Integrative analysis of common and rare variants for a more comprehensive genetic risk assessment
Integrative analysis of common and rare variants for a more comprehensive genetic risk assessment
View/Type of Scholarly Publication
DissertationDate of Exam
07.11.2024Date of Publication
12.11.2024Advisor
Krawitz, PeterCo-Referee
Fröhlich, HolgerMetadata
Show full item record
Citable Links 
Abstract
The etiology of complex traits is difficult to interpret because of their multifactorial nature. And while environmental factors play an important role in their development, genetic factors also have huge and crucial effect on the expression of complex phenotypes. However, a relevant part of the genetic landscape is yet to be discovered, despite the century long research and studies on the topic. The aim of this thesis is to investigate the role of rare pathogenic variants in complex traits and integrate their analysis with common variants for a more comprehensive genetic risk assessment.
In the first paper, we introduce an open source python package, GenRisk, that implements gene-based burden scores, focusing on rare deleterious variants, and polygenic risk scores (PRS), which are based on common variants. GenRisk’s pipeline also contains an association analysis function using different regression models and a genetic risk modeling function utilizing multiple machine learning models. In this paper, we applied the pipeline on samples from UK Biobank as a usage case example.
The second paper employs the GenRisk framework to explore 28 blood biomarkers within the UK Biobank cohort. We performed the PRS calculation using genotyping data while exome data was used for the gene-based scores (GBS) calculation. Association analysis was done using linear regression and genetic risk prediction models were also generated with either PRS, GBS, or both. We were able to show that rare pathogenic variants play an important role at an individual level, but the traditional PRS could be more informative when predicting the genetic risk at a population level.
In the last paper, we conduct a more thorough analysis on 72,469 samples from UK Biobank to investigate the rare-variants influence on male-pattern hair loss (MPHL). Novel candidate genes were identified including HEPH, CEPT1 and EIF3F, further proving that rare variants contribute to the genetic landscape of complex phenotypes like male-pattern hair loss.
In conclusion, our findings indicate that rare deleterious variants have an essential role in complex phenotypes, and can be analysed to discover new targets in these traits. Nonetheless, further investigation needs to be conducted to effectively integrate the effects of rare and common variants, ultimately improving comprehensive genetic risk assessment strategies.
In the first paper, we introduce an open source python package, GenRisk, that implements gene-based burden scores, focusing on rare deleterious variants, and polygenic risk scores (PRS), which are based on common variants. GenRisk’s pipeline also contains an association analysis function using different regression models and a genetic risk modeling function utilizing multiple machine learning models. In this paper, we applied the pipeline on samples from UK Biobank as a usage case example.
The second paper employs the GenRisk framework to explore 28 blood biomarkers within the UK Biobank cohort. We performed the PRS calculation using genotyping data while exome data was used for the gene-based scores (GBS) calculation. Association analysis was done using linear regression and genetic risk prediction models were also generated with either PRS, GBS, or both. We were able to show that rare pathogenic variants play an important role at an individual level, but the traditional PRS could be more informative when predicting the genetic risk at a population level.
In the last paper, we conduct a more thorough analysis on 72,469 samples from UK Biobank to investigate the rare-variants influence on male-pattern hair loss (MPHL). Novel candidate genes were identified including HEPH, CEPT1 and EIF3F, further proving that rare variants contribute to the genetic landscape of complex phenotypes like male-pattern hair loss.
In conclusion, our findings indicate that rare deleterious variants have an essential role in complex phenotypes, and can be analysed to discover new targets in these traits. Nonetheless, further investigation needs to be conducted to effectively integrate the effects of rare and common variants, ultimately improving comprehensive genetic risk assessment strategies.
