{"id":{"repo_id":"edinburgh","oai_identifier":"oai:era.ed.ac.uk:1842/41802"},"canonical_url":"https://search.dev.ndltd.org/etd/edinburgh/oai:era.ed.ac.uk:1842/41802","repository":{"repo_id":"edinburgh","name":"University of Edinburgh","base_url":"https://era.ed.ac.uk/server/oai/request"},"display":{"title":"Understanding and stratifying brain health through blood-based omics data","abstract":"Brain health across the lifespan is dynamic and influenced by a complex interplay of genetics and the environment. Age-related neurological diseases are a growing burden on healthcare systems and society. Individuals that do not have overt diagnoses of neurological diseases will still experience declines in cognitive ability and reductions in grey matter volumes as they age. Understanding why some brains are healthier than others may provide insight into targets for the preservation of brain health. Early identification of individuals at high-risk of neurological diseases is a priority for preventative strategies. Proteins are the effector molecules of disease in the body and are often the targets of therapeutic interventions. Blood samples can be used to derive measures of many thousands of blood proteins and are routinely collected in clinical and research settings. Another measure available from blood is DNA methylation (DNAm), which is an epigenetic mechanism that can regulate gene expression and protein levels. DNAm is thought to record the body’s response to a range of biological and environmental factors. The first aim of this thesis is to perform methylome-wide association studies (MWAS) of circulating proteins, with a focus on those related to brain health. DNAm patterns can also be used to derive proxy scores for protein levels – an approach that is somewhat analogous to polygenic scores. These proxies are known as protein epigenetic scores (or EpiScores). In some cases, protein EpiScores outperform measured proteins in associations with brain imaging and lifestyle traits, possibly due to the relative lack of stability observed across some single time point protein measurements. Protein EpiScores could represent biomarkers for risk stratification. However, this has not been examined at scale. Consequently, the second aim of this thesis is to develop a comprehensive set of protein EpiScores and evaluate them as tools for risk stratification. For neurological diseases such as Alzheimer’s dementia, damage is thought to occur in the brain decades prior to symptom presentation. Dysfunction at the blood brain barrier can facilitate leakage of proteins into the bloodstream in the early stages of neurological disease. Similarly, peripherally-produced proteins may also serve as warning signatures. Therefore, the final aim of this thesis is to conduct an assessment of blood protein signatures of incident neurological diseases and associated morbidities. In Chapters 1-3, I provide an overview of brain health and disease, blood-based molecular measures and key statistical approaches. In Chapter 4, I detail the population cohorts used in this thesis, before outlining my research aims and chosen methodologies in Chapter 5. In Chapter 6, I study serum measurements of S100 calcium-binding protein β (S100β) – a well-characterised marker with links to neuroinflammation and brain disease. I map the epigenetic and genetic signatures of this protein and test for evidence of a putative causal relationship between the protein and Alzheimer’s dementia. Chapter 7 extends this approach via a proteome-wide analysis. Instead of focusing on a single candidate biomarker, I conduct MWAS of 4,235 plasma proteins, identifying 2,928 associations (n ≥ 778 individuals). I also scan the proteome against fifteen brain health traits, identifying 405 associations involving 191 proteins. I integrate these signatures to highlight potential pathways between the methylome and proteome that may have relevance to brain health. In Chapter 8, I consider 953 possible plasma proteins for protein EpiScore development (N in the training set ranged from 706 to 944 individuals). I evaluate these EpiScores in independent populations, with 109 statistically significant EpiScores taken forward and modelled as biomarkers of incident diseases in 9,537 individuals. I build on this work in Chapter 9, where I generate protein EpiScores for GDF15 and NT-proBNP, which are two leading cardiovascular disease (CVD) markers implicated in brain health. I use a much-expanded sample size (n ≥ 16,963) to train these scores and show that they replicate protein-disease associations and associate with brain health outcomes. Finally, in Chapter 10, I test individual protein associations with 23 incident morbidities and death. I map whether proteins are markers for multiple neurological diseases, or specific to