Logic models to predict continuous outputs based on binary inputs with an application to personalized cancer therapy

Theo A. Knijnenburg, Gunnar W. Klau, Francesco Iorio, Mathew J. Garnett, Ultan McDermott, Ilya Shmulevich, Lodewyk F A Wessels

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Mining large datasets using machine learning approaches often leads to models that are hard to interpret and not amenable to the generation of hypotheses that can be experimentally tested. We present Logic Optimization for Binary Input to Continuous Output' (LOBICO), a computational approach that infers small and easily interpretable logic models of binary input features that explain a continuous output variable. Applying LOBICO to a large cancer cell line panel, we find that logic combinations of multiple mutations are more predictive of drug response than single gene predictors. Importantly, we show that the use of the continuous information leads to robust and more accurate logic models. LOBICO implements the ability to uncover logic models around predefined operating points in terms of sensitivity and specificity. As such, it represents an important step towards practical application of interpretable logic models.

Original languageEnglish
Article number36812
JournalScientific Reports
Volume6
DOIs
Publication statusPublished - 23 Nov 2016

Fingerprint

Learning systems
Genes
Cells

Cite this

Knijnenburg, T. A., Klau, G. W., Iorio, F., Garnett, M. J., McDermott, U., Shmulevich, I., & Wessels, L. F. A. (2016). Logic models to predict continuous outputs based on binary inputs with an application to personalized cancer therapy. Scientific Reports, 6, [36812]. https://doi.org/10.1038/srep36812
Knijnenburg, Theo A. ; Klau, Gunnar W. ; Iorio, Francesco ; Garnett, Mathew J. ; McDermott, Ultan ; Shmulevich, Ilya ; Wessels, Lodewyk F A. / Logic models to predict continuous outputs based on binary inputs with an application to personalized cancer therapy. In: Scientific Reports. 2016 ; Vol. 6.
@article{60f62feb229744b4b1106fe8ad7ebc39,
title = "Logic models to predict continuous outputs based on binary inputs with an application to personalized cancer therapy",
abstract = "Mining large datasets using machine learning approaches often leads to models that are hard to interpret and not amenable to the generation of hypotheses that can be experimentally tested. We present Logic Optimization for Binary Input to Continuous Output' (LOBICO), a computational approach that infers small and easily interpretable logic models of binary input features that explain a continuous output variable. Applying LOBICO to a large cancer cell line panel, we find that logic combinations of multiple mutations are more predictive of drug response than single gene predictors. Importantly, we show that the use of the continuous information leads to robust and more accurate logic models. LOBICO implements the ability to uncover logic models around predefined operating points in terms of sensitivity and specificity. As such, it represents an important step towards practical application of interpretable logic models.",
author = "Knijnenburg, {Theo A.} and Klau, {Gunnar W.} and Francesco Iorio and Garnett, {Mathew J.} and Ultan McDermott and Ilya Shmulevich and Wessels, {Lodewyk F A}",
year = "2016",
month = "11",
day = "23",
doi = "10.1038/srep36812",
language = "English",
volume = "6",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",

}

Knijnenburg, TA, Klau, GW, Iorio, F, Garnett, MJ, McDermott, U, Shmulevich, I & Wessels, LFA 2016, 'Logic models to predict continuous outputs based on binary inputs with an application to personalized cancer therapy' Scientific Reports, vol. 6, 36812. https://doi.org/10.1038/srep36812

Logic models to predict continuous outputs based on binary inputs with an application to personalized cancer therapy. / Knijnenburg, Theo A.; Klau, Gunnar W.; Iorio, Francesco; Garnett, Mathew J.; McDermott, Ultan; Shmulevich, Ilya; Wessels, Lodewyk F A.

In: Scientific Reports, Vol. 6, 36812, 23.11.2016.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Logic models to predict continuous outputs based on binary inputs with an application to personalized cancer therapy

AU - Knijnenburg, Theo A.

AU - Klau, Gunnar W.

AU - Iorio, Francesco

AU - Garnett, Mathew J.

AU - McDermott, Ultan

AU - Shmulevich, Ilya

AU - Wessels, Lodewyk F A

PY - 2016/11/23

Y1 - 2016/11/23

N2 - Mining large datasets using machine learning approaches often leads to models that are hard to interpret and not amenable to the generation of hypotheses that can be experimentally tested. We present Logic Optimization for Binary Input to Continuous Output' (LOBICO), a computational approach that infers small and easily interpretable logic models of binary input features that explain a continuous output variable. Applying LOBICO to a large cancer cell line panel, we find that logic combinations of multiple mutations are more predictive of drug response than single gene predictors. Importantly, we show that the use of the continuous information leads to robust and more accurate logic models. LOBICO implements the ability to uncover logic models around predefined operating points in terms of sensitivity and specificity. As such, it represents an important step towards practical application of interpretable logic models.

AB - Mining large datasets using machine learning approaches often leads to models that are hard to interpret and not amenable to the generation of hypotheses that can be experimentally tested. We present Logic Optimization for Binary Input to Continuous Output' (LOBICO), a computational approach that infers small and easily interpretable logic models of binary input features that explain a continuous output variable. Applying LOBICO to a large cancer cell line panel, we find that logic combinations of multiple mutations are more predictive of drug response than single gene predictors. Importantly, we show that the use of the continuous information leads to robust and more accurate logic models. LOBICO implements the ability to uncover logic models around predefined operating points in terms of sensitivity and specificity. As such, it represents an important step towards practical application of interpretable logic models.

UR - http://www.scopus.com/inward/record.url?scp=84996866116&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84996866116&partnerID=8YFLogxK

U2 - 10.1038/srep36812

DO - 10.1038/srep36812

M3 - Article

VL - 6

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

M1 - 36812

ER -