, methods, and measurements of the present invention have been specifically described according to certain embodiments, the following examples are only for illustrating the methods and components of the present invention and are not intended to limit them.

[0087] Example

[0088] Example 1: Epigenetic restriction in extraembryonic lineages reflects somatic cell transformation into cancer.

[0089] In mammals, the canonical somatic DNA methylation panorama is determined during embryonic specialization and is subsequently disrupted in many cancer types. 1–4 However, the underlying mechanisms guiding this genome-scale transformation remain elusive, and there is no clear model for its systematic acquisition or potential development utility. 5,6 Here, we analyzed overall remethylation from mouse preimplantation embryos to the early epiblast and extraembryonic ectoderm. This showed that both states yielded highly differentiated genome distributions and significantly disrupted bimodal, CpG density-dependent methylation in placental progenitor cells. 7,8 The extraembryonic genome is protected at CpG island promoters during hundreds of embryos, particularly those promoters associated with key developmental regulators and orthologously methylated in most human cancer types, including specific de novo methylation. 9 Data suggests that evolutionary innovation in extraembryonic tissues may require a shared option: the inhibition of DNA methylation as an alternative to regulation by polycomb proteins, which respond to extraembryonic cues to coordinate embryonic germ layer formation. 10 Furthermore, it has been confirmed that this decision-making is deterministically made downstream of a promiscuous and frequently oncogenic signaling pathway through novel combinations of epigenetic cofactors. Therefore, methylation of developmental gene promoters during tumorigenesis may reflect the invasive trajectory of innate genes and the spontaneous acquisition of a potential, developmentally encoded epigenetic panorama.

[0090] To compare the evolutionary pattern of the epigenetic panorama during early mammalian development, whole-genome bisulfite sequencing (WGBS) and RNA sequencing (RNA-seq) datasets were generated from mouse pre-compacted 8-cell stage embryos, the inner cell mass (ICM) and trophectoderm of the blastocyst at embryonic day (E) 3.5, and the epiblast and extraembryonic ectoderm (ExE) of the latest stage of pregnancy at E6.5, which are essentially homogeneous and undifferentiated from these progenitor cells. Figure 1A(Figure 5). Overall, the time series captured the expected transition through indifferentively hypomethylated but transcriptionally distinct blastocystic tissue, followed by a significant deviation at implantation, when approximately 80% of the genome became differentially methylated (…). Figure 6A Specifically, the extraembryonic lineage lacks typical bimodality: most CpGs are incompletely methylated compared to the epiblast, while only 1.36% are methylated in the ExE ( Figure 1B , Figure 6B ExE-specific hypomethylated or hypermethylated CpGs are segregated into different genomic compartments based on CpG density and location, preferentially enriching de novo methylation of transcription start sites (TSS) and CpG islands (CGIs) near 5' exons. Figure 1C , Figure 6C-6F Once established, these two alternative panoramas are essentially preserved in embryonic tissue or mid-trimester placenta, respectively. 11,12 ( Figure 6G ).

[0091] Notably, ExE methylated CGI (ExE hyper CGI) frequently overlaps with genes regulated by polycomb repressive complex 2 (PRC2), including master transcription factors that guide germ layer and body axis formation. Figures 7A-7B Although most targeted genes are not yet expressed in the epiblast, ExE-specific promoter methylation is generally associated with repression, including many pluripotency-specific regulators, and simultaneous loss of chromatin accessibility (Figures 7-8). Furthermore, the overall relationship between promoter methylation and gene repression is more pronounced in the ExE than in the epiblast. Figure 8C DNA methylation around these promoters is largely dispersed, with lower methylation in the flanking regions of the ExE than in the epiblast, but the largest increase is specifically observed at the TSS. Figure 1D-1E The ExE hyper CGI only reached a methylation level of ~0.25, but the methylated CpGs were distributed in 80% of the sequencing reads that fell within them, and the median methylation status of each read matched the unphased measurements. Figure 4D The most likely explanation for the consistency between per-molecule methylation and aggregate methylation is the recruitment of de novo methyltransferases across the entire population, followed by a homophasic random gain at individual CpGs, similar to various cancer systems. 13,14 Importantly, even with moderate methylation per CpG, a higher CpG density in the ExE-targeted region always results in a higher local methylation density. Figure 1E ).

[0092] Inhibition overlaps with WNT pathway effectors induced in the proximal epiblast, promoting primitive streak formation. Figure 2AHowever, ExE expresses an alternative WNT protein, which inhibits the fibroblast growth factor (Fgf) promoter via de novo methylation and specifically expresses the receptor for epiblast-secreted factors (…). Figure 2B , Figure 9A Extraembryonic panorama can be deterministically derived from these two major signaling pathways, which are concomitant in many downstream developmental processes and are frequently misregulated in cancer. To investigate this hypothesis, ICM was chosen as a model because it is indiscriminately hypomethylated with trophectoderm and can be cultured independently of FGF, while extraembryonic development rapidly deteriorates. 15 Using a combination of FGF4, the mitogen-activated protein kinase (MAPKK or MEK) inhibitor PD0325901, and the GSK3β inhibitor and WNT agonist CHIR99021 (CHIR), ICMs were cultured under four conditions. Figure 2C , Figure 9B The isolated growth halo was determined by a combination of RNA-seq and degenerate representative bisulfite sequencing (RRBS) (Figure 10, Methods section). Cells cultured in FGF4 plus CHIR gradually differentiated into two separate, morphologically distinct internal and external tissues, which were then independently isolated.

[0093] In combination, PD0325901 and CHIR constitute the "2i" condition, which is the FGF-blocked WNT activation state, maintaining overall hypomethylation in the pre-implantation sample. 16 Alternatively, exogenous FGF may be sufficient to drive genomic and CGI methylation to levels above physiological levels. Figure 2D Surprisingly, when combined with FGF, WNT agonists effectively prevented genomic remethylation but redirected CGI-level methylation to a larger subset of extraembryonic targets. Figure 2E-2F CGI targeting is specific to the exterior of the FGF-plus-CHIR growth halo, which establishes an asymmetric Fgfr2 and Fgf4 expression pattern internally, similar to what occurs in vivo. The specific overlap between in vitro promoters and ExE-methylated CGI promoters appears to reflect a progressive restriction of potential targets in early development: targets common across conditions have early developmental functions and are frequently expressed in ICM and 2i conditions, e.g. Prdm14 Targets methylated in ExE and in FGF plus CHIR, but not in FGF alone, typically include neuroectodermal regulators, such as Otx2 and Pax6 ; and ExE-specific targets are often endoderm-derived and induced by dual FGF and WNT activity, for example FoxA2 and Sox17 ( Figure 9CIt appears that ExE-like overall hypomethylation and CGI methylation can be recapitulated in vitro via WNT and FGF, but target specificity can be modulated to include multiple discrete developmental programs.