Subjects
complex phenotypes, rare variants, common variants, machine learning, bioinformatics
Classification (DDC)
004 Informatik 570 Biowissenschaften, Biologie 610 Medizin, GesundheitAldisi, Rana: Integrative analysis of common and rare variants for a more comprehensive genetic risk assessment. - Bonn, 2024. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5-79715
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5-79715
@phdthesis{handle:20.500.11811/12547,
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5-79715,
author = {{Rana Aldisi}},
title = {Integrative analysis of common and rare variants for a more comprehensive genetic risk assessment},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2024,
month = nov,
note = {The etiology of complex traits is difficult to interpret because of their multifactorial nature. And while environmental factors play an important role in their development, genetic factors also have huge and crucial effect on the expression of complex phenotypes. However, a relevant part of the genetic landscape is yet to be discovered, despite the century long research and studies on the topic. The aim of this thesis is to investigate the role of rare pathogenic variants in complex traits and integrate their analysis with common variants for a more comprehensive genetic risk assessment.
In the first paper, we introduce an open source python package, GenRisk, that implements gene-based burden scores, focusing on rare deleterious variants, and polygenic risk scores (PRS), which are based on common variants. GenRisk’s pipeline also contains an association analysis function using different regression models and a genetic risk modeling function utilizing multiple machine learning models. In this paper, we applied the pipeline on samples from UK Biobank as a usage case example.
The second paper employs the GenRisk framework to explore 28 blood biomarkers within the UK Biobank cohort. We performed the PRS calculation using genotyping data while exome data was used for the gene-based scores (GBS) calculation. Association analysis was done using linear regression and genetic risk prediction models were also generated with either PRS, GBS, or both. We were able to show that rare pathogenic variants play an important role at an individual level, but the traditional PRS could be more informative when predicting the genetic risk at a population level.
In the last paper, we conduct a more thorough analysis on 72,469 samples from UK Biobank to investigate the rare-variants influence on male-pattern hair loss (MPHL). Novel candidate genes were identified including HEPH, CEPT1 and EIF3F, further proving that rare variants contribute to the genetic landscape of complex phenotypes like male-pattern hair loss.
In conclusion, our findings indicate that rare deleterious variants have an essential role in complex phenotypes, and can be analysed to discover new targets in these traits. Nonetheless, further investigation needs to be conducted to effectively integrate the effects of rare and common variants, ultimately improving comprehensive genetic risk assessment strategies.},
url = {https://hdl.handle.net/20.500.11811/12547}
}
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5-79715,
author = {{Rana Aldisi}},
title = {Integrative analysis of common and rare variants for a more comprehensive genetic risk assessment},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2024,
month = nov,
note = {The etiology of complex traits is difficult to interpret because of their multifactorial nature. And while environmental factors play an important role in their development, genetic factors also have huge and crucial effect on the expression of complex phenotypes. However, a relevant part of the genetic landscape is yet to be discovered, despite the century long research and studies on the topic. The aim of this thesis is to investigate the role of rare pathogenic variants in complex traits and integrate their analysis with common variants for a more comprehensive genetic risk assessment.
In the first paper, we introduce an open source python package, GenRisk, that implements gene-based burden scores, focusing on rare deleterious variants, and polygenic risk scores (PRS), which are based on common variants. GenRisk’s pipeline also contains an association analysis function using different regression models and a genetic risk modeling function utilizing multiple machine learning models. In this paper, we applied the pipeline on samples from UK Biobank as a usage case example.
The second paper employs the GenRisk framework to explore 28 blood biomarkers within the UK Biobank cohort. We performed the PRS calculation using genotyping data while exome data was used for the gene-based scores (GBS) calculation. Association analysis was done using linear regression and genetic risk prediction models were also generated with either PRS, GBS, or both. We were able to show that rare pathogenic variants play an important role at an individual level, but the traditional PRS could be more informative when predicting the genetic risk at a population level.
In the last paper, we conduct a more thorough analysis on 72,469 samples from UK Biobank to investigate the rare-variants influence on male-pattern hair loss (MPHL). Novel candidate genes were identified including HEPH, CEPT1 and EIF3F, further proving that rare variants contribute to the genetic landscape of complex phenotypes like male-pattern hair loss.
In conclusion, our findings indicate that rare deleterious variants have an essential role in complex phenotypes, and can be analysed to discover new targets in these traits. Nonetheless, further investigation needs to be conducted to effectively integrate the effects of rare and common variants, ultimately improving comprehensive genetic risk assessment strategies.},
url = {https://hdl.handle.net/20.500.11811/12547}
}
The following license files are associated with this item:
