[0080] Regardless of potential genetic influences, the data indicate that inter-individual epigenetic variation at ESS probes occurs systemically and is stable over time. The inventors therefore evaluated ESS probes without regard to mQTL. Combined effects across multiple CpGs (i.e. differentially methylated regions) are more likely to demonstrate long-term stability and affect gene expression [31]. Hence, rather than analyze individual probes, the inventors focused on the 198 clusters of multiple ESS probes separated by no more than 500 bp (523 CpGs total, Table 3 in Appendix B). Analysis of expression [32] and methylation data [9] from

Grundberg el al. showed that at many ESS clusters, average methylation in adipose tissue is associated with gene expression, not only in adipose tissue but also in skin and lymphoblastoid

73689209.1 cell lines (FIG. 15). These results provide evidence that methylation at ESS loci in one tissue yields information about epigenetic regulation in additional tissues. To test for probe-specific associations between peripheral blood DNA methylation at baseline and risk of specific cancer diagnosis the inventors performed conditional logistic regression, adjusting for estimated leukocyte composition (using the Houseman algorithm [33]) and other covariates. Statistical significance of associations at the cluster level were then evaluated by permutation testing.

Relative to negative control clusters, the 198 ESS clusters were four-fold enriched for associations with later cancer (P = l.5xl0⁵). To minimize multiple testing, the inventors focused on the 10 ESS clusters with the largest number of CpG sites. Remarkably, at 7 of these, peripheral blood DNA methylation at baseline was significantly associated with later cancer (FIG. 7 and Table 4 in Appendix C); three ( SPATC1L , VTRNA2-1, and DUSP22 ) were significantly associated with multiple types of cancer. Elevated methylation in a cluster of 6 CpG sites at SPATC1L was associated with reduced risk of colorectal and prostate cancer (FIGS. 7b and 7f), and elevated methylation in a cluster of 12 CpGs at VTRNA2-1 was associated with higher risk of lung cancer and mature B-cell neoplasm (FIGS. 7d and 7e). Interestingly, elevated methylation in a cluster of 8 CpG sites at DUSP22 was associated with increased risk of mature B-cell neoplasm (FIG. 6e) yet reduced risk of urothelial cell carcinoma (FIG. 7g). The 154 negative control clusters showed few and relatively weak associations with later cancer (FIG.

16). Results are also shown for the 128 ESS clusters that included no probes with substantial mQTL (FIG. 17).

Significance of Certain Embodiments

[0081] Because they offer the potential to test the hypothesis that inter-individual epigenetic variation (in the absence of genetic variation) determines human phenotype, MZ twins have long been a focus of epigenetic investigation [34-38]. Such studies depend upon the existence of stochastic (i.e. non genetically-mediated) epigenetic differences within pairs of MZ twins. Conversely, herein has been identified a set of human genomic regions at which MZ twins exhibit non genetically-mediated epigenetic similarity. Based on the frequent occurrence of SIV in ESS regions, and their epigenetic plasticity to periconceptional environment, in particular embodiments of the disclosure ESS arises due to establishment of DNA methylation prior to embryo cleavage during MZ twinning.

73689209.1 [0082] Accordingly, at ESS loci one would expect greater epigenetic similarity in MZ twins that separate later compared to those that separate earlier. This can be tested based on chorionicity; cleavage before day 4 of gestation results in MZ twins each with their own placenta (dichorionic); later cleavage results in a shared placenta (monochorionic). In one of the earliest genome-scale studies of DNA methylation in twins, Kaminsky et al. [36] studied buccal epithelial cells and reported that monochorionic MZ twins exhibit greater epigenetic discordance than dichorionic, contrary to the inventors thesis. A slightly larger study, however, recently assessed genome-scale DNA methylation in blood and came to the exact opposite conclusion

[39]. Given that monochorionic twins share hematopoietic stem cells during fetal development

[40], blood is not the ideal tissue in which to study epigenetic correlates of chorionicity.

Definitive studies in non-blood tissues and focused on ESS regions are needed. Another predicted consequence of ESS is that estimates of methylation heritability from twin studies will be inflated relative to those from family-based designs. Indeed, whereas Grundberg et al.

estimated median genome-wide narrow-sense h² = 0.34 [9], a recent large family-based study (also using the HM450 platform) estimated an average genome-wide h² = 0.19 [41].

[0083] After decades of epigenetic studies in MZ twins, it is remarkable that ESS has not been previously reported. Despite their seemingly unsupportive findings in monochorionic vs. dichorionic twins, Kaminsky et al. proposed that in addition to their genetic identity,“epigenetic similarity at the time of blastocyst splitting may also contribute to the phenotypic similarities in MZ co-twins,” exactly as the findings suggest. The excessive h² estimates in twin studies of epigenetic heritability have, in fact, been waiting to be discovered. Grundberg et al. obtained but did not comment upon HM450 probe- specific h² estimates > 1. Likewise, in a more recent study using the HM450 array to assess genome-scale DNA methylation in whole blood of MZ and DZ twins, van Dongen et al. [42] reported 3792 probes for which their heritability model failed to converge. Of the 631 of these“NA” probes among the high-variance set from which the ESS probes were drawn, 365 (58%) are classified as ESS. Hence, two recent large studies of DNA methylation in MZ and DZ twins detected but did not explore these very interesting genomic regions.

[0084] The findings indicate complex relationships among genetic variation, ESS, and epigenetic metastability. To clarify, mQTL assesses pairwise associations between methylation at a specific CpG site and a specific genetic variant [19], while hap-ASM describes allelic biases in methylation that are associated with haplotype [17]. Because of the linkage disequilibrium

73689209.1 among neighboring SNPs and the regional correlation of CpG methylation, mQTL (specifically, cis mQTL) provides a means of assessing hap-ASM [17]. The analyses focused on mQTL because many targeted analyses of hap-ASM [21] show poor overlap with probes on the HM450 platform [17]. The results show that ESS loci are enriched for mQTL. Although this may seem to suggest that ESS is a consequence of genetic determination, we’ve provided several lines of evidence to the contrary. According to the model (FIG. 2d) mQTL is consistent with ESS, because any epigenetic state (whether under genetic influence or not) that is established prior to embryo cleavage during MZ twinning and thereafter maintained with high fidelity will exhibit ESS. Seminal studies in isogenic mice led to the concept that inter-individual variation at MEs is determined stochastically, free of genetic influence [43]. The characterization of ESS loci (many of which appear to be MEs) suggests the novel concept that establishment of epigenotype at MEs need not be completely free of genetic influence. In particular, mQTL and epigenetic

metastability appear to occur at the same loci (FIG. 5) and ESS loci - even those associated with substantial mQTL - are labile to perioconceptional environment (FIG. 6). Like nutrition [10-12] and other environmental influences [27, 44], perhaps haplotype (i.e. local sequence context) may be viewed as a determinant of the microenvironment that shifts the probability distribution of stochastic methylation processes during early embryonic development. Building upon this, the validation studies indicate allelic biases in epigenetic metastability. In the clearest example, at ZFP57 (FIG. 5d), the most common allele in the population showed greater inter-individual variation, consistent with the thesis that propensity for stochastic epigenetic variation may be both genetically inherited and evolutionarily advantageous [45].