[0094] Next, we investigated the construction of epigenetic regulators that specifically execute this transition. Although Dnmt1 and Dnmt3b are expressed in both tissues, Dnmt31 and Dnmt3a isoform 2 are cross-expressed in the ExE or epiblast and are regulated by de novo promoter methylation in the other tissue. Figure 11A-11D The truncated noncatalytic isoform of histone 3-lysine 36 (H3K36) demethylase Kdm2b is expressed during preimplantation and in the exE, while the longer isoform containing the Jumonji demethylase domain is specifically induced in the epiblast. 17 ( Figure 11E In other respects, the expression of epigenetic regulators appears relatively stable between the two tissues at this point, and their specific integration could explain this highly differentiated panorama of assembly. To compare their ability to guide total methylation and CGI methylation, we acutely disrupted zygotic CRISPR-Cas9 injection. Dnmt1 , Dnmt3a , Dnmt3b and Dnmt3l Essential PRC2 components Eed ,as well as Kdm2b (Methodology section). We found that, Dnmt1 , Dnmt3b and Dnmt3l Ablation essentially disrupted the ExE methylome, including at CGI targets, but showed no significant specificity or corresponding expression changes in these regions. Figures 3A-3B , Figure 11F-11H Compared to the epiblast matched with the sample, in Dnmt1 - The near-complete loss of methylation in invalid ExEs indicates diminished de novo activity and a greater dependence on epigenetic maintenance, despite prolonged Dnmt31 expression. Figures 3A-3B ).or, Eed - Ineffective ExE disrupts CGI methylation without affecting overall levels, suggesting that PRC2 may specifically coordinate upstream inhibition of DNMT3B as part of a novel developmental pathway. Figure 3B-3D , Figure 11F-11G Consistently, Eed-ineffective ExEs failed to suppress related genes that were induced to levels similar to those in the sample-matched epiblast. Figure 11H ).

[0095] The data indicate a point in early development where sensitivity to co-occurring growth factors guides the identification of distinct epigenomes not observed during downstream individual development. However, de novo CGI methylation is also a general feature of tissue cultures, cancer cell lines, and primary tumors, indicating potential weaknesses in somatic cells. 5,18 (Figure 4, Figures 12-13). To investigate the possible link to the subsequent re-emergence of this panorama in cancer, orthologous CGI was mapped to compare patient-matched DNA methylation profiles from the Cancer Genome Atlas (TCGA) project, age-matched chronic lymphocytic leukemia (CLL) cohorts, and data from the Encyclopedia of DNA Elements (ENCODE) and Roadmap Epigenomics projects. 14,19–21 Of the 16 cancer types with sufficient normal biopsy samples, 15 showed significant methylation of ExE hyper CGI (exoprotein chromogenic ... Figures 4A-4B When measured as a characteristic measure across patients or when examining changes in CGI levels, the signal is surprisingly strong and distinguishes cancerous tissue from normal tissue. Figure 4B (Figure 12). In at least one cancer type, 84% of ExE hyper CGIs were methylated, and they were more frequently shared as conserved pan-cancer targets. Figures 4C-4D , Figures 14A-14B Some direct and indirect evidence suggests that CGI methylation may be influenced by FGF sensing. For example, matched mutation and methylation analysis of the entire TCGA dataset (n = 10,629 cancers) showed that the mean methylation of ExE hyper CGI increased by 19.3% (from 0.275 to 0.328) when any FGF pathway member was mutated. Figure 14C Similarly, a statistical assessment of the connectivity between ExEhyper CGI and the 10 most mutated pathways in cancer revealed a significant enrichment of FGFR signaling in the disease (enrichment z-score = 3.88). Figure 14D-14E In the broader, but less internally controlled, ENCODE and Roadmap data, cancer and immortalized cell lines were clearly distinguished from primary tissues by their ExE hyper CGI methylation status. Figure 4E It is noteworthy that mature adaptive immune cells and endoderm lineages are typically more susceptible to low-level methylation in these regions, suggesting pre-existing heterogeneity even in normal populations.

[0096] A developmental epigenetic panorama of embryonic human cancer endoderm-exotomy tissues was proposed. This panorama occurs concurrently with the establishment of the first major signaling axis, can be partially guided in vitro by hypomethylated ICMs, and appears to be determined by entirely different regulation by DNMT and related cofactors. Notably, de novo methylation of CGI in the exoendothelial cell (ExE) requires PRC2, suggesting that transient biochemical interactions or upstream effects with DNMT3B determine the ExE state or trigger CGI repression. The coordination of this alternative, and presumably more persistent, repressive mechanism warrants further investigation and shares characteristics with somatic carcinogenesis. Most notably, FGF induction is mediated via RAS / MAPK / ERK signaling, which has broad potential and presumed oncogenic effects in the establishment of the cancer methylome. 22–24 Similarly, ExE showed reduced de novo methylation activity entirely guided by DNMT3B, remarkably similar to high-frequency somatic DNMT3A mutations with DNMT3B-guided CGI methylation during acute myeloid leukemia and myelodysplastic syndromes or colorectal transformation. 25–28 A transgenic mouse cancer model confirmed conserved ExE hyper CGI methylation under similar conditions. Figure 14F The extraembryonic panorama depends on intriguing cues with numerous downstream developmental functions, potentially providing opportunities for spontaneous state transitions that will not result in genetic interference later in development. If so, the likelihood of such a transition may relate to the degree of similarity between a given regulatory network and the regulatory network governing extraembryonic specialization. Whether other morphological and molecular features in placental development that appear to resemble cancer markers29,30—such as immunosuppression, tissue invasion, and angiogenesis—are part of this major epigenetic transition or occur downstream remains to be explored, but will provide a valuable developmental basis for their systematic appearance during transformation.

[0097] Additional details and supplementary information are provided by Smith et al., “Epigentic restriction of extraembryonic lineages mirrors the somatic transition to cancer,” Nature 549, 543-547 (September 28, 2017), which is incorporated herein by reference in its entirety, including extended data and supplementary information.