singular diseases. I then create ProteinScores for 10-year onset stratification of each incident outcome. ProteinScores for Alzheimer’s dementia and Parkinson’s disease are amongst the best-performing 10-year onset scores. The work done in this thesis provides information to help us identify those at the highest risk of developing neurological diseases (and associated morbidities), up to a decade prior to onset. My findings also tell us about the individual protein and DNAm patterns that associate with brain health and disease. Taken together, these results indicate that profiling epigenetic and proteomic information from our blood may improve our understanding of brain ageing. This work sits within the ethos of early detection and prevention, which should be at the heart of healthcare as we age.","abstract_html":"Brain health across the lifespan is dynamic and influenced by a complex interplay of genetics and the environment. Age-related neurological diseases are a growing burden on healthcare systems and society. Individuals that do not have overt diagnoses of neurological diseases will still experience declines in cognitive ability and reductions in grey matter volumes as they age. Understanding why some brains are healthier than others may provide insight into targets for the preservation of brain health. Early identification of individuals at high-risk of neurological diseases is a priority for preventative strategies. Proteins are the effector molecules of disease in the body and are often the targets of therapeutic interventions. Blood samples can be used to derive measures of many thousands of blood proteins and are routinely collected in clinical and research settings. Another measure available from blood is DNA methylation (DNAm), which is an epigenetic mechanism that can regulate gene expression and protein levels. DNAm is thought to record the body’s response to a range of biological and environmental factors. The first aim of this thesis is to perform methylome-wide association studies (MWAS) of circulating proteins, with a focus on those related to brain health. DNAm patterns can also be used to derive proxy scores for protein levels – an approach that is somewhat analogous to polygenic scores. These proxies are known as protein epigenetic scores (or EpiScores). In some cases, protein EpiScores outperform measured proteins in associations with brain imaging and lifestyle traits, possibly due to the relative lack of stability observed across some single time point protein measurements. Protein EpiScores could represent biomarkers for risk stratification. However, this has not been examined at scale. Consequently, the second aim of this thesis is to develop a comprehensive set of protein EpiScores and evaluate them as tools for risk stratification. For neurological diseases such as Alzheimer’s dementia, damage is thought to occur in the brain decades prior to symptom presentation. Dysfunction at the blood brain barrier can facilitate leakage of proteins into the bloodstream in the early stages of neurological disease. Similarly, peripherally-produced proteins may also serve as warning signatures. Therefore, the final aim of this thesis is to conduct an assessment of blood protein signatures of incident neurological diseases and associated morbidities. In Chapters 1-3, I provide an overview of brain health and disease, blood-based molecular measures and key statistical approaches. In Chapter 4, I detail the population cohorts used in this thesis, before outlining my research aims and chosen methodologies in Chapter 5. In Chapter 6, I study serum measurements of S100 calcium-binding protein β (S100β) – a well-characterised marker with links to neuroinflammation and brain disease. I map the epigenetic and genetic signatures of this protein and test for evidence of a putative causal relationship between the protein and Alzheimer’s dementia. Chapter 7 extends this approach via a proteome-wide analysis. Instead of focusing on a single candidate biomarker, I conduct MWAS of 4,235 plasma proteins, identifying 2,928 associations (n ≥ 778 individuals). I also scan the proteome against fifteen brain health traits, identifying 405 associations involving 191 proteins. I integrate these signatures to highlight potential pathways between the methylome and proteome that may have relevance to brain health. In Chapter 8, I consider 953 possible plasma proteins for protein EpiScore development (N in the training set ranged from 706 to 944 individuals). I evaluate these EpiScores in independent populations, with 109 statistically significant EpiScores taken forward and modelled as biomarkers of incident diseases in 9,537 individuals. I build on this work in Chapter 9, where I generate protein EpiScores for GDF15 and NT-proBNP, which are two leading