[0085] It may seem surprising that ESS loci include some genomically imprinted genes. Based on their parent-of-origin specific epigenetic regulation one would expect the mean MSE at imprinted loci to be similar in MZ and DZ twin pairs. The data at VTRNA2-1, however, (FIG. la) show this is clearly not the case. Known imprinted genes were not significantly enriched among ESS loci, but there is evidence that two more of the top hits ( PAX8 and DUSP22) are imprinted in humans, in at least some tissues [46, 47]. In particular embodiments, inter-individual variation at imprinted loci may in some cases occur stochastically; for example, the VTRNA2-1

hypomethylation that is observed in 10-20% of individuals [13, 48, 49] may reflect loss of the maternally-inherited methylation mark in the early embryo. At the population level many ESS loci exhibit clusters of three methylation states (FIG. 1 and FIG. 15). This suggests these loci behave as bistable epigenetic switches (i.e. the combination of two alleles yields three preferred

73689209.1 average states). This is actually consistent with the bimodal distribution of somatic CpG methylation genomewide (i.e. methylation at most loci is either close to 0 or close to 100%). In this regard the presence of imprinted loci - paradigmatic bistable epigenetic switches - among ESS loci is not surprising. Although identified purely on the basis of the methylation MSE ratio of adult DZ to MZ twins, ESS probes are 3-fold enriched in subtelomeric regions. This makes sense; subtelomeric regions are packed with transposable elements, known to be targets of de novo DNA methylation in the pre-implantation embryo [50]. There was a similar enrichment in the genome-wide screen for MEs [13] but filtered out most of those hits due to proximity to SNPs. The current results, showing that the subtelomeric enrichment is not associated with mQTL, suggest the considerations were overly conservative. Intriguingly, since epigenetic regulation in subtelomeric regions regulates telomere shortening [50], the Gambian data showing season of conception effects at ESS regions suggest that periconceptional events could influence the process of telomere maintenance, deregulation of which is an almost universal characteristic of aging and cancer.

[0086] Because of their early embryonic establishment and systemic inter-individual variation, ESS loci are attractive candidate regions for studies of epigenetics and disease. The HM450 array was built upon a platform initially focused on regions aberrantly methylated in cancer, motivating the focus on cancer. Methylation at three clusters ( SPATC1L , VTRNA2-1, and DUSP22) showed significant associations with two types of cancer. Little is known about the speriolin-like protein SPATC1L, but elevated methylation at the small noncoding RNA VTRNA2-1 has previously been reported to predict poor prognosis in acute myeloid leukemia [49] and other types of cancer [13], consistent with the positive association that was found between VTRNA2-1 methylation and lung cancer and mature B-cell neoplasm (FIGS. 7d & 7e). Likewise, rearrangements disrupting the dual specificity phosphatase gene DUSP22 are associated with T-cell and B-cell lymphoma [51], consistent with the finding of a positive association between methylation at DUSP22 and mature B-cell neoplasm (FIG. 7e). Methylation at DUSP22 was also negatively associated with risk of urothelial cell carcinoma (FIG. 7g), reminiscent of situations in which the same genetic variant is associated oppositely with risk of different types of cancer [52]. ZFP57 encodes a master regulator of genomic imprinting and is epigenetically labile to periconceptional nutrition [13]. The finding that elevated methylation at ZFP57 is associated with a reduced risk of later colorectal cancer (FIG. 7b) is consistent with data suggesting ZFP57 is an oncogene [53]. Likewise, PF4 (platelet factor 4) inhibits tumor

73689209.1 growth and metastasis by suppressing neovascularization [54], consistent with the positive association found between PF4 methylation and later urothelial cell carcinoma (FIG. 7g).

Despite detecting significant associations between methylation and later cancer, the effect sizes in most cases were modest. It is likely that effects of some epimutations are limited to specific cancer subtypes. Likewise, these epigenetic variants likely interact with genetic variation and environmental exposures to affect cancer risk. It is possible that some of these associations might reflect mQTL/hap-ASM association with cancer-associated SNPs identified in GW AS studies. Targeted studies in larger cohorts are needed. Nonetheless, the data indicate that individual epigenetic variation at ESS loci has phenotypic consequences: methylation in peripheral blood is associated with risk of specific cancer diagnoses years later. Broader-scale identification of ESS loci throughout the genome may enable epigenetic risk models for cancer prediction, even during early life.

[0087] The findings indicate a partial explanation for missing heritability, in specific embodiments. Because heritability is defined as the phenotypic variance explained by genetics divided by the total phenotypic variation in a population [16], rather than something heritable per se, what is actually missing is genetic variance [55]. Twin models of estimating heritability rely on the assumption that the greater phenotypic similarity of MZ relative to DZ twin pairs is attributable to their genetic identity. Hence, to the extent that epigenetic variation at ESS loci influences phenotype, estimates of heritability based on twin studies will be inflated. Indeed, twin studies often yield higher heritability estimates than family studies [56, 57]. Further, although heritability does not definitively connote transgenerational inheritance, transmission of sequence-independent epigenetic events across generations could contribute to missing heritability [55]. In this regard, genomically imprinted loci that behave as epi-alleles (such as VTRNA2-1 ) potentiate transgenerational inheritance of epigenetic traits, in particular

embodiments.

