Plutynski, Anya (2021) The Cancer Genome Atlas Project: Natural History, Experiment, or Something In-Between? [Preprint]
Text
TCGAData.hypothesisrevised3.24.20.docx Download (78kB) |
Abstract
Some (Hey, et. al., 2009) have claimed that data-driven science is an entirely new paradigm for science – that data driven science will replace hypothesis-driven and experimental sciences. Others have proposed that there is not a dichotomy between data driven and hypothesis-driven research, but rather the approaches are ‘hybridized’ (Smalheiser 2002; Kell and Oliver 2003; Strasser 2012; Keating and Cambrosio 2012; Leonelli 2012).
This raises the question, however, what exactly does it mean to “hybridize” the two. Here I consider the Cancer Genome Atlas (TCGA) as a case in point. I argue that this episode in the history of cancer genomics provides further grounds to contest the sharp contrast between purportedly “hypothesis free” or “data led” and hypothesis driven research. The enterprise of collection and cataloguing genomic data in TCGA involved refinement of methods for identification and classification, a hypothesis-driven and experimental process.
However, there was a bootstrapping element to this research. Many of the questions researchers were investigating could not be clearly defined until preliminary data were in. Yet, organizing and analyzing the data effectively itself required answers to these same questions. Researchers could thus not simply “let the data lead”; analysis of data itself was a process of reframing, and refining, hypotheses about how to separate signal from noise.
In sum, TCGA was neither strictly speaking hypothesis driven, nor simply natural history; it was scaffolding future research. In the early stages of the sequencing of cancer genomes, researchers were still mapping out the sample space.
Export/Citation: | EndNote | BibTeX | Dublin Core | ASCII/Text Citation (Chicago) | HTML Citation | OpenURL |
Social Networking: |
Item Type: | Preprint | ||||||
---|---|---|---|---|---|---|---|
Creators: |
|
||||||
Additional Information: | forthcoming Perspectives on the Human Genome Project and Genomics, Minnesota Studies in the Philosophy of Science Edited by Chris Donohue and Alan Love Minnesota Studies | ||||||
Keywords: | cancer genomics, driver genes, TCGA (cancer genome atlas), data science, hypothesis-driven | ||||||
Subjects: | General Issues > Data Specific Sciences > Biology > Molecular Biology/Genetics General Issues > Evidence |
||||||
Depositing User: | A Plutynski | ||||||
Date Deposited: | 11 May 2021 03:55 | ||||||
Last Modified: | 11 May 2021 03:55 | ||||||
Item ID: | 18999 | ||||||
Subjects: | General Issues > Data Specific Sciences > Biology > Molecular Biology/Genetics General Issues > Evidence |
||||||
Date: | 2021 | ||||||
URI: | https://philsci-archive.pitt.edu/id/eprint/18999 |
Monthly Views for the past 3 years
Monthly Downloads for the past 3 years
Plum Analytics
Actions (login required)
View Item |