Cancer as a tissue anomaly: classifying tumor transcriptomes based only on healthy data Quinn, Thomas P, Nguyen, Thin, Lee, Samuel C and Venkatesh, Svetha 2019, Cancer as a tissue anomaly: classifying tumor transcriptomes based only on healthy data, Frontiers in genetics, vol. 10, pp. 1-6, doi: 10.3389/fgene.2019.00599. Attached Files Name Description MIMEType Size Downloads Title Cancer as a tissue anomaly: classifying tumor transcriptomes based only on healthy data Author(s) Quinn, Thomas P orcid.org/0000-0003-0286-6329 Nguyen, Thin orcid.org/0000-0003-3467-8963 Lee, Samuel C orcid.org/0000-0002-6805-4051 Venkatesh, Svetha orcid.org/0000-0001-8675-6631 Journal name Frontiers in genetics Volume number 10 Article ID 599 Start page 1 End page 6 Total pages 6 Publisher Frontiers Media Place of publication Lausanne, Switzerland Publication date 2019-07 ISSN 1664-8021 Keyword(s) Science & Technology Life Sciences & Biomedicine Genetics & Heredity machine learning TCGA anomaly detection classification surveillance Summary Since the turn of the century, researchers have sought to diagnose cancer based on gene expression signatures measured from the blood or biopsy as biomarkers. This task, known as classification, is typically solved using a suite of algorithms that learn a mathematical rule capable of discriminating one group (“cases”) from another (“controls”). However, discriminatory methods can only identify cancerous samples that resemble those that the algorithm already saw during training. As such, discriminatory methods may be ill-suited for the classification of cancer: because the possibility space of cancer is definitively large, the existence of a one-of-a-kind gene expression signature is likely. Instead, we propose using an established surveillance method that detects anomalous samples based on their deviation from a learned normal steady-state structure. By transferring this method to transcriptomic data, we can create an anomaly detector for tissue transcriptomes, a “tissue detector,” that is capable of identifying cancer without ever seeing a single cancer example. As a proof-of-concept, we train a “tissue detector” on normal GTEx samples that can classify TCGA samples with >90% AUC for 3 out of 6 tissues. Importantly, we find that the classification accuracy is improved simply by adding more healthy samples. We conclude this report by emphasizing the conceptual advantages of anomaly detection and by highlighting future directions for this field of study. Language eng DOI 10.3389/fgene.2019.00599 Indigenous content off HERDC Research category C1 Refereed article in a scholarly journal Copyright notice ©2019, Quinn, Nguyen, Lee and Venkatesh Free to Read? Yes Persistent URL http://hdl.handle.net/10536/DRO/DU:30126450 Document type: Journal Article Collections: Faculty of Science, Engineering and Built Environment Centre for Pattern Recognition and Data Analytics (PRADA) Open Access Collection Related Links Link Description Link to full-text (open access)   Connect to published version Go to link with your DU access privileges     Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved. Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au. Versions Version Filter Type Wed, 17 Jul 2019, 08:27:34 EST Wed, 17 Jul 2019, 08:29:28 EST Thu, 18 Jul 2019, 05:00:37 EST Thu, 18 Jul 2019, 08:03:01 EST Mon, 22 Jul 2019, 10:29:30 EST Mon, 22 Jul 2019, 12:49:41 EST Tue, 23 Jul 2019, 12:06:51 EST Tue, 23 Jul 2019, 12:08:42 EST Tue, 23 Jul 2019, 12:09:19 EST Sat, 10 Aug 2019, 02:32:54 EST Fri, 15 Nov 2019, 16:40:29 EST Filtered Full Citation counts:   Cited 1 times in TR Web of Science Cited 1 times in Scopus Search Google Scholar Access Statistics: 86 Abstract Views, 2 File Downloads  -  Detailed Statistics Created: Mon, 22 Jul 2019, 10:29:30 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.