Mendelian randomization (MR) uses genetic variants as instrumental variables (IVs) to investigate the causal relationship between exposure and outcome. It has become widely popular due to its versatile applications in epidemiological research. Its rising popularity is largely driven by the ease of accessing summary-level data from large consortia, making it a cost-effective choice for researchers. In this thesis, we focus on three issues in MR that result in potential bias in causal inference. We first address the “winner’s curse” in MR, which arises from selecting genetic markers based on their significance or ranking. To mitigate this bias, we adapt the bootstrap-based BR-squared method to function with summary-level data. Our findings reveal that the correction methods can effectively reduce bias, albeit with an increase in variability. We then develop a method that accounts for the correlation caused by sample overlap while addressing potential bias from weak instruments. This proposed method yields stable causal estimates, although the standard errors of causal estimates may not be precisely estimated. Lastly, we introduce a novel approach for identifying invalid instrumental variables showing signs of horizontal pleiotropy. We recommend using the bootstrap method to account for the data-driven process of IV selection. Our results indicate that the bootstrap intervals approach the nominal level of coverage rate when the proportion of invalid IVs is less than 50%. / Dissertation / Doctor of Philosophy (PhD)

N/A / Thesis / Master of Science (MS)

An exciting new era has dawned for the prevention and management of CAD utilizing genetic risk variants. The recent identification of over 60 susceptibility loci for coronary artery disease (CAD) confirm not only the importance of established risk factors, but also the existence of many novel causal pathways that are expected to improve our understanding of the genetic basis of CAD and facilitate the development of new therapeutic agents over time. Concurrently, Mendelian randomization studies have provided intriguing insights on the causal relationship between CAD-related traits, and highlight the potential benefits of long-term modifications of risk factors. Lastly, genetic risk scores of CAD may serve not only as prognostic, but also as predictive markers and carry the potential to considerably improve the delivery of established prevention strategies. This review will summarize the evolution and discovery of genetic risk variants for CAD and their current and future clinical applications.

Radio Frequency IDentification (RFID) is a wireless communications technology which allows companies to secure their assets and increase the portability of information. This research was motivated by the increased commercial use of RFID technology. Existing security protocols with high levels of security have high computation requirements, and less intensive protocols can allow a tag to be tracked. The techniques proposed in this thesis result in the increase of ciphertexts available without a significant increase in processing power or storage requirements. The addition of random inputs to the generation of ciphertexts will increase the number of possible results without requiring a more advanced encryption algorithm or an increased number of stored encryption keys. Four methods of altering the plaintext/ciphertext pair (random block, set pattern, random pattern, and indexed placement) are analyzed to determine the effectiveness of each method. The number of ciphertexts generated, generation time, and generation errors were recorded to determine which of the four proposed methods would be the most beneficial in a RFID system. The comparison of these method characteristics determined that the set pattern placement method provided the best solution. The thesis also discusses how RFID transmissions appear to attackers and explains how the random inputs reduce effectiveness of current system attacks. In addition to improving the anonymity of RFID tag transmissions, the concept of authenticating random inputs is also introduced in this thesis. These methods help prevent an adversary from easily associating a tag with its transmissions, thus increasing the security of the RFID system.

Radio Frequency IDentification (RFID) is a wireless communications technology which allows companies to secure their assets and increase the portability of information. This research was motivated by the increased commercial use of RFID technology. Existing security protocols with high levels of security have high computation requirements, and less intensive protocols can allow a tag to be tracked. The techniques proposed in this thesis result in the increase of ciphertexts available without a significant increase in processing power or storage requirements. The addition of random inputs to the generation of ciphertexts will increase the number of possible results without requiring a more advanced encryption algorithm or an increased number of stored encryption keys. Four methods of altering the plaintext/ciphertext pair (random block, set pattern, random pattern, and indexed placement) are analyzed to determine the effectiveness of each method. The number of ciphertexts generated, generation time, and generation errors were recorded to determine which of the four proposed methods would be the most beneficial in a RFID system. The comparison of these method characteristics determined that the set pattern placement method provided the best solution. The thesis also discusses how RFID transmissions appear to attackers and explains how the random inputs reduce effectiveness of current system attacks. In addition to improving the anonymity of RFID tag transmissions, the concept of authenticating random inputs is also introduced in this thesis. These methods help prevent an adversary from easily associating a tag with its transmissions, thus increasing the security of the RFID system.

