{"id":{"repo_id":"cambridge","oai_identifier":"oai:www.repository.cam.ac.uk:1810/368213"},"canonical_url":"https://search.dev.ndltd.org/etd/cambridge/oai:www.repository.cam.ac.uk:1810/368213","repository":{"repo_id":"cambridge","name":"Cambridge University","base_url":"https://api.repository.cam.ac.uk/server/oai/request"},"display":{"title":"DNA methylation variability at the crossroads of stochasticity, genetics, and environment","abstract":"During mammalian development, DNA methylation at most CpG sites throughout the genome is erased; these marks are then re-established later. In humans and in mice, typically all copies of a given CpG site in a particular tissue are methylated in all individuals; a large minority are unmethylated in all cases. A substantial proportion are methylated in some but not all tissues. A smaller minority of CpG sites are only partially methylated, and amongst these a proportion have consistent methylation levels across various tissues but variable methylation levels between individuals. Loci exhibiting this latter phenomenon, often referred to as metastable epialleles, were initially identified in inbred mice, where a lack of genetic variability indicates the presence of a genuine epimutation. Furthermore, the consistency of methylation across multiple tissues suggests a timing of methylation establishment which is early in development. These initial findings were located at IAP elements, a murine-specific class of ERV transposable element. There are approximately 10,000 IAP elements in the mouse genome, but the annotation of these elements is often poor and fractured. I developed an ad hoc algorithm to systematically fix the annotations, which reduced the proportion of fractured IAP elements from 36% to 19%. Using a genome-wide dataset, I updated an existing screen of variable methylation at IAP elements and identified that there are several non-IAP ERVs and a LINE element which exhibit metastability. Identifying genuine epiphenomena in humans is challenging due to the lack of individuals having a shared homogeneous genetic background outside of monozygotic twins. Previously, using public data, I performed a screen for variably-methylated loci and found hundreds of loci showing methylation variability in the absence of local genetic variants. Methylation at these loci was then assessed in existing blood samples from over 500 children located in a region of The Gambia where annual seasonal changes in nutritional availability are known to affect circulating levels of metabolites involved in the DNA methylation pathway. Using this experimental design, I identified a testis-specific promoter having methylation which follows a date-of-conception pattern in males, indicating a possible readout of maternal nutritional status during early development. Genotype data from over 1 million SNP loci known to be prevalent in African populations was available for the majority of the individuals described in the above study. I identified that at a variably-methylated promoter of the PAX8-AS1 lncRNA gene, nearby genetic variants could restrict population-level methylation variability: individuals homozygous for the minor allele (haplotype) were all fully methylated while there was a wide range of methylation states among individuals homozygous for the major allele. I termed this phenomenon methylation variance QTL (mvQTL), and then looked for its presence throughout the genome. While canonical methylation QTL (where in a variant causes change in methylation averages but not variance) is common, the mvQTL phenomenon is confined to fewer loci. Through simulations, I found that a small number (4 to 5) of independent genetic variants which collectively determine a CpG's methylation state are sufficient to produce a near-continuous distribution of methylation. Combined with the discovery of mvQTL, these results indicate that existing human DNA methylation association studies may be severely discounting the role of genetics in driving apparent epigenotype-phenotype associations.","abstract_html":"During mammalian development, DNA methylation at most CpG sites throughout the genome is erased; these marks are then re-established later. In humans and in mice, typically all copies of a given CpG site in a particular tissue are methylated in all individuals; a large minority are unmethylated in all cases. A substantial proportion are methylated in some but not all tissues. A smaller minority of CpG sites are only partially methylated, and amongst these a proportion have consistent methylation levels across various tissues but variable methylation levels between individuals. Loci exhibiting this latter phenomenon, often referred to as metastable epialleles, were initially identified in inbred mice, where a lack of genetic variability indicates the presence of a genuine epimutation. Furthermore, the consistency of methylation across multiple tissues suggests a timing of methylation establishment which is early in development. These initial findings were located at IAP elements, a murine-specific class of ERV transposable element. There are approximately 10,000 IAP elements in the mouse genome, but the annotation of these elements is often poor and fractured. I developed an ad hoc algorithm to systematically fix the annotations, which reduced the proportion of fractured IAP elements from 36% to 19%. Using a genome-wide dataset, I updated an existing screen of variable methylation at IAP elements and identified that there are several non-IAP ERVs and a LINE element which exhibit metastability. Identifying genuine epiphenomena in humans is challenging due to the lack of individuals having a shared homogeneous genetic background outside of monozygotic twins. Previously, using public data, I performed a screen for variably-methylated loci and found hundreds of loci showing methylation variability in the absence of local genetic variants. Methylation at these loci was then assessed in existing blood samples from over 500 children located in a region of The Gambia where annual seasonal changes in nutritional