A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum

George C. Mayne; Richard J. Woodman; David I. Watson; Tim Bright; Susan Gan; Reginald V. Lord; Michael J. Bourke; Angelique Levert-Mignon; Isabell Bastian; Tanya Irvine; Ann Schloithe; Marian Martin; Lorraine Sheehan-Hennessy; Damian J. Hussey

doi:10.3390/ijms24087068

A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum

George C. Mayne, Richard J. Woodman, David I. Watson, Tim Bright, Susan Gan, Reginald V. Lord, Michael J. Bourke, Angelique Levert-Mignon, Isabell Bastian, Tanya Irvine, Ann Schloithe, Marian Martin, Lorraine Sheehan-Hennessy, Damian J. Hussey

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

42 Downloads (Pure)

Abstract

The biomarker development field within molecular medicine remains limited by the methods that are available for building predictive models. We developed an efficient method for conservatively estimating confidence intervals for the cross validation-derived prediction errors of biomarker models. This new method was investigated for its ability to improve the capacity of our previously developed method, StaVarSel, for selecting stable biomarkers. Compared with the standard cross validation method, StaVarSel markedly improved the estimated generalisable predictive capacity of serum miRNA biomarkers for the detection of disease states that are at increased risk of progressing to oesophageal adenocarcinoma. The incorporation of our new method for conservatively estimating confidence intervals into StaVarSel resulted in the selection of less complex models with increased stability and improved or similar predictive capacities. The methods developed in this study have the potential to improve progress from biomarker discovery to biomarker driven translational research.

Original language	English
Article number	7068
Number of pages	16
Journal	International Journal of Molecular Sciences
Volume	24
Issue number	8
DOIs	https://doi.org/10.3390/ijms24087068
Publication status	Published - Apr 2023

Keywords

Barrett’s oesophagus
bias–variance trade-off
biomarkers
cross validation
machine learning
microRNA
miRNA
oesophageal adenocarcinoma
robustness
stability

Access to Document

10.3390/ijms24087068Licence: CC BY

Final published versionFinal published version, 1.52 MBLicence: CC BY

Fingerprint

Dive into the research topics of 'A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum'. Together they form a unique fingerprint.

Cite this

Mayne, G. C., Woodman, R. J., Watson, D. I., Bright, T., Gan, S., Lord, R. V., Bourke, M. J., Levert-Mignon, A., Bastian, I., Irvine, T., Schloithe, A., Martin, M., Sheehan-Hennessy, L., & Hussey, D. J. (2023). A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum. International Journal of Molecular Sciences, 24(8), Article 7068. https://doi.org/10.3390/ijms24087068

Mayne, George C. ; Woodman, Richard J. ; Watson, David I. et al. / A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum. In: International Journal of Molecular Sciences. 2023 ; Vol. 24, No. 8.

@article{e53e8c332a53492e8627b8394812b0fd,

title = "A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett{\textquoteright}s Oesophagus Disease Spectrum",

abstract = "The biomarker development field within molecular medicine remains limited by the methods that are available for building predictive models. We developed an efficient method for conservatively estimating confidence intervals for the cross validation-derived prediction errors of biomarker models. This new method was investigated for its ability to improve the capacity of our previously developed method, StaVarSel, for selecting stable biomarkers. Compared with the standard cross validation method, StaVarSel markedly improved the estimated generalisable predictive capacity of serum miRNA biomarkers for the detection of disease states that are at increased risk of progressing to oesophageal adenocarcinoma. The incorporation of our new method for conservatively estimating confidence intervals into StaVarSel resulted in the selection of less complex models with increased stability and improved or similar predictive capacities. The methods developed in this study have the potential to improve progress from biomarker discovery to biomarker driven translational research.",

keywords = "Barrett{\textquoteright}s oesophagus, bias–variance trade-off, biomarkers, cross validation, machine learning, microRNA, miRNA, oesophageal adenocarcinoma, robustness, stability",

author = "Mayne, {George C.} and Woodman, {Richard J.} and Watson, {David I.} and Tim Bright and Susan Gan and Lord, {Reginald V.} and Bourke, {Michael J.} and Angelique Levert-Mignon and Isabell Bastian and Tanya Irvine and Ann Schloithe and Marian Martin and Lorraine Sheehan-Hennessy and Hussey, {Damian J.}",