cardiovascular disease (CVD) markers implicated in brain health. I use a much-expanded sample size (n ≥ 16,963) to train these scores and show that they replicate protein-disease associations and associate with brain health outcomes. Finally, in Chapter 10, I test individual protein associations with 23 incident morbidities and death. I map whether proteins are markers for multiple neurological diseases, or specific to singular diseases. I then create ProteinScores for 10-year onset stratification of each incident outcome. ProteinScores for Alzheimer’s dementia and Parkinson’s disease are amongst the best-performing 10-year onset scores. The work done in this thesis provides information to help us identify those at the highest risk of developing neurological diseases (and associated morbidities), up to a decade prior to onset. My findings also tell us about the individual protein and DNAm patterns that associate with brain health and disease. Taken together, these results indicate that profiling epigenetic and proteomic information from our blood may improve our understanding of brain ageing. This work sits within the ethos of early detection and prevention, which should be at the heart of healthcare as we age.","abstract_has_math":false,"creators":["Gadd, Danielle Alisha"],"institution":"The University of Edinburgh","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Marioni, Riccardo","Muniz Terrera, Graciela","Oyarzun, Diego","Veronique, Vitart"],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-05-22","date_published":"2024-05-22","updated_at":"2026-07-24T02:14:07Z","subjects":["DNA methylation","DNAm","methylome-wide association studies","EpiScores","biomarkers for risk stratification","risk stratification"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://dx.doi.org/10.7488/era/4525"],"render_values":[{"text":"http://dx.doi.org/10.7488/era/4525","href":"http://dx.doi.org/10.7488/era/4525","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/1842/41802","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Marioni, Riccardo","Muniz Terrera, Graciela","Oyarzun, Diego","Veronique, Vitart"]},{"key":"dc:creator","label":"Author","values":["Gadd, Danielle Alisha"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2024-05-22T10:30:44Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2024-05-22T10:30:44Z"]},{"key":"dc:date.issued","label":"Date","values":["2024-05-22"]},{"key":"dc:publisher","label":"Institution","values":["The University of Edinburgh"]},{"key":"dc:type","label":"Dc Type","values":["Thesis or Dissertation"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD Doctor of Philosophy"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["DNA methylation","DNAm","methylome-wide association studies","EpiScores","biomarkers for risk stratification","risk stratification"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1842/41802","http://dx.doi.org/10.7488/era/4525"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Brain health across the lifespan is dynamic and influenced by a complex interplay of genetics and the environment. Age-related neurological diseases are a growing burden on healthcare systems and society. Individuals that do not have overt diagnoses of neurological diseases will still experience declines in cognitive ability and reductions in grey matter volumes as they age. Understanding why some brains are healthier than others may provide insight into targets for the preservation of brain health. Early identification of individuals at high-risk of neurological diseases is a priority for preventative strategies. Proteins are the effector molecules of disease in the body and are often the targets of therapeutic interventions. Blood samples can be used to derive measures of many thousands of blood proteins and are routinely collected in clinical and research settings. Another measure available from blood is DNA methylation (DNAm), which is an epigenetic mechanism that can regulate gene expression and protein levels. DNAm is thought to record the body’s response to a range of biological and environmental factors. The first aim of this thesis is to perform methylome-wide association studies (MWAS) of circulating proteins, with a focus on those related to brain health. DNAm patterns can also be used to derive proxy scores for protein levels – an approach that is somewhat analogous to polygenic scores. These proxies are known as protein epigenetic scores (or EpiScores). In some cases, protein EpiScores outperform measured proteins in associations with brain imaging and lifestyle traits, possibly due to the relative lack of stability observed across some single time point protein measurements. Protein EpiScores could represent biomarkers for risk stratification. However, this has not been examined at