availability are known to affect circulating levels of metabolites involved in the DNA methylation pathway. Using this experimental design, I identified a testis-specific promoter having methylation which follows a date-of-conception pattern in males, indicating a possible readout of maternal nutritional status during early development. Genotype data from over 1 million SNP loci known to be prevalent in African populations was available for the majority of the individuals described in the above study. I identified that at a variably-methylated promoter of the PAX8-AS1 lncRNA gene, nearby genetic variants could restrict population-level methylation variability: individuals homozygous for the minor allele (haplotype) were all fully methylated while there was a wide range of methylation states among individuals homozygous for the major allele. I termed this phenomenon methylation variance QTL (mvQTL), and then looked for its presence throughout the genome. While canonical methylation QTL (where in a variant causes change in methylation averages but not variance) is common, the mvQTL phenomenon is confined to fewer loci. Through simulations, I found that a small number (4 to 5) of independent genetic variants which collectively determine a CpG's methylation state are sufficient to produce a near-continuous distribution of methylation. Combined with the discovery of mvQTL, these results indicate that existing human DNA methylation association studies may be severely discounting the role of genetics in driving apparent epigenotype-phenotype associations.","abstract_has_math":false,"creators":["Kessler, Noah"],"institution":"University of Cambridge","degree_name":"Doctor of Philosophy (PhD)","degree_level":"Doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Ferguson-Smith, Anne"],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023-04-05","date_published":"2023-04-05","updated_at":"2026-07-22T22:24:25Z","subjects":["dna methylation","epigenetics","genetics"],"languages":["eng"],"rights":[],"rights_urls":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/ecec7b6a-684b-4cda-a114-9363d29b86d8/download","https://www.rioxx.net/licenses/all-rights-reserved/"],"identifier_entries":[{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000000227406663"],"render_values":[{"text":"0000-0002-2740-6663","href":"https://orcid.org/0000-0002-2740-6663","code":true}]}]},"links":{"outbound_url":"https://doi.org/10.17863/CAM.108526","outbound_label":"DOI","outbound_source":"dc:identifier.doi"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Ferguson-Smith, Anne"]},{"key":"dc:creator","label":"Author","values":["Kessler, Noah"]},{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000000227406663"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.issued","label":"Date","values":["2023-04-05"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["University of Cambridge"]},{"key":"dc:relation.isreferencedby.uri","label":"Dc Relation Isreferencedby URI","values":["https://www.repository.cam.ac.uk/handle/1810/368213"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["Doctor of Philosophy (PhD)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["dna methylation","epigenetics","genetics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/ecec7b6a-684b-4cda-a114-9363d29b86d8/download","https://www.rioxx.net/licenses/all-rights-reserved/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.17863/CAM.108526"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/112921c7-11e7-4eb0-93a4-2ecb7dd28673/download"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["During mammalian development, DNA methylation at most CpG sites throughout the genome is erased; these marks are then re-established later. In humans and in mice, typically all copies of a given CpG site in a particular tissue are methylated in all individuals; a large minority are unmethylated in all cases. A substantial proportion are methylated in some but not all tissues. A smaller minority of CpG sites are only partially methylated, and amongst these a proportion have consistent methylation levels across various tissues but variable methylation levels between individuals. Loci exhibiting this latter phenomenon, often referred to as metastable epialleles, were initially identified in inbred mice, where a lack of genetic variability indicates the presence of a genuine epimutation. Furthermore, the consistency of methylation across multiple tissues suggests a timing of methylation establishment which is early in development. These initial findings were located at IAP elements, a murine-specific class of ERV transposable element. There are approximately 10,000 IAP elements in the mouse genome, but the annotation of these elements is often poor and fractured. I developed an ad hoc algorithm to systematically fix the annotations, which reduced the proportion of fractured IAP elements from 36% to 19%. Using a genome-wide dataset, I updated an existing screen of variable methylation at IAP elements and identified that there are several non-IAP ERVs and a LINE element which exhibit metastability. Identifying genuine epiphenomena in humans is challenging due to the lack of individuals having a shared homogeneous genetic background outside of monozygotic twins. Previously, using public data, I performed a screen for variably-methylated loci and found hundreds of loci showing methylation variability in the absence of local genetic variants. Methylation at these loci was then assessed in existing blood samples from over 500 children located in a region of The Gambia where annual seasonal changes in nutritional availability are known to affect circulating levels of metabolites involved in the DNA methylation pathway. Using this experimental design, I identified a testis-specific promoter having methylation which follows a date-of-conception pattern in males, indicating a possible readout of maternal nutritional status during early development. Genotype data from over 1 million SNP loci known to be prevalent in African populations was available for the majority of the