year = "2023",

month = apr,

doi = "10.3390/ijms24087068",

language = "English",

volume = "24",

journal = "International Journal of Molecular Sciences",

issn = "1661-6596",

publisher = "MDPI",

number = "8",

}

Mayne, GC, Woodman, RJ , Watson, DI , Bright, T, Gan, S, Lord, RV, Bourke, MJ, Levert-Mignon, A, Bastian, I, Irvine, T, Schloithe, A, Martin, M, Sheehan-Hennessy, L & Hussey, DJ 2023, 'A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum', International Journal of Molecular Sciences, vol. 24, no. 8, 7068. https://doi.org/10.3390/ijms24087068

A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum. / Mayne, George C.; Woodman, Richard J.; Watson, David I. et al.
In: International Journal of Molecular Sciences, Vol. 24, No. 8, 7068, 04.2023.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum

AU - Mayne, George C.

AU - Woodman, Richard J.

AU - Watson, David I.

AU - Bright, Tim

AU - Gan, Susan

AU - Lord, Reginald V.

AU - Bourke, Michael J.

AU - Levert-Mignon, Angelique

AU - Bastian, Isabell

AU - Irvine, Tanya

AU - Schloithe, Ann

AU - Martin, Marian

AU - Sheehan-Hennessy, Lorraine

AU - Hussey, Damian J.

PY - 2023/4

Y1 - 2023/4

N2 - The biomarker development field within molecular medicine remains limited by the methods that are available for building predictive models. We developed an efficient method for conservatively estimating confidence intervals for the cross validation-derived prediction errors of biomarker models. This new method was investigated for its ability to improve the capacity of our previously developed method, StaVarSel, for selecting stable biomarkers. Compared with the standard cross validation method, StaVarSel markedly improved the estimated generalisable predictive capacity of serum miRNA biomarkers for the detection of disease states that are at increased risk of progressing to oesophageal adenocarcinoma. The incorporation of our new method for conservatively estimating confidence intervals into StaVarSel resulted in the selection of less complex models with increased stability and improved or similar predictive capacities. The methods developed in this study have the potential to improve progress from biomarker discovery to biomarker driven translational research.

AB - The biomarker development field within molecular medicine remains limited by the methods that are available for building predictive models. We developed an efficient method for conservatively estimating confidence intervals for the cross validation-derived prediction errors of biomarker models. This new method was investigated for its ability to improve the capacity of our previously developed method, StaVarSel, for selecting stable biomarkers. Compared with the standard cross validation method, StaVarSel markedly improved the estimated generalisable predictive capacity of serum miRNA biomarkers for the detection of disease states that are at increased risk of progressing to oesophageal adenocarcinoma. The incorporation of our new method for conservatively estimating confidence intervals into StaVarSel resulted in the selection of less complex models with increased stability and improved or similar predictive capacities. The methods developed in this study have the potential to improve progress from biomarker discovery to biomarker driven translational research.

KW - Barrett’s oesophagus

KW - bias–variance trade-off

KW - biomarkers

KW - cross validation

KW - machine learning

KW - microRNA

KW - miRNA

KW - oesophageal adenocarcinoma

KW - robustness

KW - stability

UR - http://www.scopus.com/inward/record.url?scp=85158069289&partnerID=8YFLogxK

UR - http://purl.org/au-research/grants/NHMRC/1104281

U2 - 10.3390/ijms24087068

DO - 10.3390/ijms24087068

M3 - Article

C2 - 37108236

AN - SCOPUS:85158069289

SN - 1661-6596

VL - 24

JO - International Journal of Molecular Sciences

JF - International Journal of Molecular Sciences

IS - 8

M1 - 7068

ER -

Mayne GC, Woodman RJ , Watson DI , Bright T, Gan S, Lord RV et al. A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum. International Journal of Molecular Sciences. 2023 Apr;24(8):7068. doi: 10.3390/ijms24087068

A Method for Increasing the Robustness of Stable Feature Selection for Biomarker Discovery in Molecular Medicine Developed Using Serum Small Extracellular Vesicle Associated miRNAs and the Barrett’s Oesophagus Disease Spectrum

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this