[0098] Example 2: Haplotype-level methylation significantly reduces background noise in normal cells

[0099] Methylation dysregulation has recently been demonstrated to be frequently observed in cancer. This is one reason why single-CpG-based diagnostics have low sensitivity, as methylation can occur at nearby CpG sites. For example, the overall sensitivity of single-CpG-based diagnostics for SEPT9 in colorectal cancer is only 60%. Furthermore, diagnosis of early-stage cancers (<0.1% ctDNA) requires a background close to zero. However, when measured at a single CpG site, normal cells acquire low levels of methylation (~1%) due to random processes.

[0100] Discovering DNA methylation haplotypes provides a better option for diagnostic purposes. Here, a haplotype refers to a combination of CpG sites found on the same chromosome. Similarly, a DNA methylation haplotype represents the DNA methylation status of CpG sites on the same chromosome. In large-scale bisulfite sequencing, the DNA methylation status of thousands / millions of cells is measured. Although fragments of DNA are sequenced, each fragment is ensured to originate from a single chromosome and a single cell. Therefore, the methylation pattern of CpGs on each fragment represents a single DNA methylation haplotype.

[0101] For example, such as Figure 15A As shown, for a locus with 4 CpG sites, if 14 reads are identified covering the locus, it indicates that 14 DNA methylation haplotypes have been observed. Traditionally, only the fraction of aggregate methylation at these 4 sites is measured as the percentage of methylated CpGs divided by all measured CpGs (in this example, 0.57, 0.43, 0.43, and 0.5). Figure 15A Instead, haplotypes are examined. Specifically, these haplotypes are categorized into three groups: uniformly unmethylated haplotypes, disordered haplotypes, and uniformly methylated haplotypes. Since next-generation sequencing is typically used for DNA methylation, sequencing of readings and haplotypes can be used interchangeably.

[0102] To normalize the total sequencing depth, the proportions of uniformly unmethylated reads (PUR), unordered reads (PDR), and uniformly methylated reads (PMR) were calculated. Therefore, PUR, PDR, and PMR are always in the range of 0 to 1, and PUR + PDR + PMR = 1. Currently, PUR / PDR / PMR are calculated for reads with at least 4 Cp sites and a coverage of at least 20x.

[0103] PUR / PDR / PMR can also be calculated using other parameter settings. For example, with advancements in sequencing technology, longer reads covering more CpGs can be sequenced. Alternatively, using current technology, shorter reads can cover more CpG sites in CpG-dense regions (CpG islands, promoters, enhancers). It can also be used on sequencing reads generated by all sequencing platforms, including Sanger sequencing, next-generation sequencing, and single-molecule sequencing.

[0104] Examination of cancer-like tissue (ExE) and normal-like tissue (Epi) revealed that fully methylated reads / haplotypes were extremely rare in normal cells, thus significantly reducing background noise. Figure 15C Conversely, when average methylation is used with the current method, significant background methylation is observed. Figure 15B ).

[0105] Example 3: Detection of low-frequency tumors (0.01%) from cfDNA using simulation

[0106] It has been shown that ctDNA accounts for only 0.01% to 1% of cfDNA in plasma during the early stages of cancer. This is challenging given traditional methylation analysis methods. This paper establishes a novel method for predicting ctDNA from cfDNA with a resolution up to 0.01%, where five tumor DNA copies are present (Figure 16). Based on the simulation, the presence of 0.01% ctDNA can be predicted with a sensitivity of 100%, a specificity of 95%, and a p-value cutoff of 10. -4 .

[0107] Sequencing reads from tumor-like tissue (ExE) were mixed with reads from normal-like tissue (Epi) at fractions of 1%, 0.1%, and 0.01% to simulate tumor DNA. (Note: Only reads located in CpG islands and differentiated between tumor and normal tissue were sampled (see Example 1)). The 0.01% tumor DNA fraction mimicked the fraction of circulating tumor DNA (ctDNA) in cell-free DNA from early-stage cancer patients. Random loss of tumor reads in the simulation mimicked experimental loss during sample preparation or sequencing. In each simulated sample (a mixture of tumor-like and normal-like reads), the PMR of each CpG island was calculated and compared to the background (pure normal-like tissue). If the signal in the simulated sample was significantly higher (by rank-sum test), it was concluded that the sample contained tumor DNA.

[0108] method

[0109] Data Report

[0110] No statistical methods were used to predetermine the sample size. The experiment was not randomized, and researchers were blinded during the experiment and results evaluation.

[0111] Sample separation and library preparation

[0112] Preimplantation and postimplantation sample preparation was performed as described in Reference 31. Briefly, 5- to 8-week-old B6D2F1 hybrid female mice (Charles River) were induced sequentially with 5 IU of gestational mare gonadotropin (Sigma), followed by 5 IU of human chorionic gonadotropin (Millipore) 46 hours later, and subsequently mated with ≤6-month-old B6D2F1 male mice. For the preimplantation time point, zygotes from mated females were isolated from the oviduct the following morning (E0.5) and cultured in KSOM (Millipore) droplets under mineral oil until E2.25. Eight-cell samples were collected by carefully monitoring 4-cell embryos from ~E2, and emerging 8-cell embryos were exchanged for samples supplemented with 1 μg / ml of [amount missing]. -1 Aphidicolin (Sigma) was administered in KSOM to ensure synchronization and entry into at least the fourth replication cycle. Eight-cell embryos were collected within 4 hours of the first identifiable embryo at this stage. Prior to collection, embryos were serially transferred through acidic Tyrode's solution (Sigma) to remove the zona pellucida and carefully aspirated through a drawn glass capillary with 0.25% trypsin-EDTA (Life Technologies) to remove maternal polar bodies. E3.5 blastocysts were also treated with acidic Tyrode's solution to remove the zona pellucida and dissected the ICM and trophectoderm of matched samples using standard micromanipulation equipment (Eppendorf) and a Hamilton Thorne XYClone laser at 100% intensity with 300 μs pulses. 32 Post-implantation tissue isolation was performed. The decidua of mated female mice was isolated in the morning of E6.5, and the pregnant body was removed. The embryo was then carefully bisected along the ecto-hypocotyl axis under a stereomicroscope, and the chorionic cone was removed from the epidermal ectoderm, if visible. After isolation, the epiblast and exoblast were incubated at 4°C for 15 minutes in 0.5% trypsin and 2.5% trypsin dissolved in PBS, and then allowed to stand in KSOM at room temperature for 5–10 minutes. Finally, the embryo was aspirated through a narrow, flame-drawn glass capillary tube, and the visceral endoderm was removed, collecting only samples without obvious contamination. On average, matched exoblast and epiblast or ICM and trophectoderm samples were collected from 5–10 embryos or from 20 or more 8-cell embryos, according to the assay.

[0113] As previously stated 33DNA for whole-genome bisulfite sequencing was isolated, and libraries were prepared using the Accel-NGS Bisulfite DNA Library Kit (Swift Biosciences) according to the manufacturer's protocol. The final library was generated from 10–12 PCR cycles. RNA was purified using the RNeasy Mini Kit (Qiagen), and RNA-seq libraries were generated using the SMRTseq v4 Ultra-Low Input Kit (Clontech) according to the manufacturer's protocol with 10–11 long-range PCR cycles. Libraries were generated from 150 pg of subsequent cDNA using the Nextera XT DNA Library Preparation Kit (Illumina) and 12 PCR cycles, as described above. 34 An ATAC-seq library was generated by reacting 10 μl of a TN5 transposase mixture (Nextera DNA Library Preparation Kit, Illumina) with the mixture and incubating for 45 minutes. The reaction was stopped according to the previously described protocol 35, and the DNA was purified using silane beads (Thermo Fisher). The tagged DNA was amplified for 12–14 cycles to generate a library. The WGBS library was sequenced as a pool using the HiSeq X 10 platform (Illumina), and the RNA-seq and ATAC-seq data were sequenced using a HiSeq 2500 (Illumina).

[0114] Growth halo experiment

[0115] To generate controlled growth halo data, the intrablastocyst was isolated from the BDF1×129S1 / SvImJ strain blastocyst using an immunosurgical method 96 hours post-fertilization, as described. 31 In short, oocytes were isolated from B6D2F1 females 12–14 hours after administration of human chorionic gonadotropin (hCG), and fertilization was achieved by intracytoplasmic sperm injection of 129S1 / SvImJ sperm via piezoelectric actuation. 36 Ninety-six hours post-fertilization, the blastocyst was deprived of its zona pellucida by brief incubation in acidic Tyrode's solution, followed by 30 minutes of incubation in CO2-equilibrated KSOM with a 1:10 dilution of whole mouse antiserum (Sigma), and then the trophoblast was disrupted by incubation in a 1:10 dilution of guinea pig complement serum (Sigma). After 15 minutes at 37°C, the ICM was separated from the complement-dissolved trophoblast and could be cleanly separated by a brief pulse through a narrow glass capillary. The ICM was separated in batches of ~12 drops / drop. Once separated, the ICM was replenished with 1,000 U / ml of... -1Leukemia suppressor factor (self-made) was plated in basal N2 / B27 medium with one of the following conditions: '2i': Reagents Direct supplemented with 1 μM PD0325901 and 3 μM CHIR99021. 37 ; 'PD0325901': Supplemented with 1 μM PD0325901 and 10 ng / ml -1 BMP4 promotes growth halo expansion (Peprotech) 38 'FGF plus CHIR': Supplement with 25 ng / ml -1 Mouse recombinant FGF4 (R&D systems) and 3 μM CHIR99021; and 'FGF': supplemented with only 25 ng / ml -1 FGF4 was chosen because it is the highest-expressed member of the FGF family in preimplantation embryos, and we sought to guide specific remethylation changes based on in vivo observations. ICMs were placed in gelatin-treated tissue culture dishes coated with irradiated CF-1 line embryonic fibroblasts to promote attachment. After four days of culture, primary growth halos of the ICMs were isolated, characterized by a centrally swollen three-dimensional mass. In all cases except the 2i condition, the outer layer of differentiated cells was distinct and removed using the same strategy as described above for removing visceral endoderm from E6.5 samples. However, under FGF plus CHIR conditions, the “outer layer” was often the same size as or larger than the inner growth halo and only became distinct during the later stages of culture (see [link to relevant documentation]). Figure 9B Therefore, we collected both the internal and external portions, as they could be clearly identified as having a common ICM origin. After incubation and isolation or removal of the external cells, the growth halo was continuously washed with a few drops of KSOM under mineral oil and then flash-frozen in a minimal volume for RNA-seq and RRBS profiling analysis.

[0116] Knockout embryos were generated via zygotic CRISPR-Cas9 injection.

[0117] The zygotic injection was basically performed as described. 39 To improve the efficiency of generating invalid alleles, three separate single guide RNA (sgRNA) sequences were designed for each target to be used with the CHOPCHOP network tool. 40The absence of high-scoring off-target sites and the prioritization of these constraints as much as possible to disrupt the coding frame determined the order of the high-scoring interstitial region (ISR) sequence. The ISR sequence was input into the following oligonucleotide primer pairs and used to amplify the pX300 plasmid (Addgene): forward primer, AGTCAGTTTAATACGACTCACTATAGN19GTTTTAGAGCTAGAAATAGCAAG (SEQ ID NO: 1); reverse primer, AAAAAAAGCACCGACTCGGTGCCAC (SEQ ID NO: 2). An ISR sequence without a G-starting prefix to initiate T7 transcription was inserted, and an additional 5' G was added. Using a 200 ng gel-purified sgRNA template containing the T7 promoter, sgRNA was transcribed in vitro using the MEGAshortscript T7 Transcription Kit (Thermo Fisher), followed by purification with phenol:chloroform and ethanol precipitation. Translationally capable spCas9 RNA was transcribed in vitro from a template driven by a similarly designed T7 promoter using the mMESSAGE mMACHINE T7 Ultra kit (Thermo Fisher), and purified using the RNA Clean and Concentrator kit (Zymo Research). The RNA was resuspended in injection buffer at pH 7.4 containing 5 mM Tris-HCl and 0.1 mM EDTA. Zygotes were isolated from hormone-induced B6D2F1 females mated with B6D2F1 males as described above. Shortly after the formation of a visible pronucleus (pronuclear stage 3), 100 ng μl of cytoplasmic reticulum was injected into the zygote cytoplasm. -1 All three targeting sgRNAs were combined in a 1:1:1 ratio with 200 ng μl -1Cas9 mRNA. At E3.5, vacuolated blastocysts were transferred in groups of 10–15 fertilized eggs to a uterine horn of pseudopregnant CD-1 mice (Charles River) mated two days prior with vasectomized male Swiss-Weber (Taconic) mice. To address a ~1-day shift in developmental progression due to uterine transfer, E6.5-stage pregnant bodies were appropriately isolated 4 days post-transfer, and epiblastic and extraembryonic ectodermal tissues were separated as described above and then flash-frozen in minimal volume. Each replicate consisted of at least four embryos, and all experimental series included replicates generated from at least two rounds of zygote injection. It was noted that each replicate group included epiblastic and extraembryonic ectodermal tissues from matched embryos, and RRBS data for both portions not covering >1 million CpG with ≥5× coverage were excluded from further analysis. Disruption of the target allele was confirmed by PCR amplification from primary cDNA using primers located flanking all three pre-interstitial sequences to capture multiple simultaneous perturbations or truncations in phase.