individuals described in the above study. I identified that at a variably-methylated promoter of the PAX8-AS1 lncRNA gene, nearby genetic variants could restrict population-level methylation variability: individuals homozygous for the minor allele (haplotype) were all fully methylated while there was a wide range of methylation states among individuals homozygous for the major allele. I termed this phenomenon methylation variance QTL (mvQTL), and then looked for its presence throughout the genome. While canonical methylation QTL (where in a variant causes change in methylation averages but not variance) is common, the mvQTL phenomenon is confined to fewer loci. Through simulations, I found that a small number (4 to 5) of independent genetic variants which collectively determine a CpG's methylation state are sufficient to produce a near-continuous distribution of methylation. Combined with the discovery of mvQTL, these results indicate that existing human DNA methylation association studies may be severely discounting the role of genetics in driving apparent epigenotype-phenotype associations."]},{"key":"dc:format.checksum.md5","label":"Dc Format Checksum Md5","values":["4c481bd973298dc47d109132f73206f3","87eda9de84448d1f82354d60eee3eb5f"]},{"key":"dc:title","label":"Title","values":["DNA methylation variability at the crossroads of stochasticity, genetics, and environment"]}]}],"canonical_facts":{"dc:contributor.advisor":["Ferguson-Smith, Anne"],"dc:creator":["Kessler, Noah"],"dc:creator.authoridentifier":["0000000227406663"],"dc:date.issued":["2023-04-05"],"dc:description.abstract":["During mammalian development, DNA methylation at most CpG sites throughout the genome is erased; these marks are then re-established later. In humans and in mice, typically all copies of a given CpG site in a particular tissue are methylated in all individuals; a large minority are unmethylated in all cases. A substantial proportion are methylated in some but not all tissues. A smaller minority of CpG sites are only partially methylated, and amongst these a proportion have consistent methylation levels across various tissues but variable methylation levels between individuals. Loci exhibiting this latter phenomenon, often referred to as metastable epialleles, were initially identified in inbred mice, where a lack of genetic variability indicates the presence of a genuine epimutation. Furthermore, the consistency of methylation across multiple tissues suggests a timing of methylation establishment which is early in development. These initial findings were located at IAP elements, a murine-specific class of ERV transposable element. There are approximately 10,000 IAP elements in the mouse genome, but the annotation of these elements is often poor and fractured. I developed an ad hoc algorithm to systematically fix the annotations, which reduced the proportion of fractured IAP elements from 36% to 19%. Using a genome-wide dataset, I updated an existing screen of variable methylation at IAP elements and identified that there are several non-IAP ERVs and a LINE element which exhibit metastability. Identifying genuine epiphenomena in humans is challenging due to the lack of individuals having a shared homogeneous genetic background outside of monozygotic twins. Previously, using public data, I performed a screen for variably-methylated loci and found hundreds of loci showing methylation variability in the absence of local genetic variants. Methylation at these loci was then assessed in existing blood samples from over 500 children located in a region of The Gambia where annual seasonal changes in nutritional availability are known to affect circulating levels of metabolites involved in the DNA methylation pathway. Using this experimental design, I identified a testis-specific promoter having methylation which follows a date-of-conception pattern in males, indicating a possible readout of maternal nutritional status during early development. Genotype data from over 1 million SNP loci known to be prevalent in African populations was available for the majority of the individuals described in the above study. I identified that at a variably-methylated promoter of the PAX8-AS1 lncRNA gene, nearby genetic variants could restrict population-level methylation variability: individuals homozygous for the minor allele (haplotype) were all fully methylated while there was a wide range of methylation states among individuals homozygous for the major allele. I termed this phenomenon methylation variance QTL (mvQTL), and then looked for its presence throughout the genome. While canonical methylation QTL (where in a variant causes change in methylation averages but not variance) is common, the mvQTL phenomenon is confined to fewer loci. Through simulations, I found that a small number (4 to 5) of independent genetic variants which collectively determine a CpG's methylation state are sufficient to produce a near-continuous distribution of methylation. Combined with the discovery of mvQTL, these results indicate that existing human DNA methylation association studies may be severely discounting the role of genetics in driving apparent epigenotype-phenotype associations."],"dc:format.checksum.md5":["4c481bd973298dc47d109132f73206f3","87eda9de84448d1f82354d60eee3eb5f"],"dc:identifier.doi":["https://doi.org/10.17863/CAM.108526"],"dc:identifier.uri":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/112921c7-11e7-4eb0-93a4-2ecb7dd28673/download"],"dc:language":["eng"],"dc:publisher.institution":["University of Cambridge"],"dc:relation.isreferencedby.uri":["https://www.repository.cam.ac.uk/handle/1810/368213"],"dc:rights":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/ecec7b6a-684b-4cda-a114-9363d29b86d8/download","https://www.rioxx.net/licenses/all-rights-reserved/"],"dc:subject":["dna methylation","epigenetics","genetics"],"dc:title":["DNA methylation variability at the crossroads of stochasticity, genetics, and environment"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["Doctor of Philosophy (PhD)"]},"updated_at":"2026-07-22T22:24:25Z"}