Conclusions

[0088] Overall, the data show for the first time that, independent of their genetic identity, human MZ twin pairs share an additional level of similarity at the epigenetic level. ESS appears to result from establishment of mitotically heritable epigenetic states prior to embryo cleavage during MZ twinning. Because of ESS, human MZ twins clearly cannot be viewed as the epigenetic equivalent of isogenic inbred mice, which originate from separate zygotes. To the

73689209.1 extent that epigenetic variation at ESS loci influences human phenotype, as the data indicate, the existence of ESS establishes a link between early embryonic epigenetic development and adult disease and may call into question heritability estimates based on twin studies.

Examples of Methods

Identification and characterization of ESS and SIV probes

Analysis of twin, SIV, and mQTL data sets

[0089] Grundberg et al. used the Illumina HM450 array to assess methylation in adipose tissue from 662 female twins of European-descent, including 97 MZ pairs and 162 DZ pairs. Methylation scores were normalized by quantile normalization. SNP-associated probes were excluded, leaving 344,303 probes [9]. The analyses focused on the 34,405 probes in the top 10% by variance. To calculate the metrics used, the inventors pooled the MZ and DZ twins into a single population, and calculated probe- specific b-value ranges (max - min) from this population. Individuals in this population were randomly paired to simulate unrelated individuals (RZ). For each probe, the inventors calculated the MSE of MZ, DZ, and RZ pairs from the line of identity (i.e. the mean square difference between twins). For n twin pairs, each comprised of twins A and B,

[0090] Criteria for ESS probes were DZ/MZ > 2 and overall inter-individual b-value range (max - min) > 0.4; additionally, 14 probes with MZ/RZ MSE > 0.5 were excluded. Probe- specific h² estimates from Ref. [9] were kindly provided by Elin Grundberg. Lokk el al. used the Illumina HM450 array to assess methylation in 17 tissues from four autopsied individuals [20]. The inventors analyzed the methylation data for three tissues representing the three germ layers: gall bladder (endodermal), abdominal aorta (mesodermal), and sciatic nerve (ectodermal).

Starting with the 344,303 high-quality probes that were the basis of the Grundberg et al.

analysis, the inventors excluded any probes with missing values in any of the 12 samples (4 individuals, 3 tissues), leaving 344,151 probes. Inter-individual variation was calculated by taking the mean b value across each individual’s 3 tissues, then calculating the range of these means across all 4 individuals. Tissue-specific variation was calculated by taking the mean beta

73689209.1 value of each tissue over all individuals, then calculating the range of these means across all 3 tissues. Negative control probes were selected by maintaining the criterion of inter-individual range > 0.4 in the Grundberg el al. data set, but requiring MZ/RZ MSE > 0.5 and (in the Lokk data set) requiring tissue-specific variation to be at twice inter-individual variation (FIG. 13). Figures were made in R 3.3.1 using ggplot2 [58]. For the analysis of the Shi el al mQTF data [23], senior author Maria Fandi kindly shared with us their estimates of the proportion of methylation variance ( h²) explained by neighboring SNPs.

Validation studies

[0091] Quantitative analysis of selected candidate MEs was performed by bisulfite pyrosequencing [59] across endodermal (liver), mesodermal (kidney), and ectodermal (brain) tissue in 17 Asian cadavers [13]. Prior to use, all pyrosequencing assays were validated for linearity and sensitivity using human genomic methylation standards [12, 13]. To assess SIV, for each pyrosequencing assay methylation was averaged across multiple CpG sites for each samples, and inter-tissue correlation coefficients were calculated across the 17 cadavers (kidney vs. liver, brain vs. liver, and brain vs. kidney). Regions yielding an inter-tissue correlation of R² > 0.50 (R > 0.71) were considered positive for SIV [12]. Pyrosequencing was also used to perform SNP genotyping at specific loci [13, 14]. Associations between SNP genotype and average methylation were evaluated by linear regression (SAS), and effects of genotype on variance were evaluated by Bartlett’s test (implemented in R). Clonal bisulfite sequencing was performed as previously described [60], using appropriate primers.

Gene set enrichment analysis

[0092] For each of the probe sets analyzed ( e.g . ESS, SIV, and negative controls), associated gene sets were determined based on the HM450 probe annotations. For 24 cancer types profiled by The Cancer Genome Atlas (TCGA) [24], the inventors downloaded the RNA- Seq gene expression profiles using the firebrowse.org portal [61], selected genes with maximum FPKM across all samples exceeding 1, then generated rank file for Gene Set Enrichment Analysis (GSEA) [62] as previously described [63] by comparing the tumor samples and the adjacent normal samples. Next, GSEA was run for each cancer type rank file and each CpG- associated gene set; significance was considered for q-value<0.25. For visualization purposes, The inventors represented the Normalized Enrichment Score (NES) for each significant enrichment; by convention, NES for enrichments with Q>=0.25 were considered 0.

73689209.1 Epigenomic distribution ofCpG probes

[0093] For each of probe sets analyzed (e.g. ESS, SIV, and negative controls), genomic coordinates on the human genome build UCSC hgl9 were determined based on the HM450 probes definition. The inventors considered fifteen-states genome wide epigenomic partitions for 127 distinct epigenomes as defined by the NIH Epigenomic Roadmap Consortium [25], based on a collection of uniformly collected histone modification ChIP-Seq profiles and using the ChromHMM algorithm [64]. Using the BEDTOOLS software, the relative distribution was determined of each CpG probes set over the fifteen epigenomic states for each distinct epigenome.

Evaluating the relationship between DNA methylation and gene expression at ESS clusters

[0094] As described above, HM450 methylation data was used for subcutaneous adipose tissue from MZ and DZ twins [9]. Gene expression data in skin, adipose tissue, and

lymphoblastoid cell lines from the same set of twins was downloaded from ArrayExpress (accession # E-TABM-1140) [32, 65]. DNA methylation (b values) were first averaged across probes within each ESS cluster. Correlation between cluster-level DNA methylation and associated gene expression was evaluated using the Spearman (rank) correlation coefficient, as implemented in the Python scientific libraries.

Season of conception analyses

Sample selection and preparation

[0095] The data were collected as part of a study in The Gambia (in sub-Saharan West Africa) identifying biomarkers and understanding mechanisms for the relationship between aflatoxin exposure and child stunting. 251 blood samples (3 ml) were collected from children aged 2 years as part of the Early Nutrition and Immune Development (ENID) Trial [29].

