Inter-species pathway perturbation prediction via data-driven detection of functional homology

Christoph Hafemeister, Roberto Romero, Erhan Bilal, Pablo Meyer, Raquel Norel, Kahn Rhrissorrakrai, Richard Bonneau, Adi L. Tarca

Research output: Contribution to journalArticle

Abstract

Motivation: Experiments in animal models are often conducted to infer how humans will respond to stimuli by assuming that the same biological pathways will be affected in both organisms. The limitations of this assumption were tested in the IMPROVER Species Translation Challenge, where 52 stimuli were applied to both human and rat cells and perturbed pathways were identified. In the Inter-species Pathway Perturbation Prediction sub-challenge, multiple teams proposedmethods to use rat transcription data from 26 stimuli to predict human gene set and pathway activity under the same perturbations. Submissions were evaluated using three performance metrics on data from the remaining 26 stimuli. Results: We present two approaches, ranked second in this challenge, that do not rely on sequence-based orthology between rat and human genes to translate pathway perturbation state but instead identify transcriptional response orthologs across a set of training conditions. The translation from rat to human accomplished by these socalled direct methods is not dependent on the particular analysis method used to identify perturbed gene sets. In contrast, machine learning-based methods require performing a pathway analysis initially and then mapping the pathway activity between organisms. Unlike most machine learning approaches, direct methods can be used to predict the activation of a human pathway for a new (test) stimuli, even when that pathway was never activated by a training stimuli.

Original languageEnglish (US)
Pages (from-to)501-508
Number of pages8
JournalBioinformatics
Volume31
Issue number4
DOIs
StatePublished - Feb 15 2015

Fingerprint

Data-driven
Rats
Homology
Pathway
Perturbation
Genes
Prediction
Learning systems
Transcription
Gene
Direct Method
Animals
Chemical activation
Machine Learning
Predict
Animal Model
Animal Models
Performance Metrics
Human
Activation

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Computational Mathematics
  • Statistics and Probability

Cite this

Hafemeister, C., Romero, R., Bilal, E., Meyer, P., Norel, R., Rhrissorrakrai, K., ... Tarca, A. L. (2015). Inter-species pathway perturbation prediction via data-driven detection of functional homology. Bioinformatics, 31(4), 501-508. https://doi.org/10.1093/bioinformatics/btu570

Inter-species pathway perturbation prediction via data-driven detection of functional homology. / Hafemeister, Christoph; Romero, Roberto; Bilal, Erhan; Meyer, Pablo; Norel, Raquel; Rhrissorrakrai, Kahn; Bonneau, Richard; Tarca, Adi L.

In: Bioinformatics, Vol. 31, No. 4, 15.02.2015, p. 501-508.

Research output: Contribution to journalArticle

Hafemeister, C, Romero, R, Bilal, E, Meyer, P, Norel, R, Rhrissorrakrai, K, Bonneau, R & Tarca, AL 2015, 'Inter-species pathway perturbation prediction via data-driven detection of functional homology', Bioinformatics, vol. 31, no. 4, pp. 501-508. https://doi.org/10.1093/bioinformatics/btu570
Hafemeister C, Romero R, Bilal E, Meyer P, Norel R, Rhrissorrakrai K et al. Inter-species pathway perturbation prediction via data-driven detection of functional homology. Bioinformatics. 2015 Feb 15;31(4):501-508. https://doi.org/10.1093/bioinformatics/btu570
Hafemeister, Christoph ; Romero, Roberto ; Bilal, Erhan ; Meyer, Pablo ; Norel, Raquel ; Rhrissorrakrai, Kahn ; Bonneau, Richard ; Tarca, Adi L. / Inter-species pathway perturbation prediction via data-driven detection of functional homology. In: Bioinformatics. 2015 ; Vol. 31, No. 4. pp. 501-508.
@article{33a885c416cb44f0a4143c6983028ec0,
title = "Inter-species pathway perturbation prediction via data-driven detection of functional homology",
abstract = "Motivation: Experiments in animal models are often conducted to infer how humans will respond to stimuli by assuming that the same biological pathways will be affected in both organisms. The limitations of this assumption were tested in the IMPROVER Species Translation Challenge, where 52 stimuli were applied to both human and rat cells and perturbed pathways were identified. In the Inter-species Pathway Perturbation Prediction sub-challenge, multiple teams proposedmethods to use rat transcription data from 26 stimuli to predict human gene set and pathway activity under the same perturbations. Submissions were evaluated using three performance metrics on data from the remaining 26 stimuli. Results: We present two approaches, ranked second in this challenge, that do not rely on sequence-based orthology between rat and human genes to translate pathway perturbation state but instead identify transcriptional response orthologs across a set of training conditions. The translation from rat to human accomplished by these socalled direct methods is not dependent on the particular analysis method used to identify perturbed gene sets. In contrast, machine learning-based methods require performing a pathway analysis initially and then mapping the pathway activity between organisms. Unlike most machine learning approaches, direct methods can be used to predict the activation of a human pathway for a new (test) stimuli, even when that pathway was never activated by a training stimuli.",
author = "Christoph Hafemeister and Roberto Romero and Erhan Bilal and Pablo Meyer and Raquel Norel and Kahn Rhrissorrakrai and Richard Bonneau and Tarca, {Adi L.}",
year = "2015",
month = "2",
day = "15",
doi = "10.1093/bioinformatics/btu570",
language = "English (US)",
volume = "31",
pages = "501--508",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - Inter-species pathway perturbation prediction via data-driven detection of functional homology

