| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Immediate Transcriptional Effects in Humans1,2
* School of Medicine, University of California, San Francisco, CA 94143;
School of Computing, Queen’s University, Kingston, Ontario, Canada; and
University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, NJ 07103
IFN-
effectively controls clinical exacerbations and magnetic resonance imaging activity in most multiple sclerosis patients. However, its mechanism of action has not been yet fully elucidated. In this study we used DNA microarrays to analyze the longitudinal transcriptional profile of blood cells within a week of IFN- administration. Using differential expression and gene ontology analyses we found evidence of a general decrease in the cellular activity of T lymphocytes resembling the endogenous antiviral response of IFNs. In contrast, most of the differentially expressed genes (DEGs) from untreated individuals were involved in cellular physiological processes. We then used mutual information (MI) to build networks of coregulated genes in both treated and untreated individuals. Interestingly, the connectivity distribution (k) of networks generated with high MI values displayed scale-free properties. Conversely, the observed k for networks generated with suboptimal MI values approximated a Poisson distribution, suggesting that MI captures biologically relevant interactions. Gene networks from individuals treated with IFN- revealed a tight core of immune- and apoptosis-related genes associated with higher values of MI. In contrast, networks obtained from untreated individuals primarily reflected cellular housekeeping functions. Finally, we trained a neural network to reverse engineer the directionality of the main interactions observed at the biological process level. This is the first study that incorporates network analysis to investigate gene regulation in response to a therapeutic drug in humans. Implications of this method in the creation of personalized models of response to therapy are discussed.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was funded by National Institutes of Health Grant 1R01 AI42911 and National Multiple Sclerosis Society Grant NMSS CA 1035-A-7.
2 The data presented in this article have been submitted in the Gene Expression Omnibus under accession number GSE5678.
3 Address correspondence and reprint requests to Dr. Sergio E. Baranzini, 513 Parnassus Avenue S-256, San Francisco, CA 94143. E-mail address: sebaran{at}cgl.ucsf.edu
4 Abbreviations used in this paper: MI, mutual information; DEG, differentially expressed gene; FDR, false discovery rate; GO, gene ontology; SA, sensitivity analysis; O, observed; E, expected; O/E, observed to expected (ratio); SSE, sum squared error; TMI, threshold of MI.
This article has been cited by other articles:
|
|
 |
|
  X. Liu and M. C. K. Yang Identifying temporally differentially expressed genes through functional principal components analysis Biostat., October 1, 2009; 10(4): 667 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Han and L Steinman Systems biology for identification of molecular networks in multiple sclerosis Multiple Sclerosis, May 1, 2009; 15(5): 529 - 530. [PDF]
|
|
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