Bisulfite conversion and DNA methylation assay

[0096] DNA was extracted from white blood cells following a previously described protocol [12]. An additional 6 samples were included as technical replicates. Genome-scale methylation profiles were obtained using HM450 Infinium methylation bead arrays (Illumina,

73689209.1 San Diego, USA). DNA samples (500 ng) were bisulfite-modified using the EZ DNA Methylation kit (Zymo Research, D5001) following manufacturer’s instructions for the HM450 array. Modified DNA was stored at -20C until used. Amplification, labelling, hybridization and scanning was performed as previously described [13].

Methylation data QC and normalization

[0097] Methylation data pre-processing was performed using the R/Bioconductor minfi package [66], along with other functions and bespoke R scripts as appropriate (R version 3.2.3; Bioconductor version 3.2). Briefly, data for 485,512 HM450 probes measured in 257 samples were imported from raw ID AT files. Analysis of internal HM450 bisulfite conversion control probes revealed one sample with poor bisulfite conversion efficiency which was excluded.

Functional normalization [67] was used to reduce unwanted technical variation using control probes on the array. 7 samples with discordant sex were removed following sex prediction based on median values of measurement on X and Y chromosomes using the minfi addSex() function. Using a probe detection p-value threshold of 0.01, 5 samples failing in >1% of probes were removed, along with 32,488 probes failing in 1 or more samples. All technical replicates showed beta-value Pearson correlations > 0.994 and visual inspection of replicate correlation scatterplots revealed no anomalies. Following removal of technical replicates and X and Y chromosome probes, methylation data for 442,869 probes measured in 239 individuals remained. Correction for differences in HM450 beta-value distributions due to type-I and type-II probes on the array was conducted using the Beta Mixture Quantile Dilation (BMIQ) method [68]. Finally, 28,509 cross-reactive probes [69] and 41,334 probes within lObp of common (MAF > 1%) African SNPs identified using the R Illumina450ProbeVariants.db were removed. After all QC and filtering, 373,026 probes remained.

Identification ofCpGs associated with Gambian Season of Conception

[0098] Statistical analysis was performed to identify HM450 probes associated with Gambian season of conception, described hereafter as‘season of conception differentially methylated probes’ (SoC-DMPs). This analysis was restricted to 128 individuals conceived at the peak of either the Gambian dry (February- April) or rainy (July- September) seasons (based on date of birth). These seasonal windows have been used in previous studies investigating seasonal effects on DNA methylation [12, 13]. Robust linear regression using the R rim function was used to model the association between SoC and DNA methylation (measured as HM450 beta-values),

73689209.1 in order to account for potential heteroscedasticity and influential outliers [70, 71]. The regression model included infant sex, and the first 3 principal components identified in an unsupervised principal components analysis of genome-wide methylation (FIG. 18). The model was additionally adjusted for the effects of cell heterogeneity using an established method that uses methylation data to estimate the relative proportions of 6 white blood cell types [33].

Additional analyses were performed i) without cell composition adjustment to assess sensitivity to cell composition effects; and ii) with the inclusion of one further principal component (PC 8) which was associated with SoC which would be expected to dilute the SoC effect (hence providing a more conservative estimate of SoC-associated ME enrichment). A correction for multiple testing was applied by controlling the false discovery rate (FDR).

Enrichment analysis

[0099] Probes with an FDR < 10% were identified as SoC-DMPs. Different sets of HM450 probes were tested for SoC-DMP enrichment (FDR < 10%) using Fisher’s exact test, against a background of all 373,026 probes passing QC and filtering steps. Enrichment results for the main analysis are presented in FIG. 6A. Additional enrichment tests were performed without adjustment for cell composition and with the inclusion of one further principal component (see previous section).

Identification of ESS clusters associated with cancer risk

Sample collection, data generation, and quality control

[0100] Methylation data were available for participants in one of seven case-control studies of breast, colorectal, kidney, lung, mature B-cell malignancies, prostate or urothelial cancer [72-74] nested within the Melbourne Collaborative Cohort Study [30]. DNA was extracted from samples of peripheral blood mononuclear cells (PBMC), buffy coats or dried blood spots (DBS) stored on Guthrie card diagnostic cellulose filter paper. Samples were collected at recruitment to the cohort (baseline) or at follow-up approximately ten years later. Cases and controls were individually matched to cases on age (they had to be free of cancer at an age within one year of the age at diagnosis of the corresponding case), sex, ethnicity and blood DNA source (DBS, PBMC or buffy coat). For all but the colorectal cancer study, controls were matched to cases on year of birth. For the lung cancer study, controls were matched on smoking

73689209.1 status at the time of blood collection. Case-control pairs were placed at random positions on a same chip of the assay to minimize batch effects.

[0101] DNA was extracted from mononuclear cells using QIAamp mini spin columns (Qiagen, Hilden, Germany). Dried blood spot DNA was extracted as previously described [75]. Briefly, twenty blood spots of 3.2 mm diameter were punched from the Guthrie card and lysed in phosphate buffered saline using TissueLyser (Qiagen). The resulting supernatant was processed using Qiagen mini spin columns according to the manufacturer’s protocol. The quality and quantity of DNA was assessed using the Quant-iT™ Picogreen® dsDNA assay measured on the Qubit® Fluorometer (Life Technologies, Grand Island, NY), with a minimum of 0.3 pg DNA considered acceptable for methylation analysis.

[0102] Bisulfite conversion was performed using the Zymo Gold single tube kit (EZ DNA Methylation-Gold kit, Zymo Research, Irvine, CA) according to the manufacturer’s instructions. Post-conversion quality control was performed using SYBR Green-based quantitative PCR, an in-house assay designed to determine the success of bisulphite conversion by comparing amplification of the test sample with positive and negative controls. All samples were processed in the same laboratory on 96-well plates, each using eight HM450 BeadChips to assay batches of 12 samples.

[0103] The methylation data were background corrected and normalized based on internal control probes using the manufacturer’s background correction, using the R library minfi [66]. The inventors also applied subset-quantile within-array normalization (SWAN) [76] to correct for technical discrepancies between type I and type II probes on the assay. A b-value (interpreted as percentage methylation) was calculated for each CpG site using minfi.

Methylation measures with a detection p-value higher than 0.01 were considered missing.

Samples with more than 5% missing values were excluded; then, CpGs that were missing for more than 20% of samples were excluded b-values were transformed into M-values for analysis: M=1o_b2(b/(1-b)).

