상세정보
Export to Refworks부가기능
Two Novel Algorithms for Extracting Patient Specific Information from Tumor Biopsies
상세 프로파일
| 자료유형 | 학위논문 |
|---|---|
| 서명/저자사항 | Two Novel Algorithms for Extracting Patient Specific Information from Tumor Biopsies. |
| 개인저자 | Klein, Michael. |
| 단체저자명 | Yale University. Computational Biology and Bioinformatics. |
| 발행사항 | [S.l.]: Yale University., 2019. |
| 발행사항 | Ann Arbor: ProQuest Dissertations & Theses, 2019. |
| 형태사항 | 251 p. |
| 기본자료 저록 | Dissertations Abstracts International 81-03B. Dissertation Abstract International |
| ISBN | 9781088315965 |
| 학위논문주기 | Thesis (Ph.D.)--Yale University, 2019. |
| 일반주기 |
Source: Dissertations Abstracts International, Volume: 81-03, Section: B.
Advisor: Zhao, Hongyu |
| 이용제한사항 | This item must not be sold to any third party vendors. |
| 요약 | GRAPE Abstract: Personalizing treatment regimes based on gene expression profiles of individual tumors will facilitate management of cancer. Although many methods have been developed to identify pathways perturbed in tumors, the results are often not generalizable across independent datasets due to the presence of platform/batch effects. There is a need to develop methods that are robust to platform/batch effects and able to identify perturbed pathways in individual samples.We present Gene-Ranking Analysis of Pathway Expression (GRAPE) as a novel method to identify gene sets with abnormal expression in individual samples. GRAPE first defines a template consisting of an ordered set of pathway genes to characterize the normative state of a pathway based on the relative rankings of gene expression levels across a set of reference samples. This template can be used to assess whether a sample conforms to or deviates from the typical behavior of the reference samples for this pathway. We demonstrate that GRAPE classifiers achieve superior robustness and generalizability across different datasets compared to existing methods. A powerful feature of GRAPE is the ability to represent individual gene expression profiles as a vector of pathways scores. We present applications to the analyses of breast cancer subtypes and different colonic diseases.GRAPE templates offer a novel approach for summarizing the behavior of gene-sets across a collection of gene expression profiles. GRAPE pathway scores enable identification of abnormal gene-set behavior in individual samples using a non-competitive approach that is fundamentally distinct from popular enrichment-based methods. GRAPE may be an appropriate tool for researchers seeking to identify individual samples displaying abnormal gene-set behavior as well as to explore differences in the consensus gene-set behavior of groups of samples.CRSO Abstract: Understanding the genetic causes of cancer in individual patients is critical to developing effective personalized treatments. There is compelling evidence that it takes multiple genetic alterations to transform a normal cell into a cancer cell. Computational approaches have been developed to detect putative driver events by comparing the observed to expected event frequencies in the population. These algorithms fill a crucial need of prioritizing candidate alterations and weeding out the majority of false positives. However, we do not understand how different drivers cooperate and we lack statistical methods that address this important need. Going one step further, we lack methods for determining which combinations of alterations drive cancer formation in individual patients.We present Cancer Rule Set Optimization (CRSO) as a method for predicting the combinations of alterations that cooperate to drive tumor formation in individual patients. CRSO is developed upon a theoretical mathematical framework in which a cancer rule is defined to be a collection of two or more genetic alterations that collectively drive cancer when they co-occur in an individual patient. All observed alterations are initially assumed to be passengers and are associated with patient-specific passenger probabilities representing the probability of the alteration occurring in a particular patient by chance under neutral selection. A rule set is a proposed collection of rules that encompass all of the different ways for cancer to happen in the patient population. Under a proposed rule set each patient is systematically assigned to at most one rule, and the alterations that comprise the assigned rule are designated as drivers for that patient. CRSO is an optimization problem over the set of all possible rule sets. A statistical penalty is determined for each rule set based on the designation of events as passengers or drivers. A stochastic optimization procedure is developed to determine a rule set of fixed size K that minimizes the statistical penalty.We applied CRSO to 19 TCGA cancer types. For each cancer type a core rule set was identified that balances minimizing the statistical penalty and minimizing rule set size. Many of the most important rules involve cooperation between mutations and copy number variations. We find many rules that are identified as drivers in multiple cancer types, some of which contain lesser-known driver alterations. In several cancers statistically significant survival differences are observed between groups of patients assigned to different rules that would not be identifiable by consideration of individual alterations. We highlight examples in glioma, liver cancer and melanoma that have potential significance as clinical biomarkers. Our results suggest that classifying patients according to driver rules identified by CRSO has advantages compared to classifying patients according to individual driver alterations. |
| 일반주제명 | Bioinformatics. |
| 언어 | 영어 |
| 바로가기 | 
: 이 자료의 원문은 한국교육학술정보원에서 제공합니다. |