AU - Hafemeister, Christoph

AU - Romero, Roberto

AU - Bilal, Erhan

AU - Meyer, Pablo

AU - Norel, Raquel

AU - Rhrissorrakrai, Kahn

AU - Bonneau, Richard

AU - Tarca, Adi L.

PY - 2015/2/15

Y1 - 2015/2/15

N2 - Motivation: Experiments in animal models are often conducted to infer how humans will respond to stimuli by assuming that the same biological pathways will be affected in both organisms. The limitations of this assumption were tested in the IMPROVER Species Translation Challenge, where 52 stimuli were applied to both human and rat cells and perturbed pathways were identified. In the Inter-species Pathway Perturbation Prediction sub-challenge, multiple teams proposedmethods to use rat transcription data from 26 stimuli to predict human gene set and pathway activity under the same perturbations. Submissions were evaluated using three performance metrics on data from the remaining 26 stimuli. Results: We present two approaches, ranked second in this challenge, that do not rely on sequence-based orthology between rat and human genes to translate pathway perturbation state but instead identify transcriptional response orthologs across a set of training conditions. The translation from rat to human accomplished by these socalled direct methods is not dependent on the particular analysis method used to identify perturbed gene sets. In contrast, machine learning-based methods require performing a pathway analysis initially and then mapping the pathway activity between organisms. Unlike most machine learning approaches, direct methods can be used to predict the activation of a human pathway for a new (test) stimuli, even when that pathway was never activated by a training stimuli.

AB - Motivation: Experiments in animal models are often conducted to infer how humans will respond to stimuli by assuming that the same biological pathways will be affected in both organisms. The limitations of this assumption were tested in the IMPROVER Species Translation Challenge, where 52 stimuli were applied to both human and rat cells and perturbed pathways were identified. In the Inter-species Pathway Perturbation Prediction sub-challenge, multiple teams proposedmethods to use rat transcription data from 26 stimuli to predict human gene set and pathway activity under the same perturbations. Submissions were evaluated using three performance metrics on data from the remaining 26 stimuli. Results: We present two approaches, ranked second in this challenge, that do not rely on sequence-based orthology between rat and human genes to translate pathway perturbation state but instead identify transcriptional response orthologs across a set of training conditions. The translation from rat to human accomplished by these socalled direct methods is not dependent on the particular analysis method used to identify perturbed gene sets. In contrast, machine learning-based methods require performing a pathway analysis initially and then mapping the pathway activity between organisms. Unlike most machine learning approaches, direct methods can be used to predict the activation of a human pathway for a new (test) stimuli, even when that pathway was never activated by a training stimuli.

UR - http://www.scopus.com/inward/record.url?scp=84928983700&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84928983700&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/btu570

DO - 10.1093/bioinformatics/btu570

M3 - Article

VL - 31

SP - 501

EP - 508

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 4

ER -