TY - JOUR
T1 - Gene filtering strategies for machine learning guided biomarker discovery using neonatal sepsis RNA-seq data
AU - Parkinson, Edward
AU - Liberatore, Federico
AU - Watkins, W. John
AU - Andrews, Robert
AU - Edkins, Sarah
AU - Hibbert, Julie
AU - Strunk, Tobias
AU - Currie, Andrew
AU - Ghazal, Peter
N1 - Funding Information:
Project Sepsis is funded by a Ser Cymru grant from Welsh Government and EU/ERDF funds to PG. PROTECT was funded by the NHMRC and Perth Children’s Hospital Foundation. JH supported by the Wesfarmers Centre of Vaccines and Infectious Diseases.
Publisher Copyright:
Copyright © 2023 Parkinson, Liberatore, Watkins, Andrews, Edkins, Hibbert, Strunk, Currie and Ghazal.
PY - 2023/4/11
Y1 - 2023/4/11
N2 - Machine learning (ML) algorithms are powerful tools that are increasingly being used for sepsis biomarker discovery in RNA-Seq data. RNA-Seq datasets contain multiple sources and types of noise (operator, technical and non-systematic) that may bias ML classification. Normalisation and independent gene filtering approaches described in RNA-Seq workflows account for some of this variability and are typically only targeted at differential expression analysis rather than ML applications. Pre-processing normalisation steps significantly reduce the number of variables in the data and thereby increase the power of statistical testing, but can potentially discard valuable and insightful classification features. A systematic assessment of applying transcript level filtering on the robustness and stability of ML based RNA-seq classification remains to be fully explored. In this report we examine the impact of filtering out low count transcripts and those with influential outliers read counts on downstream ML analysis for sepsis biomarker discovery using elastic net regularised logistic regression, L1-reguarlised support vector machines and random forests. We demonstrate that applying a systematic objective strategy for removal of uninformative and potentially biasing biomarkers representing up to 60% of transcripts in different sample size datasets, including two illustrative neonatal sepsis cohorts, leads to substantial improvements in classification performance, higher stability of the resulting gene signatures, and better agreement with previously reported sepsis biomarkers. We also demonstrate that the performance uplift from gene filtering depends on the ML classifier chosen, with L1-regularlised support vector machines showing the greatest performance improvements with our experimental data.
AB - Machine learning (ML) algorithms are powerful tools that are increasingly being used for sepsis biomarker discovery in RNA-Seq data. RNA-Seq datasets contain multiple sources and types of noise (operator, technical and non-systematic) that may bias ML classification. Normalisation and independent gene filtering approaches described in RNA-Seq workflows account for some of this variability and are typically only targeted at differential expression analysis rather than ML applications. Pre-processing normalisation steps significantly reduce the number of variables in the data and thereby increase the power of statistical testing, but can potentially discard valuable and insightful classification features. A systematic assessment of applying transcript level filtering on the robustness and stability of ML based RNA-seq classification remains to be fully explored. In this report we examine the impact of filtering out low count transcripts and those with influential outliers read counts on downstream ML analysis for sepsis biomarker discovery using elastic net regularised logistic regression, L1-reguarlised support vector machines and random forests. We demonstrate that applying a systematic objective strategy for removal of uninformative and potentially biasing biomarkers representing up to 60% of transcripts in different sample size datasets, including two illustrative neonatal sepsis cohorts, leads to substantial improvements in classification performance, higher stability of the resulting gene signatures, and better agreement with previously reported sepsis biomarkers. We also demonstrate that the performance uplift from gene filtering depends on the ML classifier chosen, with L1-regularlised support vector machines showing the greatest performance improvements with our experimental data.
KW - gene signature stability
KW - independent gene filtering
KW - machine learning with RNA-seq
KW - neonatal sepsis
KW - transcriptomic sepsis biomarkers
UR - http://www.scopus.com/inward/record.url?scp=85153514567&partnerID=8YFLogxK
U2 - 10.3389/fgene.2023.1158352
DO - 10.3389/fgene.2023.1158352
M3 - Article
C2 - 37113992
AN - SCOPUS:85153514567
SN - 1664-8021
VL - 14
JO - Frontiers in Genetics
JF - Frontiers in Genetics
M1 - 1158352
ER -