scale. Consequently, the second aim of this thesis is to develop a comprehensive set of protein EpiScores and evaluate them as tools for risk stratification. For neurological diseases such as Alzheimer’s dementia, damage is thought to occur in the brain decades prior to symptom presentation. Dysfunction at the blood brain barrier can facilitate leakage of proteins into the bloodstream in the early stages of neurological disease. Similarly, peripherally-produced proteins may also serve as warning signatures. Therefore, the final aim of this thesis is to conduct an assessment of blood protein signatures of incident neurological diseases and associated morbidities. In Chapters 1-3, I provide an overview of brain health and disease, blood-based molecular measures and key statistical approaches. In Chapter 4, I detail the population cohorts used in this thesis, before outlining my research aims and chosen methodologies in Chapter 5. In Chapter 6, I study serum measurements of S100 calcium-binding protein β (S100β) – a well-characterised marker with links to neuroinflammation and brain disease. I map the epigenetic and genetic signatures of this protein and test for evidence of a putative causal relationship between the protein and Alzheimer’s dementia. Chapter 7 extends this approach via a proteome-wide analysis. Instead of focusing on a single candidate biomarker, I conduct MWAS of 4,235 plasma proteins, identifying 2,928 associations (n ≥ 778 individuals). I also scan the proteome against fifteen brain health traits, identifying 405 associations involving 191 proteins. I integrate these signatures to highlight potential pathways between the methylome and proteome that may have relevance to brain health. In Chapter 8, I consider 953 possible plasma proteins for protein EpiScore development (N in the training set ranged from 706 to 944 individuals). I evaluate these EpiScores in independent populations, with 109 statistically significant EpiScores taken forward and modelled as biomarkers of incident diseases in 9,537 individuals. I build on this work in Chapter 9, where I generate protein EpiScores for GDF15 and NT-proBNP, which are two leading cardiovascular disease (CVD) markers implicated in brain health. I use a much-expanded sample size (n ≥ 16,963) to train these scores and show that they replicate protein-disease associations and associate with brain health outcomes. Finally, in Chapter 10, I test individual protein associations with 23 incident morbidities and death. I map whether proteins are markers for multiple neurological diseases, or specific to singular diseases. I then create ProteinScores for 10-year onset stratification of each incident outcome. ProteinScores for Alzheimer’s dementia and Parkinson’s disease are amongst the best-performing 10-year onset scores. The work done in this thesis provides information to help us identify those at the highest risk of developing neurological diseases (and associated morbidities), up to a decade prior to onset. My findings also tell us about the individual protein and DNAm patterns that associate with brain health and disease. Taken together, these results indicate that profiling epigenetic and proteomic information from our blood may improve our understanding of brain ageing. This work sits within the ethos of early detection and prevention, which should be at the heart of healthcare as we age."]},{"key":"dc:title","label":"Title","values":["Understanding and stratifying brain health through blood-based omics data"]}]}],"canonical_facts":{"dc:contributor.advisor":["Marioni, Riccardo","Muniz Terrera, Graciela","Oyarzun, Diego","Veronique, Vitart"],"dc:creator":["Gadd, Danielle Alisha"],"dc:date.accessioned":["2024-05-22T10:30:44Z"],"dc:date.available":["2024-05-22T10:30:44Z"],"dc:date.issued":["2024-05-22"],"dc:description.abstract":["Brain health across the lifespan is dynamic and influenced by a complex interplay of genetics and the environment. Age-related neurological diseases are a growing burden on healthcare systems and society. Individuals that do not have overt diagnoses of neurological diseases will still experience declines in cognitive ability and reductions in grey matter volumes as they age. Understanding why some brains are healthier than others may provide insight into targets for the preservation of brain health. Early identification of individuals at high-risk of neurological diseases is a priority for preventative strategies. Proteins are the effector molecules of disease in the body and are often the targets of therapeutic interventions. Blood samples can be used to derive measures of many thousands of blood proteins and are routinely collected in clinical and research settings. Another measure available from blood is DNA methylation (DNAm), which is an epigenetic