Logistic regression and permutation analyses

[0104] For each CpG probe set, their clustering structure was first determined by considering all CpGs within 500bp of each other; groups of at least 2 such CpGs were considered clusters. ESS, SIV, and negative control cluster annotations are provided in Table 3

73689209.1 in Appendix B. For each of the 7 case-control cohorts described above, normalized DNA methylation data at the CpG probe level were obtained. Since methylation was measured in peripheral blood DNA, cell type composition estimates using established methods [33] were also included for each sample (specifically, proportions of B-cells, granulocytes, monocytes, NK,

CD4 T-cells, and CD8 T-cells). Clinical data variables indicating body mass index (BMI), alcohol consumption, and smoking status were included for each subject. Many ESS probes showed highly non-normal methylation distributions within each cohort. To avoid incorrect assumptions about the data distribution, the M- values for each probe were rank- normalized in ascending order across all samples using the R statistical system. Using conditional logistic regression as implemented in the R survival package, the inventors determined for each probe the significance of the association between methylation rank and cancer status, in a model including both cell type proportion and the clinical variables described above. For the purposes of permutation testing (see below), associations were considered statistically significant at p<0.05.

[0105] These probe- specific P values were then utilized to evaluate the statistical significance of associations at the cluster level. Analysis was focused on the top 10 clusters by total number of CpGs. The significance was assessed of two event types. For each cluster (C), cancer type (7), and random assignment of the case-control status for each matched pair (S_md), the inventors denoted the number of significant probes at p<0.05 with concordant coefficients as determined by conditional logistic regression as N(C,T,S_md ) The inventors denote the actual case-control status from the MCCS cohort as N(C,T,S_0bs ) The inventors likewise denote the minimum p-value obtained across all the probes in a cluster, for the randomly-assigned and actual case-control pairing as P_mm(C,T,S_md) and P_min(C,T,S_0bs). The event was defined:

Ϊ) N( C,T,S rnd ) >= N(C,T,Sobs) and Pmin(C,T,Smd) <= Pmin(C,T,Sobs).

[0106] Next, for each random assignment S_md the inventors defined the event

Recurrence(C,S_md ) as the number of cancer types for which a cluster C contains at least two significant probes (p<0.05) with concordant coefficients. The corresponding value for the actual case-control status is Recurrence(C,S_0bs). The event was defined: ii)) Recurrence(C,Smd) >= Recurrence(C,S_abs)·

[0107] The null hypothesis is that both i) and ii) occur by chance. Similar to widely used methods such as GSEA [62], permutation testing was used to establish a null distribution for S.

73689209.1 The inventors generated 20,000 permutations for each individual cancer site, by keeping the sample pairing (as indicated by the patient id) but randomly assigning the case / control status within each pair. For each permutation S the inventors applied conditional logistic regression for each cancer type, and counted events i) and ii) as described above. The inventors assigned each event a p-value corresponding to the relative number of permutations (out of 20,000) for which events i) or ii) were observed. Statistical significance was achieved at the FDR<0.25 level, across the top 10 most CpG-rich clusters.

REFERENCES

[0108] All publications mentioned in this specification are indicative of the level of those skilled in the art to which the invention pertains. All publications herein are incorporated by reference to the same extent as if each individual publication was specifically and individually indicated to be incorporated by reference in their entirety.

1. Jaenisch R, Bird A: Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat Genet 2003, 33 Suppl:245-254.

2. Holliday R: The inheritance of epigenetic defects. Science 1987, 238:163-170.

3. Baylin SB, Jones PA: A decade of exploring the cancer epigenome - biological and translational implications. Nat Rev Cancer 2011, 11:726-734.

4. Bimey E, Smith GD, Greally JM: Epigenome-wide Association Studies and the Interpretation of Disease -Omics. PLoS Genet 2016, l2:el006l05.

5. Rakyan VK, Down TA, Balding DJ, Beck S: Epigenome-wide association studies for common human diseases. Nat Rev Genet 2011, 12:529-541.

6. Rakyan VK, Blewitt ME, Druker R, Preis JI, Whitelaw E: Metastable epialleles in mammals. Trends Genet 2002, 18:348-351.

7. Duhl DM, Vrieling H, Miller KA, Wolff GL, Barsh GS: Neomorphic agouti mutations in obese yellow mice. NatGenet 1994, 8:59-65.

8. Morgan HD, Sutherland HG, Martin DI, Whitelaw E: Epigenetic inheritance at the agouti locus in the mouse. Nat Genet 1999, 23:314-318.

73689209.1 9. Grundberg E, Meduri E, Sandling JK, Hedman AK, Keildson S, Buil A, Busche S, Yuan W, Nisbet J, Sekowska M, et al: Global analysis of DNA methylation variation in adipose tissue from twins reveals links to disease-associated variants in distal regulatory elements. Am J Hum Genet 2013, 93:876-890.

10. Waterland RA, Jirtle RL: Transposable elements: targets for early nutritional effects on epigenetic gene regulation. Mol Cell Biol 2003, 23:5293-5300.

11. Waterland RA, Dolinoy DC, Lin JR, Smith CA, Shi X, Tahiliani KG: Maternal methyl supplements increase offspring DNA methylation at Axin fused. Genesis 2006, 44:401- 406.

12. Dominguez-Salas P, Moore SE, Baker MS, Bergen AW, Cox SE, Dyer RA, Fulford AJ, Guan Y, Laritsky E, Silver MJ, et al: Maternal nutrition at conception modulates DNA methylation of human metastable epialleles. Nat Commun 2014, 5:3746.

13. Silver MJ, Kessler NJ, Hennig BJ, Dominguez-Salas P, Laritsky E, Baker MS, Coarfa C, Hernandez- Vargas H, Castelino JM, Routledge MN, et al: Independent genomewide screens identify the tumor suppressor VTRNA2-1 as a human epiallele responsive to

periconceptional environment. Genome Biol 2015, 16:118.

14. Waterland RA, Kellermayer R, Laritsky E, Rayco-Solon P, Harris RA, Travisano M, Zhang W, Torskaya MS, Zhang J, Shen L, et al: Season of conception in rural gambia affects DNA methylation at putative human metastable epialleles. PLoS Genet 2010, 6:el00l252.

15. Visscher PM, Benyamin B, White I: The use of linear mixed models to estimate variance components from data on twin pairs by maximum likelihood. Twin Res 2004, 7:670- 674.

