Establishing a many-cytokine signature via multivariate anomaly detection

K. Dingle*, A. Zimek, F. Azizieh, A. R. Ansari

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Establishing a cytokine signature associated to some medical condition is an important task in immunology. Increasingly, large numbers of cytokines are used for signatures, via lists of reference ranges for each individual cytokine or ratios of cytokines. Here we argue that this common approach has weaknesses, especially when many different cytokines are analysed. Instead, we propose that establishing signatures can be framed as a multivariate anomaly detection problem, and hence exploit the many statistical methods available for this. In this framework, whether or not a given subject’s profile matches the cytokine signature of some condition is determined by whether or not the profile is typical of reference samples of that condition, as judged by an anomaly detection algorithm. We examine previously published cytokine data sets associated to pregnancy complications, brain tumours, and rheumatoid arthritis, as well as normal healthy control samples, and test the performance of a range of anomaly detection algorithms on these data, identifying the best performing methods. Finally, we suggest that this anomaly detection approach could be adopted more widely for general multi-biomarker signatures.

Original languageEnglish
Article number9684
JournalScientific Reports
Volume9
Issue number1
Number of pages13
ISSN2045-2322
DOIs
Publication statusPublished - 1. Dec 2019

Fingerprint

Pregnancy Complications
Allergy and Immunology
Reference Values
Datasets

Cite this

Dingle, K. ; Zimek, A. ; Azizieh, F. ; Ansari, A. R. / Establishing a many-cytokine signature via multivariate anomaly detection. In: Scientific Reports. 2019 ; Vol. 9, No. 1.
@article{d8d8a26912b94638aeb91e9f17053878,
title = "Establishing a many-cytokine signature via multivariate anomaly detection",
abstract = "Establishing a cytokine signature associated to some medical condition is an important task in immunology. Increasingly, large numbers of cytokines are used for signatures, via lists of reference ranges for each individual cytokine or ratios of cytokines. Here we argue that this common approach has weaknesses, especially when many different cytokines are analysed. Instead, we propose that establishing signatures can be framed as a multivariate anomaly detection problem, and hence exploit the many statistical methods available for this. In this framework, whether or not a given subject’s profile matches the cytokine signature of some condition is determined by whether or not the profile is typical of reference samples of that condition, as judged by an anomaly detection algorithm. We examine previously published cytokine data sets associated to pregnancy complications, brain tumours, and rheumatoid arthritis, as well as normal healthy control samples, and test the performance of a range of anomaly detection algorithms on these data, identifying the best performing methods. Finally, we suggest that this anomaly detection approach could be adopted more widely for general multi-biomarker signatures.",
author = "K. Dingle and A. Zimek and F. Azizieh and Ansari, {A. R.}",
year = "2019",
month = "12",
day = "1",
doi = "10.1038/s41598-019-46097-9",
language = "English",
volume = "9",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

Establishing a many-cytokine signature via multivariate anomaly detection. / Dingle, K.; Zimek, A.; Azizieh, F.; Ansari, A. R.

In: Scientific Reports, Vol. 9, No. 1, 9684, 01.12.2019.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Establishing a many-cytokine signature via multivariate anomaly detection

AU - Dingle, K.

AU - Zimek, A.

AU - Azizieh, F.

AU - Ansari, A. R.

PY - 2019/12/1

Y1 - 2019/12/1

N2 - Establishing a cytokine signature associated to some medical condition is an important task in immunology. Increasingly, large numbers of cytokines are used for signatures, via lists of reference ranges for each individual cytokine or ratios of cytokines. Here we argue that this common approach has weaknesses, especially when many different cytokines are analysed. Instead, we propose that establishing signatures can be framed as a multivariate anomaly detection problem, and hence exploit the many statistical methods available for this. In this framework, whether or not a given subject’s profile matches the cytokine signature of some condition is determined by whether or not the profile is typical of reference samples of that condition, as judged by an anomaly detection algorithm. We examine previously published cytokine data sets associated to pregnancy complications, brain tumours, and rheumatoid arthritis, as well as normal healthy control samples, and test the performance of a range of anomaly detection algorithms on these data, identifying the best performing methods. Finally, we suggest that this anomaly detection approach could be adopted more widely for general multi-biomarker signatures.

AB - Establishing a cytokine signature associated to some medical condition is an important task in immunology. Increasingly, large numbers of cytokines are used for signatures, via lists of reference ranges for each individual cytokine or ratios of cytokines. Here we argue that this common approach has weaknesses, especially when many different cytokines are analysed. Instead, we propose that establishing signatures can be framed as a multivariate anomaly detection problem, and hence exploit the many statistical methods available for this. In this framework, whether or not a given subject’s profile matches the cytokine signature of some condition is determined by whether or not the profile is typical of reference samples of that condition, as judged by an anomaly detection algorithm. We examine previously published cytokine data sets associated to pregnancy complications, brain tumours, and rheumatoid arthritis, as well as normal healthy control samples, and test the performance of a range of anomaly detection algorithms on these data, identifying the best performing methods. Finally, we suggest that this anomaly detection approach could be adopted more widely for general multi-biomarker signatures.

U2 - 10.1038/s41598-019-46097-9

DO - 10.1038/s41598-019-46097-9

M3 - Journal article

VL - 9

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 9684

ER -