[0118] Dual RNA-seq and RRBS profiling analysis

[0119] As previously stated and modified 41 Genomic DNA and mRNA were purified from samples with low input volumes. In short, cells were mixed with 15 μl of a solution containing 1 U μl of... -1 SUPERase Mix the RNase inhibitor (ThermoFisher), 1% β-mercaptoethanol (Sigma), and RLT plus buffer (Qiagen) and transfer to one well of a 96-well DNA LoBind plate (Eppendorf). Add 10 μl of M-280 streptavidin bead-conjugated reverse transcription primer to each sample, and incubate the reaction in a thermal cycler at 72°C for 3 minutes, followed by gentle rotation incubation at room temperature for 25 minutes. Genomic DNA and mRNA are then isolated using a DynaMag-96 edge magnet (ThermoFisher). As previously described. 41 Bead-tagged mRNA was reverse transcribed, and the genomic DNA in the supernatant was transferred to a fresh 96-well DNA LoBind plate. Following reverse transcription, the DNA was processed according to the Smart-seq2 protocol. 42 cDNA was amplified by PCR to generate an RNA-seq library. The resulting RNA-seq library was then merged and sequenced using an Illumina Hiseq 2500 sequencer.

[0120] Genomic DNA was isolated using 1x Agencourt AMPure beads (Beckman Coulter) and eluted with 15 μl of low Tris-EDTA buffer. (Based on previous reports with modifications.) 43 RRBS libraries were generated. We used CutSmart buffer (New England Biolabs) for all three enzymatic reactions, including MspI digestion, end repair / A-tailing, and T4 DNA ligation. To minimize DNA loss, the DNA purification step was omitted after each enzymatic reaction. Briefly, genomic DNA was digested at 37°C with 16 units of MspI (New England Biolabs) for 80 minutes, followed by heat inactivation at 65°C for 15 minutes. End repair and A-tailing were performed on the digested DNA fragments by adding 4 units of Klenow fragment (3′ → 5′ exo-) (New England Biolabs), 0.03 mM dCTP, 0.03 mM dGTP, and 0.3 mM dATP; the reactions were carried out at 30°C for 25 minutes and then at 37°C for 25 minutes, followed by inactivation at 70°C for 10 minutes. We then ligated the A-tailed DNA fragment to the indicated adaptor overnight at 16°C by adding 2,000 U of T4 DNA ligase, 0.75 mM ATP, and 7 nM adaptor. The T4 ligase was heat-inactivated at 65°C for 15 min before pooling the libraries. To remove adaptor dimers, the library pool was cleared with 1.8x AMPure beads, and the DNA fragment with the adaptor tag was eluted into 30 μl of low-Tris-EDTA buffer. The DNA fragment with the adaptor tag was bisulfite-converted using the Qiagen EpiTect Rapid Bisulfite Conversion Kit, following the manufacturer's instructions with minor modifications. We extended the bisulfite conversion time from 10 min for 2 cycles to 20 min for 2 cycles to achieve a bisulfite conversion rate >99%. Bisulfite-converted DNA fragments were PCR amplified using the following thermal cycler settings: 98°C for 45 sec, 6 cycles of 98°C for 20 sec, 58°C for 30 sec, 72°C for 1 min, followed by 8–10 cycles of 98°C for 20 sec, 65°C for 30 sec, 72°C for 1 min, and a final extension at 72°C for 5 min. The PCR-amplified library DNA was cleaned using 1.3x AMPure beads, and the RRBS library was sequenced with paired ends for 2 × 100 cycles. Only instances with > 1 million CpG coverage from the epiblast and ExE of a given replicate sample were included for downstream analysis.

[0121] For each sample, prepare 10 μl of M-280 streptavidin beads (ThermoFisher) according to the manufacturer's recommendations. Specifically, after washing sequentially with solutions A (0.1 M NaOH, 0.05 M NaCl) and B (0.1 M NaCl), resuspend the beads in 10 μl of 2x binding and washing buffer (10 mM Tris-HCl, 1 mM EDTA, 2 M NaCl), and then mix with an equal volume of 2 μM reverse transcription primers. 41 The mixture was incubated at room temperature with gentle rotation for 15 minutes. The bead-bound reverse transcription primers were collected using a magnet and then resuspended in 10 μl of binding buffer (10 mM Tris-HCl (pH 8.0), 167 mM NaCl, 0.05% Tween-20).

[0122] Estimating methylation levels

[0123] The methylation level of cytosine in each sample is estimated as the number of report C reads divided by the total number of report C or T reads. The methylation level of a single CpG is limited to those CpGs with at least five-fold coverage. For a 100 bp tile, reads of all CpGs within that tile with more than five-fold coverage are pooled and used to estimate the methylation level as described for a single CpG. The CpG density of a given single CpG is the number of CpGs 50 bp upstream and downstream of that CpG. The CpG density of a 100 bp tile is the number of CpGs within that tile. The reported methylation level of a sample is the average methylation of all reads pooled from replicate samples.

[0124] Genomic characteristics

[0125] LINE, LTR, and SINE annotations were downloaded from the UCSC (University of California, Santa Cruz) browser (mm9) RepeatMasker track. CGI annotations were downloaded from the UCSC browser (mm9) CpGIslands track. Gene annotations (exons, 5' exons, introns) were downloaded from the UCSC browser (mm9) RefSeq track. Promoter (TSS) were defined as ±2 kb of the RefSeq annotation. The corresponding human annotations are the annotations for hg19 downloaded from the UCSC browser. In each case, the methylation level of an individual trait was estimated by averaging the methylation of all CpGs within the trait that covered more than five times. CGIs were assigned to a given TSS (CGI promoter) that included the CGIs of the annotation within this boundary. Methylation was estimated against the 'core TSS' sequence defined as ±1 kb of the RefSeq annotation and only included CpGs measured at ≥5× in either of the two sample types (WGBS) or pooled sample type (RRBS). Figure 2B , Figure 7F and Figure 9C This includes all isoform promoters and reports the most distinct alternative TSSs. Methylation levels for all annotated TSSs are calculated and reported in this manner, along with an estimate of the average transcripts per million (TPM) of the gene reported for all relevant TSSs.