16. Visscher PM, Hill WG, Wray NR: Heritability in the genomics era— concepts and misconceptions. Nat Rev Genet 2008, 9:255-266.

17. Do C, Lang CF, Lin J, Darbary H, Krupska I, Gaba A, Petukhova L, Vonsattel JP, Gallagher MP, Goland RS, et al: Mechanisms and Disease Associations of Haplotype-Dependent Allele-Specific DNA Methylation. Am J Hum Genet 2016, 98:934-955.

73689209.1 18. Titlestad IL, Kyvik KO, Kristensen T, Lillevang S: HLA haplotypes in dizygotic twin pairs: are dizygotic twins more similar than sibs? Twin Res 2002, 5:287-288.

19. Yet I, Tsai PC, Castillo-Femandez JE, Carnero-Montoro E, Bell JT: Genetic and environmental impacts on DNA methylation levels in twins. Epigenomics 2016, 8:105-117.

20. Lokk K, Modhukur V, Rajashekar B, Martens K, Magi R, Kolde R, Koltsina M, Nilsson TK, Vilo J, Salumets A, Tonisson N: DNA methylome profiling of human tissues identifies global and tissue- specific methylation patterns. Genome Biol 2014, l5:r54.

21. Do C, Shearer A, Suzuki M, Terry MB, Gelernter J, Greally JM, Tycko B:

Genetic-epigenetic interactions in cis: a major focus in the post-GWAS era. Genome Biol 2017, 18:120.

22. Volkov P, Olsson AH, Gillberg L, Jorgensen SW, Brons C, Eriksson KF, Groop L, Jansson PA, Nilsson E, Ronn T, et al: A Genome-Wide mQTL Analysis in Human Adipose Tissue Identifies Genetic Variants Associated with DNA Methylation, Gene Expression and Metabolic Traits. PLoS One 2016, H:e0l57776.

23. Shi J, Marconett CN, Duan J, Hyland PL, Li P, Wang Z, Wheeler W, Zhou B, Campan M, Lee DS, et al: Characterizing the genetic basis of methylome diversity in histologically normal human lung tissue. Nat Commun 2014, 5:3365.

24. Cancer Genome Atlas Research N, Weinstein JN, Collisson EA, Mills GB, Shaw KR, Ozenberger BA, Ellrott K, Shmulevich I, Sander C, Stuart JM: The Cancer Genome Atlas Pan-Cancer analysis project. Nat Genet 2013, 45: 1113-1120.

25. Roadmap Epigenomics C, Kundaje A, Meuleman W, Ernst J, Bilenky M, Yen A, Heravi-Moussavi A, Kheradpour P, Zhang Z, Wang J, et al: Integrative analysis of 111 reference human epigenomes. Nature 2015, 518:317-330.

26. Dolinoy DC, Huang D, Jirtle RL: Maternal nutrient supplementation counteracts bisphenol A-induced DNA hypomethylation in early development. Proc Natl Acad Sci U SA 2007, 104:13056-13061.

73689209.1 27. Estill MS, Bolnick JM, Waterland RA, Bolnick AD, Diamond MP, Krawetz SA: Assisted reproductive technology alters deoxyribonucleic acid methylation profiles in bloodspots of newborn infants. Fertil Steril 2016, 106:629-639 e6l0.

28. Kuhnen P, Handke D, Waterland RA, Hennig BJ, Silver M, Fulford AJ,

Dominguez-Salas P, Moore SE, Prentice AM, Spranger J, et al: Inter-individual Variation in DNA Methylation at a Putative POMC Metastable Epiallele Is Associated with Obesity. Cell Metab 2016, 24:502-509.

29. Moore SE, Fulford AJ, Darboe MK, Jobarteh ML, Jarjou LM, Prentice AM: A randomized trial to investigate the effects of pre-natal and infant nutritional supplementation on infant immune development in rural Gambia: the ENID trial: Early Nutrition and Immune Development. BMC Pregnancy Childbirth 2012, 12:107.

30. Giles GG, English DR: The Melbourne Collaborative Cohort Study. IARC Sci Publ 2002, 156:69-70.

31. Bock C: Analysing and interpreting DNA methylation data. Nat Rev Genet 2012, 13:705-719.

32. Grundberg E, Small KS, Hedman AK, Nica AC, Buil A, Keildson S, Bell JT, Yang TP, Meduri E, Barrett A, et al: Mapping cis- and trans-regulatory effects across multiple tissues in twins. Nat Genet 2012, 44:1084-1089.

33. Jaffe AE, Irizarry RA: Accounting for cellular heterogeneity is critical in epigenome-wide association studies. Genome Biol 2014, l5:R3l.

34. Bell JT, Spector TD: A twin approach to unraveling epigenetics. Trends Genet 2011, 27:116-125.

35. Fraga MF, Ballestar E, Paz MF, Ropero S, Setien F, Ballestar ML, Heine-Suner D, Cigudosa JC, Urioste M, Benitez J, et al: Epigenetic differences arise during the lifetime of monozygotic twins. Proc Natl Acad Sci U SA 2005, 102:10604-10609.

36. Kaminsky ZA, Tang T, Wang SC, Ptak C, Oh GH, Wong AH, Feldcamp LA, Virtanen C, Halfvarson J, Tysk C, et al: DNA methylation profiles in monozygotic and dizygotic twins. Nat Genet 2009, 41:240-245.

73689209.1 37. Oates NA, van Vliet J, Duffy DL, Kroes HY, Martin NG, Boomsma DI,

Campbell M, Coulthard MG, Whitelaw E, Chong S: Increased DNA methylation at the AXIN1 gene in a monozygotic twin from a pair discordant for a caudal duplication anomaly. Am J Hum Genet 2006, 79:155-162.

38. Wong AH, Gottesman, II, Petronis A: Phenotypic differences in genetically identical organisms: the epigenetic perspective. Hum Mol Genet 2005, 14 Spec No l:Rl 1-18.

39. Bui M, Benyamin B, Shah S, Henders AK, Martin NG, Montgomery GW, McRae AF: Sharing a Placenta is Associated With a Greater Similarity in DNA Methylation in

Monochorionic Versus Dichorionic Twin Pars in Blood at Age 14. Twin Res Hum Genet 2015, 18:680-685.