In practice, quantum systems are not completely isolated from their environment and the resulting system-environment interaction can lead to information leakage from the system. As a result, if a quantum system is to be used for storing or manipulating information, one would like to characterize these environmental noise effects. Such a characterization affords one the ability to design robust methods for preserving the information contained in the system. Unfortunately, completely characterizing the noise in a realistic amount of time is impossible for even moderately large systems. In this thesis we discuss methods and diagnostics for partially characterizing quantum noise processes that are especially useful in quantum information and computation. We present a randomized benchmarking protocol that provides a scalable method for determining important properties of the noise affecting the set of gates used on a quantum information processor. We also prove various properties of the quantum gate fidelity, which is a useful state-dependent measure of the distance between two quantum operations, and an important diagnostic of the noise affecting a quantum process. Some non-intuitive generic features of quantum operations acting on large-dimensional quantum systems are also presented.

Radio Frequency IDentification (RFID) is a wireless communications technology which allows companies to secure their assets and increase the portability of information. This research was motivated by the increased commercial use of RFID technology. Existing security protocols with high levels of security have high computation requirements, and less intensive protocols can allow a tag to be tracked. The techniques proposed in this thesis result in the increase of ciphertexts available without a significant increase in processing power or storage requirements. The addition of random inputs to the generation of ciphertexts will increase the number of possible results without requiring a more advanced encryption algorithm or an increased number of stored encryption keys. Four methods of altering the plaintext/ciphertext pair (random block, set pattern, random pattern, and indexed placement) are analyzed to determine the effectiveness of each method. The number of ciphertexts generated, generation time, and generation errors were recorded to determine which of the four proposed methods would be the most beneficial in a RFID system. The comparison of these method characteristics determined that the set pattern placement method provided the best solution. The thesis also discusses how RFID transmissions appear to attackers and explains how the random inputs reduce effectiveness of current system attacks. In addition to improving the anonymity of RFID tag transmissions, the concept of authenticating random inputs is also introduced in this thesis. These methods help prevent an adversary from easily associating a tag with its transmissions, thus increasing the security of the RFID system.

In practice, quantum systems are not completely isolated from their environment and the resulting system-environment interaction can lead to information leakage from the system. As a result, if a quantum system is to be used for storing or manipulating information, one would like to characterize these environmental noise effects. Such a characterization affords one the ability to design robust methods for preserving the information contained in the system. Unfortunately, completely characterizing the noise in a realistic amount of time is impossible for even moderately large systems. In this thesis we discuss methods and diagnostics for partially characterizing quantum noise processes that are especially useful in quantum information and computation. We present a randomized benchmarking protocol that provides a scalable method for determining important properties of the noise affecting the set of gates used on a quantum information processor. We also prove various properties of the quantum gate fidelity, which is a useful state-dependent measure of the distance between two quantum operations, and an important diagnostic of the noise affecting a quantum process. Some non-intuitive generic features of quantum operations acting on large-dimensional quantum systems are also presented.

Radio Frequency IDentification (RFID) is a wireless communications technology which allows companies to secure their assets and increase the portability of information. This research was motivated by the increased commercial use of RFID technology. Existing security protocols with high levels of security have high computation requirements, and less intensive protocols can allow a tag to be tracked. The techniques proposed in this thesis result in the increase of ciphertexts available without a significant increase in processing power or storage requirements. The addition of random inputs to the generation of ciphertexts will increase the number of possible results without requiring a more advanced encryption algorithm or an increased number of stored encryption keys. Four methods of altering the plaintext/ciphertext pair (random block, set pattern, random pattern, and indexed placement) are analyzed to determine the effectiveness of each method. The number of ciphertexts generated, generation time, and generation errors were recorded to determine which of the four proposed methods would be the most beneficial in a RFID system. The comparison of these method characteristics determined that the set pattern placement method provided the best solution. The thesis also discusses how RFID transmissions appear to attackers and explains how the random inputs reduce effectiveness of current system attacks. In addition to improving the anonymity of RFID tag transmissions, the concept of authenticating random inputs is also introduced in this thesis. These methods help prevent an adversary from easily associating a tag with its transmissions, thus increasing the security of the RFID system.

The similarities between generations of living subjects are often quantified by heritability. By distinguishing genotypic variation, or variation due to parental pairings, from phenotypic variation, or normal intraspecies variation, the heritability of traits can be estimated. Due to the multivariate nature of many traits, such as size and shape, computation of heritability can be difficult. Also, assessment of the variation of the heritability estimate is extremely difficult. This study uses nonparametric methods, namely the randomization test and the bootstrap, to obtain both a measure of the extremity of the observed heritability and an assessment of the uncertainty.