Home > Research > Publications & Outputs > A mechanism-aware and multiomic machine-learnin...

Links

Text available via DOI:

View graph of relations

A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth. / Culley, Christopher; Vijayakumar, Supreeta; Zampieri, Guido; Angione, Claudio.

In: Proceedings of the National Academy of Sciences of the United States of America, Vol. 117, No. 31, 04.08.2020, p. 18869-18879.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Culley, C, Vijayakumar, S, Zampieri, G & Angione, C 2020, 'A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth', Proceedings of the National Academy of Sciences of the United States of America, vol. 117, no. 31, pp. 18869-18879. https://doi.org/10.1073/pnas.2002959117

APA

Culley, C., Vijayakumar, S., Zampieri, G., & Angione, C. (2020). A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth. Proceedings of the National Academy of Sciences of the United States of America, 117(31), 18869-18879. https://doi.org/10.1073/pnas.2002959117

Vancouver

Culley C, Vijayakumar S, Zampieri G, Angione C. A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth. Proceedings of the National Academy of Sciences of the United States of America. 2020 Aug 4;117(31):18869-18879. https://doi.org/10.1073/pnas.2002959117

Author

Culley, Christopher ; Vijayakumar, Supreeta ; Zampieri, Guido ; Angione, Claudio. / A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth. In: Proceedings of the National Academy of Sciences of the United States of America. 2020 ; Vol. 117, No. 31. pp. 18869-18879.

Bibtex

@article{27c919d3a94642bb900592e313a139c7,
title = "A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth",
abstract = "Metabolic modeling and machine learning are key components in the emerging next generation of systems and synthetic biology tools, targeting the genotype–phenotype–environment relationship. Rather than being used in isolation, it is becoming clear that their value is maximized when they are combined. However, the potential of integrating these two frameworks for omic data augmentation and integration is largely unexplored. We propose, rigorously assess, and compare machine-learning–based data integration techniques, combining gene expression profiles with computationally generated metabolic flux data to predict yeast cell growth. To this end, we create strain-specific metabolic models for 1,143 Saccharomyces cerevisiae mutants and we test 27 machine-learning methods, incorporating state-of-the-art feature selection and multiview learning approaches. We propose a multiview neural network using fluxomic and transcriptomic data, showing that the former increases the predictive accuracy of the latter and reveals functional patterns that are not directly deducible from gene expression alone. We test the proposed neural network on a further 86 strains generated in a different experiment, therefore verifying its robustness to an additional independent dataset. Finally, we show that introducing mechanistic flux features improves the predictions also for knockout strains whose genes were not modeled in the metabolic reconstruction. Our results thus demonstrate that fusing experimental cues with in silico models, based on known biochemistry, can contribute with disjoint information toward biologically informed and interpretable machine learning. Overall, this study provides tools for understanding and manipulating complex phenotypes, increasing both the prediction accuracy and the extent of discernible mechanistic biological insights.",
author = "Christopher Culley and Supreeta Vijayakumar and Guido Zampieri and Claudio Angione",
year = "2020",
month = aug,
day = "4",
doi = "10.1073/pnas.2002959117",
language = "English",
volume = "117",
pages = "18869--18879",
journal = "Proceedings of the National Academy of Sciences of the United States of America",
issn = "0027-8424",
publisher = "National Academy of Sciences",
number = "31",

}

RIS

TY - JOUR

T1 - A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth

AU - Culley, Christopher

AU - Vijayakumar, Supreeta

AU - Zampieri, Guido

AU - Angione, Claudio

PY - 2020/8/4

Y1 - 2020/8/4

N2 - Metabolic modeling and machine learning are key components in the emerging next generation of systems and synthetic biology tools, targeting the genotype–phenotype–environment relationship. Rather than being used in isolation, it is becoming clear that their value is maximized when they are combined. However, the potential of integrating these two frameworks for omic data augmentation and integration is largely unexplored. We propose, rigorously assess, and compare machine-learning–based data integration techniques, combining gene expression profiles with computationally generated metabolic flux data to predict yeast cell growth. To this end, we create strain-specific metabolic models for 1,143 Saccharomyces cerevisiae mutants and we test 27 machine-learning methods, incorporating state-of-the-art feature selection and multiview learning approaches. We propose a multiview neural network using fluxomic and transcriptomic data, showing that the former increases the predictive accuracy of the latter and reveals functional patterns that are not directly deducible from gene expression alone. We test the proposed neural network on a further 86 strains generated in a different experiment, therefore verifying its robustness to an additional independent dataset. Finally, we show that introducing mechanistic flux features improves the predictions also for knockout strains whose genes were not modeled in the metabolic reconstruction. Our results thus demonstrate that fusing experimental cues with in silico models, based on known biochemistry, can contribute with disjoint information toward biologically informed and interpretable machine learning. Overall, this study provides tools for understanding and manipulating complex phenotypes, increasing both the prediction accuracy and the extent of discernible mechanistic biological insights.

AB - Metabolic modeling and machine learning are key components in the emerging next generation of systems and synthetic biology tools, targeting the genotype–phenotype–environment relationship. Rather than being used in isolation, it is becoming clear that their value is maximized when they are combined. However, the potential of integrating these two frameworks for omic data augmentation and integration is largely unexplored. We propose, rigorously assess, and compare machine-learning–based data integration techniques, combining gene expression profiles with computationally generated metabolic flux data to predict yeast cell growth. To this end, we create strain-specific metabolic models for 1,143 Saccharomyces cerevisiae mutants and we test 27 machine-learning methods, incorporating state-of-the-art feature selection and multiview learning approaches. We propose a multiview neural network using fluxomic and transcriptomic data, showing that the former increases the predictive accuracy of the latter and reveals functional patterns that are not directly deducible from gene expression alone. We test the proposed neural network on a further 86 strains generated in a different experiment, therefore verifying its robustness to an additional independent dataset. Finally, we show that introducing mechanistic flux features improves the predictions also for knockout strains whose genes were not modeled in the metabolic reconstruction. Our results thus demonstrate that fusing experimental cues with in silico models, based on known biochemistry, can contribute with disjoint information toward biologically informed and interpretable machine learning. Overall, this study provides tools for understanding and manipulating complex phenotypes, increasing both the prediction accuracy and the extent of discernible mechanistic biological insights.

U2 - 10.1073/pnas.2002959117

DO - 10.1073/pnas.2002959117

M3 - Journal article

VL - 117

SP - 18869

EP - 18879

JO - Proceedings of the National Academy of Sciences of the United States of America

JF - Proceedings of the National Academy of Sciences of the United States of America

SN - 0027-8424

IS - 31

ER -