Skip to Main Content
 

Global Search Box

 
 
 

ETD Abstract Container

Abstract Header

Topics in Computational and Statistical Genomics: Exploring Alternatives to the Wald Test and Identifying Deleterious Mutations in Human Diseases.

GNONA, KOMLA MESSAN
http://rave.ohiolink.edu/etdc/view?acc_num=osu1650489241265134

Abstract Details

2022, Doctor of Philosophy, Ohio State University, Biophysics.
Genetic variants are the footprints of the changes that occur in DNA sequences and have been linked to various human diseases. However, with every human birth, hundreds of novel variants also enter the genome. In mendelian diseases, the associated variants have high effects and traditional methods are successful at identifying them as opposed to non-mendelian diseases. The work presented here expands our knowledge on the use of statistical and computational approaches to identify and prioritize the pathogenicity of genetic variants in non-mendelian conditions. In Chapter 1&2 we presented an overview of the modern challenges in associations studies amid the advent of high throughput sequencing. In chapter 3&4, we investigated the widely used Wald test (WT) as it pertains to limited case control studies. We showed that WT exhibits a drawback, namely, the Hauck Donner effect (HDE), that decreases its power to zero as the distance between the point estimate and its expectation increases because the numerator of the Wald statistic is scaled by a biased estimator of its standard deviation. Moreover, we showed that this aberrant behavior of the WT is rather common in genetic burden studies because of the skewness and private levels that arise due in part to the presence of rare variants and proposed a 3-step approach to circumvent the HDE in limited samples. We then used our approach in a real case-control association study of prematurity to successfully identify gene candidates that can explain the underlying molecular basis for neonatal complications. In chapter 5, we demonstrated the practical utility of correctly identifying and assessing the effect of variants using SAVER, a novel comparative genomics approach based on sequence homology. We specifically used SAVER to provide a blueprint for predicting existing and novel polymorphisms in malic enzyme 2, a gene involved in cancers and neurodegenerative disorders.
Igor Jouline (Committee Chair)
Christopher Barlett (Committee Member)
Leif Nelin (Committee Member)
William Stewart (Committee Member)
134 p.
Bioinformatics; Biophysics; Biostatistics; Genetics; Statistics
Wald Test, Genetic Variants, Burden test, evolutionary analysis, p-values, maximum likelihood, epilepsy, bronchopulmonary dysplasia, complex diseases, gwas.

Recommended Citations

Citations

  • GNONA, K. M. (2022). Topics in Computational and Statistical Genomics: Exploring Alternatives to the Wald Test and Identifying Deleterious Mutations in Human Diseases. [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1650489241265134

    APA Style (7th edition)

  • GNONA, KOMLA. Topics in Computational and Statistical Genomics: Exploring Alternatives to the Wald Test and Identifying Deleterious Mutations in Human Diseases. . 2022. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1650489241265134.

    MLA Style (8th edition)

  • GNONA, KOMLA. "Topics in Computational and Statistical Genomics: Exploring Alternatives to the Wald Test and Identifying Deleterious Mutations in Human Diseases. ." Doctoral dissertation, Ohio State University, 2022. http://rave.ohiolink.edu/etdc/view?acc_num=osu1650489241265134

    Chicago Manual of Style (17th edition)

osu1650489241265134
29
© 2022, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.