國立虎尾科技大學 |

Integrative Polygenic Risk Prediction of Coronary Artery Disease.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Integrative Polygenic Risk Prediction of Coronary Artery Disease./
Author:	Chen, Shang-Fu.
Description:	1 online resource (630 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
Contained By:	Dissertations Abstracts International85-12B.
Subject:	Biology. -
Online resource:	click for full text (PQDT)
ISBN:	9798383164327

Integrative Polygenic Risk Prediction of Coronary Artery Disease.
Chen, Shang-Fu.

Integrative Polygenic Risk Prediction of Coronary Artery Disease. - 1 online resource (630 pages)

Source: Dissertations Abstracts International, Volume: 85-12, Section: B.

Thesis (Ph.D.)--The Scripps Research Institute, 2024.

Includes bibliographical references

Coronary artery disease (CAD) remains a leading cause of global morbidity and mortality, characterized by its multifactorial etiology and the challenge it poses for effective prevention. Current guidelines for primary prevention, or the prevention of a first acute event, include relatively simple risk assessment and leave substantial room for improvement both for risk ascertainment and selection of prevention strategies. This thesis aims to leverage the advances in the power of extensive data available from biobanks to build a comprehensive risk predictive model and suggest a step forward in refining risk assessment and personalized prevention approaches for CAD. The contemporary risk equations are basic to identify patients at high risk. These risk scores are mainly driven by age and sex, with slight modifications for specific health indicators, lifestyle choices, and existing diseases. Yet, these methods incompletely recapitulate genetic predisposition through granular categories like family history and ethnic background. While this serves as a starting point, it falls short to quantify an individual's genetic susceptibility and does not distinguish patients at high polygenic risk . Despite being grounded in epidemiological evidence; the current methods miss out on the opportunity to customize prevention plans based on a person's unique genetic makeup.Similarly, cardiac calcium scoring , by detecting calcification, provides crucial insights into arterial health and aids in the decision-making process for statin therapy, particularly for individuals in the intermediate risk category where treatment pathways are not clear-cut. Despite its significant value, the application of this method is restricted; it is not generally recommended for those with low clinical risk scores, and its utility varies across different population groups. This predicament reflects similar constraints faced by other tests, such as lipoprotein panels, assessing cholesterol levels; estimated glomerular filtration rate (eGFR), which evaluates kidney function based on serum creatinine; and high-sensitivity C-reactive protein (hsCRP) , that indicates inflammation conditions. Each test offers distinct perspectives on cardiovascular risk factors, yet they share the common challenge of how to effectively incorporate information form varied health indicators into a unified risk assessment framework. This emphasizes the need for a broader, more inclusive approach to evaluating cardiovascular risk. While it improves the understanding of the disease's present condition, it should be integrated with a full risk evaluation, promoting a collaborative approach to risk screening. Such an approach should encompass a wide spectrum of risk factors, thereby optimizing the allocation of medical resources beyond a singular diagnostic tool. The emergence of polygenic risk scores (PRS) represents a significant advancement, offering the potential to overcome these traditional limitations in CAD risk prediction . Derived from a one-time DNA sequencing, PRS calculates an individual's risk based on the cumulative effect of an individual's genetic variants, each weighted according to its association with the disease10. This method, efficient and grounded in the individual's genetic blueprint from the earliest stages of life, allows for the early identification of those at elevated risk for CAD. Recent studies highlight the ability of PRS to significantly improve risk stratification beyond traditional models, identifying individuals at high risk of CAD who might benefit from more proactive preventive treatments, even in the absence of conventional risk factors. However, the integration of PRS into clinical practice faces challenges, particularly in its interpretation and application, which currently involves categorizing individuals to identify those at high risk, alongside established risk factors and diagnostic tools. Moreover, the clinical utility of PRS for personalized prevention strategies depends on its accuracy in predicting CAD risk across individuals with different background. This requires not only comprehensive validation but also proper recalibration, potentially involving nonlinear integration techniques. This emphasizes the critical need to embed PRS within advanced models to refine and adapt these predictions effectively, thereby optimizing the deployment of personalized preventive interventions.Altogether, artificial intelligence (AI) offers promising advancements in utilizing high-dimensional data to enhance the understanding of the complex interplay of CAD risk factors and develop a generalized risk prediction model. Beyond its application in cardiac imaging, AI's potential in multimodal risk prediction, incorporating standard biomarkers, genetic and other omics data, various biosensors, and electronic health records (EHRs), is being increasingly recognized. However, clinical explanation of AI models remains an area in need of further exploration, particularly in translating complex risk predictions into actionable therapeutic interventions rather than abstract probabilities.To satisfy the critical need, the recent availability of nation-scale biobank datasets marks a transformative moment, offering the extensive, diverse data that AI models crave. These biobanks not only feed the data-hungry AI algorithms but also enrich the models with a variety of features that reflect the complex interplay of genetic, lifestyle, and environmental factors influencing CAD risk. This rich dataset foundation allows AI to advance from generating probabilistic risk assessments to enabling the testing and validation of actionable effects, thereby making personalized, data-driven prevention strategies a reality.The overarching goal of this thesis is to apply the state-of-art machine learning techniques to develop a CAD risk assessment tool, with the capability to prioritize the tailored actions for CAD primary prevention.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798383164327Subjects--Topical Terms:

599573
Biology.
Subjects--Index Terms:

Coronary artery diseaseIndex Terms--Genre/Form:

554714
Electronic books.

Integrative Polygenic Risk Prediction of Coronary Artery Disease.
LDR:07408ntm a22004097 4500 001 1150211
005 20241022111615.5
006 m o d
007 cr bn ---uuuuu
008 250605s2024 xx obm 000 0 eng d
020 $a 9798383164327
035 $a (MiAaPQ)AAI31300675
035 $a AAI31300675
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Chen, Shang-Fu. $3 1476652
245 1 0 $a Integrative Polygenic Risk Prediction of Coronary Artery Disease.
264 0 $c 2024
300 $a 1 online resource (630 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
500 $a Advisor: Torkamani, Ali.
502 $a Thesis (Ph.D.)--The Scripps Research Institute, 2024.
504 $a Includes bibliographical references
520 $a Coronary artery disease (CAD) remains a leading cause of global morbidity and mortality, characterized by its multifactorial etiology and the challenge it poses for effective prevention. Current guidelines for primary prevention, or the prevention of a first acute event, include relatively simple risk assessment and leave substantial room for improvement both for risk ascertainment and selection of prevention strategies. This thesis aims to leverage the advances in the power of extensive data available from biobanks to build a comprehensive risk predictive model and suggest a step forward in refining risk assessment and personalized prevention approaches for CAD. The contemporary risk equations are basic to identify patients at high risk. These risk scores are mainly driven by age and sex, with slight modifications for specific health indicators, lifestyle choices, and existing diseases. Yet, these methods incompletely recapitulate genetic predisposition through granular categories like family history and ethnic background. While this serves as a starting point, it falls short to quantify an individual's genetic susceptibility and does not distinguish patients at high polygenic risk . Despite being grounded in epidemiological evidence; the current methods miss out on the opportunity to customize prevention plans based on a person's unique genetic makeup.Similarly, cardiac calcium scoring , by detecting calcification, provides crucial insights into arterial health and aids in the decision-making process for statin therapy, particularly for individuals in the intermediate risk category where treatment pathways are not clear-cut. Despite its significant value, the application of this method is restricted; it is not generally recommended for those with low clinical risk scores, and its utility varies across different population groups. This predicament reflects similar constraints faced by other tests, such as lipoprotein panels, assessing cholesterol levels; estimated glomerular filtration rate (eGFR), which evaluates kidney function based on serum creatinine; and high-sensitivity C-reactive protein (hsCRP) , that indicates inflammation conditions. Each test offers distinct perspectives on cardiovascular risk factors, yet they share the common challenge of how to effectively incorporate information form varied health indicators into a unified risk assessment framework. This emphasizes the need for a broader, more inclusive approach to evaluating cardiovascular risk. While it improves the understanding of the disease's present condition, it should be integrated with a full risk evaluation, promoting a collaborative approach to risk screening. Such an approach should encompass a wide spectrum of risk factors, thereby optimizing the allocation of medical resources beyond a singular diagnostic tool. The emergence of polygenic risk scores (PRS) represents a significant advancement, offering the potential to overcome these traditional limitations in CAD risk prediction . Derived from a one-time DNA sequencing, PRS calculates an individual's risk based on the cumulative effect of an individual's genetic variants, each weighted according to its association with the disease10. This method, efficient and grounded in the individual's genetic blueprint from the earliest stages of life, allows for the early identification of those at elevated risk for CAD. Recent studies highlight the ability of PRS to significantly improve risk stratification beyond traditional models, identifying individuals at high risk of CAD who might benefit from more proactive preventive treatments, even in the absence of conventional risk factors. However, the integration of PRS into clinical practice faces challenges, particularly in its interpretation and application, which currently involves categorizing individuals to identify those at high risk, alongside established risk factors and diagnostic tools. Moreover, the clinical utility of PRS for personalized prevention strategies depends on its accuracy in predicting CAD risk across individuals with different background. This requires not only comprehensive validation but also proper recalibration, potentially involving nonlinear integration techniques. This emphasizes the critical need to embed PRS within advanced models to refine and adapt these predictions effectively, thereby optimizing the deployment of personalized preventive interventions.Altogether, artificial intelligence (AI) offers promising advancements in utilizing high-dimensional data to enhance the understanding of the complex interplay of CAD risk factors and develop a generalized risk prediction model. Beyond its application in cardiac imaging, AI's potential in multimodal risk prediction, incorporating standard biomarkers, genetic and other omics data, various biosensors, and electronic health records (EHRs), is being increasingly recognized. However, clinical explanation of AI models remains an area in need of further exploration, particularly in translating complex risk predictions into actionable therapeutic interventions rather than abstract probabilities.To satisfy the critical need, the recent availability of nation-scale biobank datasets marks a transformative moment, offering the extensive, diverse data that AI models crave. These biobanks not only feed the data-hungry AI algorithms but also enrich the models with a variety of features that reflect the complex interplay of genetic, lifestyle, and environmental factors influencing CAD risk. This rich dataset foundation allows AI to advance from generating probabilistic risk assessments to enabling the testing and validation of actionable effects, thereby making personalized, data-driven prevention strategies a reality.The overarching goal of this thesis is to apply the state-of-art machine learning techniques to develop a CAD risk assessment tool, with the capability to prioritize the tailored actions for CAD primary prevention.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Biology. $3 599573
650 4 $a Bioinformatics. $3 583857
653 $a Coronary artery disease
653 $a Genotype imputation
653 $a Machine learning
653 $a Meta-prediction
653 $a Polygenic risk prediction
653 $a Prospective prediction
655 7 $a Electronic books. $2 local $3 554714
690 $a 0306
690 $a 0715
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a The Scripps Research Institute. $b Computational Biology/Bioinformatics. $3 1476653
773 0 $t Dissertations Abstracts International $g 85-12B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31300675 $z click for full text (PQDT)