Limits...
Inter-species inference of gene set enrichment in lung epithelial cells from proteomic and large transcriptomic datasets.

Hormoz S, Bhanot G, Biehl M, Bilal E, Meyer P, Norel R, Rhrissorrakrai K, Dayarian A - Bioinformatics (2014)

Bottom Line: Translating findings in rodent models to human models has been a cornerstone of modern biology and drug development.However, in many cases, a naive 'extrapolation' between the two species has not succeeded.In spite of this difference, we were able to develop a robust algorithm to predict gene set activation in NHBE with high accuracy using simple analytical methods.

View Article: PubMed Central - PubMed

Affiliation: Kavli Institute for Theoretical Physics, Kohn Hall, University of California, Santa Barbara, CA 93106, USA, Department of Physics, Department of Molecular Biology and Biochemistry, Busch Campus, Rutgers University, Piscataway, NJ 08854, USA, Johann Bernoulli Institute for Mathematics and Computer Science, University of Groningen, 9700 AK Groningen, The Netherlands and IBM T.J. Watson Research Center, Computational Biology, Yorktown Heights, NY 10003, USA.

Show MeSH

Related in: MedlinePlus

We predict gene set enrichment in human bronchial epithelial cells under a diverse set of stimuli (B) from measurements of gene set enrichment in rat cells under the same stimuli. We considered two distinct approaches: (i) a direct method where the algorithm was trained only on the gene set measurements (set A), and a direct prediction was made on the enrichment scores of set B. (ii) An indirect method where the gene expression levels were used for training and prediction. A GSEA was then used to infer the gene set enrichment scores. Blue boxes are the available data; red is to be predicted
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4325538&req=5

btu569-F1: We predict gene set enrichment in human bronchial epithelial cells under a diverse set of stimuli (B) from measurements of gene set enrichment in rat cells under the same stimuli. We considered two distinct approaches: (i) a direct method where the algorithm was trained only on the gene set measurements (set A), and a direct prediction was made on the enrichment scores of set B. (ii) An indirect method where the gene expression levels were used for training and prediction. A GSEA was then used to infer the gene set enrichment scores. Blue boxes are the available data; red is to be predicted

Mentions: The outline of the article is as follows. First, we describe how the raw data were processed to identify statistically significant signals. Next, we discuss inter-species correlations (human and rat) for both gene sets and gene expression level data. In principle, there are two possible strategies available to predict differential expression (enrichment) of human gene sets from the rat data (see Fig. 1): (i) one possibility is to use a direct method, where the human gene set enrichment scores are learned from those in rat training data (Rat A/Human A) and then applied to the rat test data gene set scores (Rat B) to predict human gene set scores; (ii) the second possibility is to use an indirect method, where human gene expression levels are learned from gene expression levels in rat in the training data and applied to the Rat B gene expression test data to get human expression predictions. Finally, the predicted Human B gene expression levels could be used in a gene set enrichment algorithm (GSEA) (Subramanian et al., 2005) to predict human gene set scores. Although we attempted both methods, for our final prediction, we used only the direct method [method (i)]. Below, we outline our reasons for choosing the direct method followed by a detailed description of the prediction algorithm we used for our predictions.Fig. 1.


Inter-species inference of gene set enrichment in lung epithelial cells from proteomic and large transcriptomic datasets.

Hormoz S, Bhanot G, Biehl M, Bilal E, Meyer P, Norel R, Rhrissorrakrai K, Dayarian A - Bioinformatics (2014)

We predict gene set enrichment in human bronchial epithelial cells under a diverse set of stimuli (B) from measurements of gene set enrichment in rat cells under the same stimuli. We considered two distinct approaches: (i) a direct method where the algorithm was trained only on the gene set measurements (set A), and a direct prediction was made on the enrichment scores of set B. (ii) An indirect method where the gene expression levels were used for training and prediction. A GSEA was then used to infer the gene set enrichment scores. Blue boxes are the available data; red is to be predicted
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4325538&req=5

btu569-F1: We predict gene set enrichment in human bronchial epithelial cells under a diverse set of stimuli (B) from measurements of gene set enrichment in rat cells under the same stimuli. We considered two distinct approaches: (i) a direct method where the algorithm was trained only on the gene set measurements (set A), and a direct prediction was made on the enrichment scores of set B. (ii) An indirect method where the gene expression levels were used for training and prediction. A GSEA was then used to infer the gene set enrichment scores. Blue boxes are the available data; red is to be predicted
Mentions: The outline of the article is as follows. First, we describe how the raw data were processed to identify statistically significant signals. Next, we discuss inter-species correlations (human and rat) for both gene sets and gene expression level data. In principle, there are two possible strategies available to predict differential expression (enrichment) of human gene sets from the rat data (see Fig. 1): (i) one possibility is to use a direct method, where the human gene set enrichment scores are learned from those in rat training data (Rat A/Human A) and then applied to the rat test data gene set scores (Rat B) to predict human gene set scores; (ii) the second possibility is to use an indirect method, where human gene expression levels are learned from gene expression levels in rat in the training data and applied to the Rat B gene expression test data to get human expression predictions. Finally, the predicted Human B gene expression levels could be used in a gene set enrichment algorithm (GSEA) (Subramanian et al., 2005) to predict human gene set scores. Although we attempted both methods, for our final prediction, we used only the direct method [method (i)]. Below, we outline our reasons for choosing the direct method followed by a detailed description of the prediction algorithm we used for our predictions.Fig. 1.

Bottom Line: Translating findings in rodent models to human models has been a cornerstone of modern biology and drug development.However, in many cases, a naive 'extrapolation' between the two species has not succeeded.In spite of this difference, we were able to develop a robust algorithm to predict gene set activation in NHBE with high accuracy using simple analytical methods.

View Article: PubMed Central - PubMed

Affiliation: Kavli Institute for Theoretical Physics, Kohn Hall, University of California, Santa Barbara, CA 93106, USA, Department of Physics, Department of Molecular Biology and Biochemistry, Busch Campus, Rutgers University, Piscataway, NJ 08854, USA, Johann Bernoulli Institute for Mathematics and Computer Science, University of Groningen, 9700 AK Groningen, The Netherlands and IBM T.J. Watson Research Center, Computational Biology, Yorktown Heights, NY 10003, USA.

Show MeSH
Related in: MedlinePlus