[0126] Identification of differentially methylated loci and regions

[0127] For WGBS data, differentially methylated loci were identified using the DSS package, which uses biological duplicate samples and information from CpG sites in the genome to stabilize dispersion parameters for estimation. 44 For a given comparison, only CpGs covering at least five-fold CpGs across all samples are considered. A 5% false discovery rate (FDR) cutoff is used to identify differentially methylated CpGs. A CGI is defined as differentially methylated if it is covered by at least five CpGs and 80% of them are significantly hypermethylated / hypomethylated. For the TCGA Illumina Infinium HumanMethylation450K BeadChip data, given that most cancer types have more than 20 cancer and normal samples, the Wilcoxon rank-sum test is used to identify differentially methylated CpGs with an FDR cutoff of 5%. All statistical tests throughout the study are two-sided. A CGI is defined as differentially methylated if 80% of the covered CpGs are significantly hypermethylated / hypomethylated. For the RRBS data, a cutoff of 10% difference in CGI level methylation is used to predict differential methylation.

[0128] Gene expression analysis

[0129] TopHat2, used for assembling mm9 from the mouse genome, was used for alignment with default settings. Isotype-level expression was quantified using Kallisto, which performs spurious alignments of cDNA reads from transcripts. Gene-level expression was estimated as the sum of the expression of relevant isotypes. Refseq mRNA sequences were downloaded from the UCSC Genome Explorer. Expression levels are reported as per million transcripts (TPM).

[0130] Pathway enrichment

[0131] Pathway enrichment was performed using the GSEA online tool via hypergeometric tests. P-values ​​were adjusted for multiple hypothesis testing, with a 5% cutoff, based on Benjamini and Hochberg. Regulation via PRC2 in human ES cells was obtained from reference 45.

[0132] Connectivity analysis

[0133] We used GRAIL (gene relationships across implicated loci)46 to examine whether a query gene is functionally related to a set of seed genes. GRAIL uses text mining to quantify the correlation between two genes in the genome to construct a total gene network. It has been shown that genes functioning along the same pathway tend to be distributed in coherent subnetworks. In this study, we constructed a subnetwork using ExE hyper CGI-related genes that were significantly enriched in several pathways. To predict whether a query gene is functionally related to the ExEhyper subnetwork, we projected the gene into the total network and examined whether the connection between the gene and the subnetwork was random or statistically significant.

[0134] ATAC–seq data processing

[0135] Reads were aligned to the mouse genome mm9 using BWA with default parameters. Duplicates were removed using the PicDuplicates function in the Picard toolkit. Reads with low mapping quality (<10) or located on mitochondrial chromosomes were removed. Insert density was generated using NucleoATAC and normalized to the total number of inserts in each sample. 47 .

[0136] Ortholog mapping between humans and mice

[0137] The mouse mm9 CGI was mapped to the human hg19 region using liftOver with the chain file mm9ToHg19.over.chain. The human orthologous CGI was then defined as the CGI closest to the mapped region.

[0138] Data availability

[0139] All datasets are stored in the Gene Expression Omnibus and are accessible under GSE84236. Other data include: Roadmap and ENCODE samples from the RnBeads Methylome Resource (rnbeads.mpiinf.mpg.de / methylomes.php), adult mouse tissues from GSE42836, and CLL and normal B lymphocytes from CSE58889.

[0140]

[0141] References

[0142] 1. Smith, ZD & Meissner, A. DNA methylation: roles in mammalian development. Nat. Rev. Genet. 14, 204–220 (2013).

[0143] 2. Ohm, JE, et al. A stem cell-like chromatin pattern may predispose tumorsuppressor genes to DNA hypermethylation and heritable silencing. Nat. Genet. 39, 237–242 (2007).

[0144] 3. Schlesinger, Y. et al. Polycomb-mediated methylation on Lys27 of histone H3 premarker genes for de novo methylation in cancer. de novo (methylation in cancer). Nat. Genet. 39, 232–236 (2007).

[0145] 4. Widschwendter, M. et al. The role of epigenetic stem cell signature in cancer. Nat. Genet. 39, 157–158 (2007).

[0146] 5. Feinberg, AP, Ohlsson, R. & Henikoff, S. The epigenetic progenitor origin of human cancer. Nat. Rev. Genet. 7, 21–33 (2006).

[0147] 6. Flavahan, WA, Gaskell, E. & Bernstein, BE. Epigenetic plasticity and the hallmarks of cancer. Science 357, eaal2380 (2017).

[0148] 7. Schroeder, DI et al. The human placentamethylome. Proc. Natl Acad. Sci. USA 110, 6037–6042 (2013).

[0149] 8. Branco, MR, et al. Maternal DNA methylation regulates early trophoblast development. Dev.Cell 36, 152–163 (2016).

[0150] 9. Deaton, AM & Bird, A. CpG islands and the regulation of transcription. Genes Dev. 25, 1010–1022 (2011).

[0151] 10. Arnold, SJ & Robertson, EJ: Making a commitment: cell lineage allocation and axispatterning in the early mouse embryo. Nat. Rev. Mol. Cell Biol. 10, 91–103 (2009).

[0152] 11. Hon, GC, et al. Epigenetic memory at embryonic enhancers identified in DNA methylation maps from adult mouse tissues. Nat. Genet. 45, 1198–1206 (2013).

[0153] 12. Ziller, MJ, and others Charting a dynamic DNA methylation landscape of the human genome. Nature 500, 477–481 (2013).

[0154] 13. Landan, G. et al. Epigenetic polymorphism and the stochastic formation of differentially methylated regions in normal and cancerous tissues. Nat. Genet. 44, 1207–1214 (2012).

[0155] 14. Landau, DA, et al. Locally disordered methylation forms the basis of intratumor methylome variation in chronic lymphocytic leukemia. Cancer Cell 26, 813–825 (2014).

[0156] 15. Arman, E., Haffner-Krausz, R., Chen, Y., Heath, JK & Lonai, P. Targeted disruption of fibroblast growth factor (FGF) receptor 2 suggests a role for FGF signaling in pregastrulation mammalian development. Proc. Natl Acad. Sci. USA 95, 5082–5087 (1998).

[0157] 16. Leitch, HG, et al. Naive pluripotency is associated with global DNA hypomethylation. Nat. Struct. Mol. Biol. 20, 311–316 (2013).

[0158] 17. Boulard, M., Edwards, JR & Bestor, TH. Abnormal X chromosome inactivation and sex-specific gene dysregulation after ablation of FBXL10. Epigenet. Chromatin September 22, 2016

[0159] 18. Meissner, A. et al. Genome-scale DNA methylation maps of pluripotent and differentiated cells. Nature 454, 766–770 (2008).