40. Weksberg R, Shuman C, Caluseriu O, Smith AC, Fei YF, Nishikawa J, Stockley TF, Best F, Chitayat D, Olney A, et al: Discordant KCNQlOTl imprinting in sets of

monozygotic twins discordant for Beckwith- Wiedemann syndrome. Hum Mol Genet 2002, 11:1317-1325.

41. McRae AF, Powell JE, Henders AK, Bowdler F, Hemani G, Shah S, Painter JN, Martin NG, Visscher PM, Montgomery GW: Contribution of genetic variation to

transgenerational inheritance of DNA methylation. Genome Biol 2014, l5:R73.

42. van Dongen J, Nivard MG, Willemsen G, Hottenga JJ, Helmer Q, Dolan CV, Ehli EA, Davies GE, van Iterson M, Breeze CE, et al: Genetic and environmental influences interact with age and sex in shaping the human methylome. Nat Commun 2016, 7:11115.

43. Rakyan VK: Metastable epialleles in mammals. Trends in genetics 2002, 18:348-

351.

44. Dolinoy DC, Weidman JR, Waterland RA, Jirtle RE: Maternal genistein alters coat color and protects Avy mouse offspring from obesity by modifying the fetal epigenome. Environ Health Perspect 2006, 114:567-572.

45. Feinberg AP, Irizarry RA: Evolution in health and medicine Sackler colloquium: Stochastic epigenetic variation as a driving force of development, evolutionary adaptation, and disease. Proc Natl Acad Sci U SA 2010, 107 Suppl 1:1757-1764.

73689209.1 46. Green BB, Kappil M, Lambertini L, Armstrong DA, Guerin DJ, Sharp AJ, Lester BM, Chen J, Marsit CJ: Expression of imprinted genes in placenta is associated with infant neurobehavioral development. Epigenetics 2015, 10:834-841.

47. Kukurba KR, Zhang R, Li X, Smith KS, Knowles DA, How Tan M, Piskol R,

Lek M, Snyder M, Macarthur DG, et al: Allelic expression of deleterious protein-coding variants across human tissues. PLoS Genet 2014, l0:el004304.

48. Romanelli V, Nakabayashi K, Vizoso M, Moran S, Iglesias-Platas I, Sugahara N, Simon C, Hata K, Esteller M, Court F, Monk D: Variable maternal methylation overlapping the nc886/vtRNA2-l locus is locked between hypermethylated repeats and is frequently altered in cancer. Epigenetics 2014, 9:783-790.

49. Treppendahl MB, Qiu X, Sogaard A, Yang X, Nandrup-Bus C, Hother C, Andersen MK, Kjeldsen L, Mollgard L, Hellstrom-Lindberg E, et al: Allelic methylation levels of the noncoding VTRNA2-1 located on chromosome 5q3l.l predict outcome in AML. Blood 2012, 119:206-216.

50. Blasco MA: The epigenetic regulation of mammalian telomeres. Nat Rev Genet 2007, 8:299-309.

51. Hapgood G, Savage KJ: The biology and management of systemic anaplastic large cell lymphoma. Blood 2015, 126:17-25.

52. Kamper-Jorgensen M, Biggar RJ, Tjonneland A, Hjalgrim H, Kroman N, Rostgaard K, Stamper CL, Olsen A, Andersen AM, Gadi VK: Opposite effects of

microchimerism on breast and colon cancer. Eur J Cancer 2012, 48:2227-2235.

53. Tada Y, Yamaguchi Y, Kinjo T, Song X, Akagi T, Takamura H, Ohta T, Yokota T, Koide H: The stem cell transcription factor ZFP57 induces IGF2 expression to promote anchorage-independent growth in cancer cells. Oncogene 2015, 34:752-760.

54. Lippi G, Favaloro EJ: Recombinant platelet factor 4: a therapeutic, anti-neoplastic chimera? Semin Thromb Hemost 2010, 36:558-569.

73689209.1 55. Eichler EE, Flint J, Gibson G, Kong A, Leal SM, Moore JH, Nadeau JH: Missing heritability and strategies for finding the underlying causes of complex disease. Nat Rev Genet 2010, 11:446-450.

56. Costa AM, Breitenfeld L, Silva AJ, Pereira A, Izquierdo M, Marques MC:

Genetic inheritance effects on endurance and muscle strength: an update. Sports Med 2012, 42:449-458.

57. Gordon H, Trier Moller F, Andersen V, Harbord M: Heritability in inflammatory bowel disease: from the first twin study to genome-wide association studies. Inflamm Bowel Dis 2015, 21:1428-1434.

58. Wickham H: ggplot2: Elegant Graphics for Data Analysis. Springer- Verlag New York; 2009.

59. Shen L, Guo Y, Chen X, Ahmed S, Issa JP: Optimizing annealing temperature overcomes bias in bisulfite PCR methylation analysis. Biotechniques 2007 , 42:48-58.

60. Waterland RA, Kellermayer R, Rached MT, Tatevian N, Gomes MV, Zhang J, Zhang L, Chakravarty A, Zhu W, Laritsky E, et al: Epigenomic profiling indicates a role for DNA methylation in early postnatal liver development. Hum Mol Genet 2009, 18:3026-3038.

61. Deng M, Bragelmann J, Kryukov I, Saraiva-Agostinho N, Pemer S: FirebrowseR: an R client to the Broad Institute's Firehose Pipeline. Database (Oxford) 2017, 2017.

62. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP: Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U SA 2005, 102:15545-15550.

63. He B, Lanz RB, Fiskus W, Geng C, Yi P, Hartig SM, Rajapakshe K, Shou J, Wei L, Shah SS, et al: GATA2 facilitates steroid receptor coactivator recruitment to the androgen receptor complex. Proc Natl Acad Sci U SA 2014, 111:18261-18266.

64. Ernst J, Kheradpour P, Mikkelsen TS, Shoresh N, Ward LD, Epstein CB, Zhang X, Wang L, Issner R, Coyne M, et al: Mapping and analysis of chromatin state dynamics in nine human cell types. Nature 2011, 473:43-49.

73689209.1 65. Yang J, Huang T, Petralia F, Long Q, Zhang B, Argmann C, Zhao Y, Mobbs CV, Consortium GT, Schadt EE, et al: Synchronized age-related gene expression changes across multiple tissues in human and the link to complex diseases. Sci Rep 2015, 5:15145.

66. Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, Irizarry RA: Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics 2014, 30:1363-1369.