mechanism that can regulate gene expression and protein levels. DNAm is thought to record the body’s response to a range of biological and environmental factors. The first aim of this thesis is to perform methylome-wide association studies (MWAS) of circulating proteins, with a focus on those related to brain health. DNAm patterns can also be used to derive proxy scores for protein levels – an approach that is somewhat analogous to polygenic scores. These proxies are known as protein epigenetic scores (or EpiScores). In some cases, protein EpiScores outperform measured proteins in associations with brain imaging and lifestyle traits, possibly due to the relative lack of stability observed across some single time point protein measurements. Protein EpiScores could represent biomarkers for risk stratification. However, this has not been examined at scale. Consequently, the second aim of this thesis is to develop a comprehensive set of protein EpiScores and evaluate them as tools for risk stratification. For neurological diseases such as Alzheimer’s dementia, damage is thought to occur in the brain decades prior to symptom presentation. Dysfunction at the blood brain barrier can facilitate leakage of proteins into the bloodstream in the early stages of neurological disease. Similarly, peripherally-produced proteins may also serve as warning signatures. Therefore, the final aim of this thesis is to conduct an assessment of blood protein signatures of incident neurological diseases and associated morbidities. In Chapters 1-3, I provide an overview of brain health and disease, blood-based molecular measures and key statistical approaches. In Chapter 4, I detail the population cohorts used in this thesis, before outlining my research aims and chosen methodologies in Chapter 5. In Chapter 6, I study serum measurements of S100 calcium-binding protein β (S100β) – a well-characterised marker with links to neuroinflammation and brain disease. I map the epigenetic and genetic signatures of this protein and test for evidence of a putative causal relationship between the protein and Alzheimer’s dementia. Chapter 7 extends this approach via a proteome-wide analysis. Instead of focusing on a single candidate biomarker, I conduct MWAS of 4,235 plasma proteins, identifying 2,928 associations (n ≥ 778 individuals). I also scan the proteome against fifteen brain health traits, identifying 405 associations involving 191 proteins. I integrate these signatures to highlight potential pathways between the methylome and proteome that may have relevance to brain health. In Chapter 8, I consider 953 possible plasma proteins for protein EpiScore development (N in the training set ranged from 706 to 944 individuals). I evaluate these EpiScores in independent populations, with 109 statistically significant EpiScores taken forward and modelled as biomarkers of incident diseases in 9,537 individuals. I build on this work in Chapter 9, where I generate protein EpiScores for GDF15 and NT-proBNP, which are two leading cardiovascular disease (CVD) markers implicated in brain health. I use a much-expanded sample size (n ≥ 16,963) to train these scores and show that they replicate protein-disease associations and associate with brain health outcomes. Finally, in Chapter 10, I test individual protein associations with 23 incident morbidities and death. I map whether proteins are markers for multiple neurological diseases, or specific to singular diseases. I then create ProteinScores for 10-year onset stratification of each incident outcome. ProteinScores for Alzheimer’s dementia and Parkinson’s disease are amongst the best-performing 10-year onset scores. The work done in this thesis provides information to help us identify those at the highest risk of developing neurological diseases (and associated morbidities), up to a decade prior to onset. My findings also tell us about the individual protein and DNAm patterns that associate with brain health and disease. Taken together, these results indicate that profiling epigenetic and proteomic information from our blood may improve our understanding of brain ageing. This work sits within the ethos of early detection and prevention, which should be at the heart of healthcare as we age."],"dc:identifier.uri":["https://hdl.handle.net/1842/41802","http://dx.doi.org/10.7488/era/4525"],"dc:language.iso":["en"],"dc:publisher":["The University of Edinburgh"],"dc:subject":["DNA methylation","DNAm","methylome-wide association studies","EpiScores","biomarkers for risk stratification","risk stratification"],"dc:title":["Understanding and stratifying brain health through blood-based omics data"],"dc:type":["Thesis or Dissertation"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["PhD Doctor of Philosophy"]},"updated_at":"2026-07-24T02:14:07Z"}