[0160] 19. The ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74 (2012).

[0161] 20. Hoadley, KA et al. Multiplatform analysis of 12 cancer types reveals molecular classification within and across tissues of origin. Cell 158, 929–944 (2014).

[0162] 21. Kundaje, A. et al. Integrative analysis of 111 reference human epigenomes. Nature 518, 317–330 (2015).

[0163] 22. MacLeod, AR, Rouleau, J. & Szyf, M. Regulation of DNA methylation by the Ras signaling pathway. J. Biol. Chem. 270, 11327–11337 (1995).

[0164] 23. Lu, CW et al. Ras-MAPK signaling promotes trophectoderm formation from embryonic stem cells and mouse embryos. Nat. Genet. 40, 921–926 (2008).

[0165] 24. Serra, RW, Fang, M., Park, SM, Hutchinson, L. & Green, MRA. KRAS-directed transcriptional silencing pathway that mediates the CpG island methylator phenotype. eLife 3, e02313 (2014).

[0166] 25. Ley, TJ, and others Acute myeloid leukemia DNMT3A mutation( DNMT3A mutations inacute myeloid leukemia). N. Engl. J. Med. 363, 2424–2433 (2010).

[0167] 26. Walter, MJ, and others Relapsed in patients with myelodysplastic syndrome DNMT3A Mutation (Recurrent) DNMT3A mutations in patients with myelodysplastic syndromes). Leukemia 25, 1153–1158 (2011)

[0168] 27. Rhee, I. et al. DNMT1 and DNMT3b cooperate to silence genes in human cancer cells. Nature 416, 552–556 (2002)

[0169] 28. Lin, H. et al. By deleting Dnmt3b Suppression of intestinal neoplasm by deletion of Dnmt3b ). Mol. Cell. Biol. 26, 2976–2983 (2006).

[0170] 29. Novakovic, B. & Saffery, R. Placental pseudo-malignancy from a DNA methylation perspective: unanswered questions and future directions. Front. Genet 4,285 (2013).

[0171] 30. Hanahan, D. & Weinberg, RA. Hallmarks of cancer: the next generation. Cell 144, 646–674 (2011).

[0172] 31. Smith, ZD et al. DNA methylation dynamics of the human preimplantation embryo. Nature 511, 611–615 (2014).

[0173] 32. Chenoweth, JG & Tesar, PJ. Isolation and maintenance of mouse epiblast stem cells. Methods Mol. Biol .636, 25–44 (2010).

[0174] 33. Smith, ZD et al. A unique regulatory phase of DNA methylation in the early mammalian embryo. Nature 484, 339–344 (2012).

[0175] 34. Buenrostro, JD, Giresi, PG, Zaba, LC, Chang, HY & Greenleaf, WJ. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat. Methods 10, 1213–1218 (2013).

[0176] 35. Lara-Astiaso, D. et al. Immunogenetics. Chromatin state dynamics during blood formation. Science 345, 943–949 (2014).

[0177] 36. Yoshida, N. & Perry, AC. Piezo-actuated mouse intracytoplasmic sperm injection (ICSI). Nat. Protocols 2,296–304 (2007).

[0178] 37. Ying, QL et al. The ground state of embryonic stem cell self-renewal. Nature 453, 519–523 (2008)

[0179] 38. Ying, QL, Nichols, J., Chambers, I. & Smith, A. BMP induction of Id proteins suppresses differentiation and sustains embryonic stem cell self-renewal incollaboration with STAT3. Cell 115, 281–292 (2003).

[0180] 39. Wang, H. et al. One-step generation of mice carrying mutations in multiple genes by CRISPR / Cas-mediated genome engineering. Cell 153, 910–918 (2013).

[0181] 40. Labun, K., Montague, TG, Gagnon, JA, Thyme, SB & Valen, E. CHOPCHOP v2: a web tool for the next generation of CRISPR genome engineering. Nucleic Acids Res. 44, W272–W276 (2016).

[0182] 41. Macaulay, IC et al. G&T-seq: parallel sequencing of single-cell genomes and transcriptomes. Nat. Methods 12, 519–522 (2015).

[0183] 42. Picelli, S. et al. Full-length RNA-seq from single cells using Smart-seq2. Nat. Protocols 9, 171–181 (2014).

[0184] 43. Gu, H. et al. Preparation of reduced representation bisulfite sequencing libraries for genome-scale DNA methylation profiling. Nat. Protocols 6, 468–481 (2011).

[0185] 44. Wu, H. et al. Detection of differentially methylated regions from whole-genome bisulfite sequencing data without replication. Nucleic Acids Res. 43, e141 (2015).

[0186] 45. Ben-Porath, I. et al. An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat. Genet. 40, 499–507 (2008).

[0187] 46. ​​Raychaudhuri, S. et al. Identifying relationships among genomic disease regions: predicting genes at pathogenic SNP associations and rare deletions. PLoS Genet. 5, e1000534 (2009)

[0188] 47. Schep, AN et al. Structured nucleosome fingerprints enable high-resolution mapping of chromatin architecture within regulatory regions. Genome Res. 25, 1757–1770 (2015).

[0189] 48. Ciruna, BG & Rossant, J. T-box gene Eomesodermin Expression of the T-box gene during early mouse development Eomesodermin during early mousedevelopment). Mech. Dev. 81, 199–203 (1999).

[0190] 49. Ralston, A. & Rossant, J. Cdx2 acts downstream of cell polarization to cell-autonomously promote trophectoderm fate in the early mouse embryo. Dev. Biol. 313, 614–629 (2008).

[0191] 50. Savory, JG et al. Cdx2 regulates posterior development through non-Hox targets. Development 136, 4099–4110 (2009).

[0192] 51. Donnison, M. et al. The Elf5 mutant suffers from a loss of the extraembryonic ectoderm, leading to a defect in embryonic pattern formation. Elf5 mutants lead to defects inembryonic patterning). Development 132, 2299–2308 (2005).

[0193] 52. Goldin, SN & Papaioannou, VE. Paracrine action of FGF4 during periimplantation development maintains trophectoderm and primitive endoderm. Genesis 36, 40–47 (2003).

[0194] 53. Kang, M., Piliszek, A., Artus, J. & Hadjantonakis, AK. FGF4 is required for lineage restriction and salt-and-pepper distribution of primitive endoderm factors, but not their initial expression in the mouse. Development 140, 267–279 (2013)

[0195] 54. Nichols, J., Silva, J., Roode, M. & Smith, A. Suppression of Erk signaling promotes ground state pluripotency in the mouse embryo. Development 136, 3215–3222 (2009).

