Performance of gene expression-based single sample predictors for assessment of clinicopathological subgroups and molecular subtypes in cancers : a case comparison study in non-small cell lung cancer
(2019) In Briefings in Bioinformatics 21(2). p.729-740- Abstract
- The development of multigene classifiers for cancer prognosis, treatment prediction, molecular subtypes or clinicopathological groups has been a cornerstone in transcriptomic analyses of human malignancies for nearly two decades. However, many reported classifiers are critically limited by different preprocessing needs like normalization and data centering. In response, a new breed of classifiers, single sample predictors (SSPs), has emerged. SSPs classify samples in an N-of-1 fashion, relying on, e.g. gene rules comparing expression values within a sample. To date, several methods have been reported, but there is a lack of head-to-head performance comparison for typical cancer classification problems, representing an unmet methodological... (More)
- The development of multigene classifiers for cancer prognosis, treatment prediction, molecular subtypes or clinicopathological groups has been a cornerstone in transcriptomic analyses of human malignancies for nearly two decades. However, many reported classifiers are critically limited by different preprocessing needs like normalization and data centering. In response, a new breed of classifiers, single sample predictors (SSPs), has emerged. SSPs classify samples in an N-of-1 fashion, relying on, e.g. gene rules comparing expression values within a sample. To date, several methods have been reported, but there is a lack of head-to-head performance comparison for typical cancer classification problems, representing an unmet methodological need in cancer bioinformatics. To resolve this need, we performed an evaluation of two SSPs [k-top-scoring pair classifier (kTSP) and absolute intrinsic molecular subtyping (AIMS)] for two case examples of different magnitude of difficulty in non-small cell lung cancer: gene expression–based classification of (i) tumor histology and (ii) molecular subtype. Through the analysis of ~2000 lung cancer samples for each case example (n = 1918 and n = 2106, respectively), we compared the performance of the methods for different sample compositions, training data set sizes, gene expression platforms and gene rule selections. Three main conclusions are drawn from the comparisons: both methods are platform independent, they select largely overlapping gene rules associated with actual underlying tumor biology and, for large training data sets, they behave interchangeably performance-wise. While SSPs like AIMS and kTSP offer new possibilities to move gene expression signatures/predictors closer to a clinical context, they are still importantly limited by the difficultness of the classification problem at hand. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/4a4dc7ab-8e33-4e3b-8550-64bfc4cd51f9
- author
- Cirenajwis, Helena LU ; Lauss, Martin LU ; Planck, Maria LU ; Vallon-Christersson, Johan LU and Staaf, Johan LU
- organization
- publishing date
- 2019-02-04
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Briefings in Bioinformatics
- volume
- 21
- issue
- 2
- pages
- 12 pages
- publisher
- Oxford University Press
- external identifiers
-
- pmid:30721923
- scopus:85083812129
- ISSN
- 1477-4054
- DOI
- 10.1093/bib/bbz008
- language
- English
- LU publication?
- yes
- id
- 4a4dc7ab-8e33-4e3b-8550-64bfc4cd51f9
- date added to LUP
- 2019-05-27 22:35:49
- date last changed
- 2023-04-09 13:32:29
@article{4a4dc7ab-8e33-4e3b-8550-64bfc4cd51f9, abstract = {{The development of multigene classifiers for cancer prognosis, treatment prediction, molecular subtypes or clinicopathological groups has been a cornerstone in transcriptomic analyses of human malignancies for nearly two decades. However, many reported classifiers are critically limited by different preprocessing needs like normalization and data centering. In response, a new breed of classifiers, single sample predictors (SSPs), has emerged. SSPs classify samples in an N-of-1 fashion, relying on, e.g. gene rules comparing expression values within a sample. To date, several methods have been reported, but there is a lack of head-to-head performance comparison for typical cancer classification problems, representing an unmet methodological need in cancer bioinformatics. To resolve this need, we performed an evaluation of two SSPs [k-top-scoring pair classifier (kTSP) and absolute intrinsic molecular subtyping (AIMS)] for two case examples of different magnitude of difficulty in non-small cell lung cancer: gene expression–based classification of (i) tumor histology and (ii) molecular subtype. Through the analysis of ~2000 lung cancer samples for each case example (n = 1918 and n = 2106, respectively), we compared the performance of the methods for different sample compositions, training data set sizes, gene expression platforms and gene rule selections. Three main conclusions are drawn from the comparisons: both methods are platform independent, they select largely overlapping gene rules associated with actual underlying tumor biology and, for large training data sets, they behave interchangeably performance-wise. While SSPs like AIMS and kTSP offer new possibilities to move gene expression signatures/predictors closer to a clinical context, they are still importantly limited by the difficultness of the classification problem at hand.}}, author = {{Cirenajwis, Helena and Lauss, Martin and Planck, Maria and Vallon-Christersson, Johan and Staaf, Johan}}, issn = {{1477-4054}}, language = {{eng}}, month = {{02}}, number = {{2}}, pages = {{729--740}}, publisher = {{Oxford University Press}}, series = {{Briefings in Bioinformatics}}, title = {{Performance of gene expression-based single sample predictors for assessment of clinicopathological subgroups and molecular subtypes in cancers : a case comparison study in non-small cell lung cancer}}, url = {{http://dx.doi.org/10.1093/bib/bbz008}}, doi = {{10.1093/bib/bbz008}}, volume = {{21}}, year = {{2019}}, }