000009633 001__ 9633 000009633 005__ 20240411123232.0 000009633 0247_ $$2DOI$$a10.6083/g158bj19x 000009633 037__ $$aIR 000009633 245__ $$aMachine learning for cancer subtyping: sampling effects on predictive accuracy and feature selection 000009633 260__ $$bOregon Health and Science University 000009633 269__ $$a2022-04-21 000009633 336__ $$aAbstract 000009633 520__ $$aAccurate classification of cancer subtypes based on genomic data is a key component of precision oncology. Subtyping allows for better alignment of therapeutics with the specific biology of an individual patient's tumor. Training predictive models for classification requires collection, sequencing, and labeling of patient samples which incurs significant cost and thus motivates an estimation of the minimum number of samples required to attain a particular classification accuracy. In this work, a minimum sample size estimation strategy for machine learning cancer subtype prediction is developed by combining a subsampling method for learning curve generation with an inverse power law curve fitting method. 000009633 540__ $$fCC BY 000009633 542__ $$fIn copyright - joint owners 000009633 650__ $$aMachine Learning$$011449 000009633 650__ $$aNeoplasms$$022758 000009633 6531_ $$acancer subtypes 000009633 6531_ $$apower law curve fitting 000009633 6531_ $$afeature engineering 000009633 6531_ $$athe cancer genome atlas 000009633 6531_ $$aprecision oncology 000009633 691__ $$aSchool of Medicine$$041369 000009633 692__ $$aDepartment of Biomedical Engineering$$041397 000009633 692__ $$aOHSU Knight Cancer Institute$$041488 000009633 7001_ $$aKarlberg, Brian$$uOregon Health and Science University$$041354 000009633 7001_ $$aLee, Jordan$$uOregon Health and Science University$$041354 000009633 7001_ $$aWong, Chris$$uOregon Health and Science University$$041354 000009633 7001_ $$aStuart, Josh$$uOregon Health and Science University$$041354 000009633 7001_ $$aEllrott, Kyle$$uOregon Health and Science University$$041354 000009633 711__ $$aResearch Week$$uOregon Health and Science University$$d2022 000009633 8564_ $$9b27f3670-2ba1-4a59-bb46-4ed388550d6b$$s55177$$uhttps://digitalcollections.ohsu.edu/record/9633/files/Karlberg_sample_count_abstract_2022-04-20.pdf 000009633 905__ $$a/rest/prod/g1/58/bj/19/g158bj19x 000009633 980__ $$aResearch Week