67. Fortin JP, Labbe A, Lemire M, Zanke BW, Hudson TJ, Fertig EJ, Greenwood CMT, Hansen KD: Functional normalization of 450k methylation array data improves replication in large cancer studies. Genome Biol 2014, 15.

68. Teschendorff AE, Marabita F, Lechner M, Bartlett T, Tegner J, Gomez-Cabrero D: A beta-mixture quantile normalization method for correcting probe design bias in Illumina Infinium 450k DNA methylation data. Bioinformatics 2013, 29.

69. Chen Y, Lemire M, Choufani S, Butcher DT, Grafodatskaya D, Zanke BW, Gallinger S, Hudson TJ, Weksberg R: Discovery of cross-reactive probes and polymorphic CpGs in the Illumina Infinium HumanMethylation450 microarray. Epigenetics 2013, 8:203-209.

70. Joubert BR, Felix JF, Yousefi P, Bakulski KM, Just AC, Breton C, Reese SE, Markunas CA, Richmond RC, Xu CJ, et al: DNA Methylation in Newborns and Maternal Smoking in Pregnancy: Genome-wide Consortium Meta- analysis. Am J Hum Genet 2016, 98:680-696.

71. Engel SM, Joubert BR, Wu MC, Olshan AF, Haberg SE, Ueland PM, Nystad W, Nilsen RM, Vollset SE, Peddada SD, London SJ: Neonatal genome- wide methylation patterns in relation to birth weight in the Norwegian Mother and Child Cohort. Am J Epidemiol 2014, 179:834-842.

72. Dugue PA, Brinkman MT, Milne RL, Wong EM, FitzGerald LM, Bassett JK, Joo JE, Jung CH, Makalic E, Schmidt DF, et al: Genome-wide measures of DNA methylation in peripheral blood and the risk of urothelial cell carcinoma: a prospective nested case-control study. Br J Cancer 2016, 115:664-673.

73689209.1 73. Severi G, Southey MC, English DR, Jung CH, Lonie A, McLean C, Tsimiklis H, Hopper JL, Giles GG, Baglietto L: Epigenome-wide methylation in DNA from peripheral blood as a marker of risk for breast cancer. Breast Cancer Res Treat 2014, 148:665-673.

74. Wong Doo N, Makalic E, Joo JE, Vajdic CM, Schmidt DF, Wong EM, Jung CH, Severi G, Park DJ, Chung J, et al: Global measures of peripheral blood-derived DNA

methylation as a risk factor in the development of mature B-cell neoplasms. Epigenomics 2016.

75. Joo JE: The use of DNA from archival dried blood spots with the Infinium HumanMethylation450 array. BMC biotechnology 2013, 13:23.

76. Maksimovic J, Gordon L, Oshlack A: SWAN: Subset-quantile within array normalization for illumina infinium HumanMethylation450 BeadChips. Genome Biol 2012, l3:R44.

77. Machiela MJ, Chanock SJ: LDlink: a web-based application for exploring population- specific haplotype structure and linking correlated alleles of possible functional variants. Bioinformatics 2015, 31:3555-3557.

[0102] Although the present invention and its advantages have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims. Moreover, the scope of the present application is not intended to be limited to the particular embodiments of the process, machine, manufacture, composition of matter, means, methods and steps described in the specification. As one of ordinary skill in the art will readily appreciate from the disclosure of the present invention, processes, machines, manufacture, compositions of matter, means, methods, or steps, presently existing or later to be developed that perform substantially the same function or achieve substantially the same result as the corresponding embodiments described herein may be utilized according to the present invention. Accordingly, the appended claims are intended to include within their scope such processes, machines, manufacture, compositions of matter, means, methods, or steps.

73689209.1 APPENDIX A

Table 1: Annotated list of ESS and SIV probes

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1 APPENDIX B

Table 3: ESS clusters

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

73689209.1

APPENDIX C

Table 4: MCCS permutation testing results for 10 most CpG-rich ESS clusters

73689209.1

73689209.1

73689209.1 APPENDIX D

Table 5: CpG sites in Cancer-Associated Clusters

73689209.1

73689209.1

Claims

CLAIMS What is claimed is:

1. A method of identifying a risk for a subject to develop cancer, comprising the step of determining from a genomic DNA sample from the subject the methylation status of one or more loci selected from the group consisting of ZFP57, SPATC1L, OR2L-13, VTRNA2-1, DUSP22, HCG4B, PF4, and a combination thereof.

2. The method of claim 1, wherein determining the methylation status comprises identifying methylation at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, or more CpG sites in one or more of the loci.

3. The method of claim 1 or 2, wherein the loci include all of ZFP57, SPATC1L, OR2L-13, VTRNA2-1, DUSP22, HCG4B, and PF4.

4. The method of claim 1 or 2, wherein the loci include 6, 5, 4, 3, or 2 of ZFP57, SPATC1L, OR2L-13, VTRNA2-1, DUSP22, HCG4B, and PF4.

5. The method of any one of claims 1-4, wherein the individual is an adult, adolescent, child, or infant.

6. The method of any one of claims 1-5, wherein the cancer is cancer of the lung, breast, brain, colon, pancreas, uterus, bone, skin, endometrium, testes, uterus, spleen, liver, kidney, stomach, thyroid, gall bladder, blood, prostate, hematopoietic lineages, esophagus or is urothelial cancer.

7. The method of any one of claims 1-6, wherein the sample comprises peripheral blood, biopsy, hair, follicles, fingernails, saliva, cheek scrapings, cerebrospinal fluid, urine, nipple aspirate, fecal material, semen, sputum, mucus, or a combination thereof.

8. The method of any one of claims 1-7, wherein the method further comprises the step of obtaining a sample from the individual.

9. The method of any one of claims 1-8, wherein the method further comprises the step of providing a therapeutic and/or a preventative therapy to the individual.

10. The method of claim 9, wherein the therapeutic comprises surgery, drug, radiation, immunotherapy, hormone therapy, or a combination thereof.

73689209.1

11. The method of claim 9, wherein the preventative therapy comprises watchful waiting, surgery, drug, radiation, immunotherapy, hormone therapy, or a combination thereof.

12. The method of any one of claims 1-11, wherein the genomic DNA sample is from a tissue or body fluid other than the tissue or body fluid at risk for development of the cancer.

73689209.1