[0196] 55. Auclair, G., Guibert, S., Bender, A. & Weber, M. Ontogeny of CpG island methylation and specificity of DNMT3 methyltransferases during embryonic development in the mouse. Genome Biol. 15,545 (2014).

[0197] 56. Smallwood, SA, et al. Dynamic CpG island methylation landscape in oocytes and preimplantation embryos. Nat. Genet. 43, 811–814 (2011).

[0198] 57. Ooi, SK et al. DNMT3L connects unmethylated lysine 4 of histone H3 to de novo DNA methylation. de novo methylation of DNA). Nature 448, 714–717 (2007).

[0199] 58. He, J. et al. Kdm2b maintains murine embryonic stem cell status by recruiting the PRC1 complex to CpG islands of developmental genes. Nat. Cell Biol. 15, 373–384 (2013).

[0200] 59. Wu, X., Johansen, JV & Helin, K. Fbxl10 / Kdm2b recruits polycomb repressive complex 1 to CpG islands and regulates H2A ubiquitylation. Mol. Cell 49, 1134–1146 (2013).

[0201] 60. Blackledge, NP, et al. Variant PRC1 complex-dependent H2A ubiquitylation drives PRC2 recruitment and polycomb domain formation. Cell 157, 1445–1459 (2014).

[0202] 61. Boulard, M., Edwards, JR & Bestor, TH FBXL10 protects polycomb-bound genes from hypermethylation. Nat. Genet. 47, 479–485 (2015).

[0203] 62. Irizarry, RA et al. The human colon cancer methylome shows similar hypomethylation and hypermethylation at conserved tissue-specific CpG island shores. Nat. Genet. 41, 178–186 (2009)

[0204] 63. Steine, EJ, et al. Genes methylated by DNA methyltransferase 3b are similar in mouse intestinal and human colon cancer. J. Clin. Invest. 121, 1748–1752 (2011).

[0205] 64. Schulze, I. et al. Increased DNA methylation of Dnmt3b targets impairs leukemia. Blood 127, 1575–1586 (2016).

[0206] 65. Yang, L. et al. DNMT3A loss drives enhancer hypomethylation in FLT3-ITD-associated leukemias. Cancer Cell 29, 922–934 (2016); erratum 30, 363–365, (2016).

[0207] 66. Mayle, A. et al. Dnmt3a Loss can easily lead to malignant transformation of mouse hematopoietic stem cells. Dnmt3a losspredisposes murine hematopoietic stem cells to malignant transformation). Blood 125, 629–638 (2015).

[0208] 67. Haney, SL, et al. Promoter hypomethylation and expression are conserved in mouse chronic lymphocytic leukemia induced by decreased or inactivated Dnmt3a. Cell Rep.15，1190–1201 (2016).

Claims

1. A method for preparing a sequencing read dataset for detecting at least one CpG island (CGI) of circulating tumor DNA (ctDNA) from a patient sample, the method comprising: a. Treat cell-free DNA (cfDNA) obtained from patient samples with bisulfite; b. Amplify at least one CGI from bisulfite-treated cfDNA, wherein the at least one CGI is selected from Table 1; c. Sequencing the amplified DNA using next-generation sequencing (NGS) to generate a dataset containing multiple sequencing reads with at least one CGI.

2. The method according to claim 1, further comprising: d. Calculate the proportion (PMR) of at least one CGI consistent methylated read from multiple sequencing reads.

3. The method according to claim 2, further comprising: e. Detect the presence of circulating tumor DNA (ctDNA) in the sample when the PMR of at least one CGI is greater than that of normal tissue or epiblast control background.

4. The method of claim 1, wherein the processing, amplification, and sequencing steps are performed as part of a degenerate representative bisulfite sequencing process.

5. The method according to claim 1, wherein the sample is selected from plasma, urine, stool, menstrual fluid, and lymph.

6. The method of claim 1, wherein the sample is obtained from an individual diagnosed with cancer, having cancer, at risk of developing cancer, or suspected of having cancer.

7. The method according to claim 6, wherein the cancer is selected from bladder urothelial carcinoma, invasive breast carcinoma, colonic adenocarcinoma, colorectal adenocarcinoma, esophageal cancer, head and neck squamous cell carcinoma, clear cell renal carcinoma, papillary renal carcinoma, hepatocellular carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, prostate adenocarcinoma, stomach and esophageal cancer, thyroid cancer, endometrial cancer, and chronic lymphocytic leukemia.

8. The method of claim 3, wherein the presence of ctDNA indicates the presence of a tumor.

9. The method of claim 3, wherein the method is capable of detecting 0.01% ctDNA in the sample.

10. The method of claim 3, wherein the presence of ctDNA in the sample is detected with a sensitivity greater than 80%.

11. The method of claim 3, wherein the presence of ctDNA in the sample is detected with a specificity of greater than 75%.

12. The method of claim 3, wherein the presence of ctDNA in the sample is detected with 100% sensitivity and 95% specificity.

13. The method of claim 1, wherein the patient sample is a first patient sample obtained before the patient receives cancer treatment, and the method further comprises: f. Cell-free DNA (cfDNA) obtained from a second patient sample obtained from the same patient after cancer treatment was treated with bisulfite; g. Amplifying at least one CGI identical to that amplified from the cfDNA of a second patient sample treated with bisulfite as from the cfDNA of a first patient sample treated with bisulfite; h. The amplified DNA was sequenced using next-generation sequencing (NGS) to generate a second dataset containing a second set of multiple sequencing reads from at least one identical CGI from a second patient sample.

14. The method of claim 13, further comprising: i. Calculate the proportion (PMR) of at least one coherently methylated read from multiple sequencing reads to detect the amount of ctDNA in the first patient sample; j. Calculate the proportion (PMR) of at least one CGI consistent methylation read from the second set of multiple sequencing reads to detect the amount of ctDNA in the second patient sample; as well as k. Compare the amount of ctDNA obtained from the first patient sample with the amount of ctDNA obtained from the second patient sample; l. ctDNA increases the negative response of the indicator subject to cancer treatment, and ctDNA decreases the positive response of the indicator subject to cancer treatment.

15. A method for disrupting the methylation of CpG islands, the method comprising reducing the expression of Eed, Dnmtl, Dnmt3l, Dnmt3b or PRC2.

16. The method of claim 15, wherein the reduction is achieved through genome modification. Eed , Dnmtl , Dnmt3l , Dnmt3b or PRC2 The expression.

17. The method of claim 16, wherein the genome modification is performed using a CRISPR / Cas system.

Images