000043709 001__ 43709 000043709 005__ 20240920130349.0 000043709 0247_ $$2doi$$a10.6083/bpxhc43709 000043709 037__ $$aETD 000043709 041__ $$aeng 000043709 245__ $$aMechanistic deep learning for perturbation biology: application to precision oncology 000043709 260__ $$bOregon Health and Science University 000043709 269__ $$a2024-09-19 000043709 336__ $$aDissertation 000043709 502__ $$bPh.D. 000043709 502__ $$gBioinformatics & Computational Biomedicine 000043709 520__ $$aThe work presented in this dissertation focuses on the development of robust deep learning models to predict in-vitro cancer drug responses, with utility in precision oncology research tasks such as drug repurposing and prioritization of disease-specific drug combinations. This work addresses two critical shortcomings in the current approaches to predicting cancer drug responses with deep learning: 1) data quality issues inherent in high-throughput drug screening datasets by developing algorithms for detection of atypical or low-quality data and 2) improved prediction and utility of perturbation biology models by developing algorithms that operate on mechanistic prior knowledge. 000043709 536__ $$oNational Cancer Institute$$cT32 CA106195; T32CA106195 000043709 536__ $$oNational Institute of Allergy and Infectious Diseases / National Library of Medicine $$cT15LM007088 000043709 536__ $$oOregon Clinical & Translational Research Institute$$cTL1TR002371 000043709 540__ $$fCC BY 000043709 542__ $$fIn copyright - single owner 000043709 650__ $$aDeep Learning$$012734 000043709 650__ $$aBiomarkers, Pharmacological$$037754 000043709 650__ $$aPrecision Medicine$$038927 000043709 6531_ $$agraph learning 000043709 6531_ $$aprecision oncology 000043709 6531_ $$adrug repurposing 000043709 6531_ $$aperturbation biology 000043709 6531_ $$amechanistic modeling 000043709 691__ $$aSchool of Medicine$$041369 000043709 692__ $$aDepartment of Medical Informatics and Clinical Epidemiology$$041422 000043709 7001_ $$aEvans, Nathaniel J.$$uOregon Health and Science University$$041354$$10000-0003-2245-8904 000043709 7201_ $$aMcWeeney, Shannon$$uOregon Health and Science University$$041354$$10000-0001-8333-6607$$eAcademic advisor$$7Personal 000043709 7201_ $$aSong, Xubo $$uOregon Health and Science University$$041354$$eChair$$7Personal 000043709 7201_ $$aWu, Guanming $$uOregon Health and Science University$$041354$$10000-0001-8196-1177$$eAdvisor$$7Personal 000043709 7201_ $$aMills, Gordon B. $$uOregon Health and Science University$$041354$$10000-0002-0144-9614$$eAdvisor$$7Personal 000043709 7201_ $$aMooney, Michael$$uOregon Health and Science University$$041354$$10000-0003-1372-8722$$eAdvisor$$7Personal 000043709 789__ $$eHas part$$whttps://doi.org/10.1101/2024.02.28.582164$$2DOI 000043709 789__ $$eHas part$$w https://doi.org/10.48550/arXiv.2405.08217$$2DOI 000043709 792__ $$a<p>Sections of my dissertation have alternated versions previously published as preprints: </p> <p>1. Evans, Nathaniel J., et al. "Data Valuation with Gradient Similarity." ArXiv (2024).</p> <p>2. Evans, Nathaniel J., et al. "Graph Structured Neural Networks for Perturbation Biology." bioRxiv (2024).</p> 000043709 8564_ $$yMechanistic deep learning for perturbation biology: Application to precision oncology$$974247bec-861f-42b6-be3f-6f55b1dd917d$$s11902381$$uhttps://digitalcollections.ohsu.edu/record/43709/files/Evans.Nathaniel.2024.pdf 000043709 909CO $$ooai:digitalcollections.ohsu.edu:43709$$pstudent-work 000043709 980__ $$aTheses and Dissertations 000043709 981__ $$aPublished$$b2024-09-19