Dr. Jesse Gillis – 2009 Recipient Research Update

MIND foundation of BC Progress Report for Jesse Gillis (2009 Competition)


Dr. Jesse Gillis

Dr. Jesse Gillis

This has been a productive year, with four first-authored publications (and one other in review), two publications close to submission, two international conferences attended (one in BC), and a number of collaborations initiated resulting from our ongoing work – these activities are listed in more detail on the following page.

The goal of my project is to characterize the role genes play in schizophrenia, but to do by particularly considering the role context plays in gene function. Specifically, not to consider the role gene A plays in schizophrenia, but to consider the role gene A working with genes B-F play in schizophrenia. We have available in the Pavlidis lab, a large set of expression data, including human schizophrenia brain expression data, and the main method I proposed to examine gene function was through “differential coexpression”. In essence, this is a method of examining gene network rewiring – how the interactions of genes change – with a change in state. In addition to developing a methodology to accomplish this rigorously (i.e., Gillis and Pavlidis, 2009), we have also integrated this with neuroanatomical connectivity data to reveal that changing gene networks are reflected in neuroanatomy – and specifically what changes may be linked to schizophrenia (publication in preparation).

A large portion of this year was also spent in examining what appears to be a crucial bias in the field underlying gene function prediction and network analysis (Gillis and Pavlidis, 2010). In essence, we examine the difference between saying a gene is the best candidate for a function and a gene is specifically a candidate for a function – it turns out that the best candidates for most functions come from the same very small set of genes (so this must assessed if trying to determine a gene’s specific role in schizophrenia). This was a necessary control before beginning our own analysis and is work that has excited collaborative interest and we believe will be of broad significance.

I would like to again express my gratitude to the MIND foundation of BC for their support.


Sept 2009 – Sept 2010

Gillis, J., P. Pavlidis (2010). Multifunctionality drives gene characterization: A re-evaluation of hubs and promiscuity in gene function prediction. PloS ONE (in review).

Gillis, J., M. Mistry, P. Pavlidis (2010). Gene function analysis in complex data sets using ErmineJ. Nature Protocols, 5(6):1148-1159.

Gillis, J., L. Zhang, F Skinner (2010). Spatial coherence and stationarity of local field potentials in an isolated whole hippocampal preparation in vitro. Journal of Computational Neuroscience, Jan 16: 1-12

Gillis, J., P. Pavlidis (2009). A methodology for the analysis of differential coexpression across the human lifespan. BMC Bioinformatics, 10:305.

Gillis, J. (2009). Time–Frequency Methods and Brain Rhythm Signal Processing. In J.L. Velazquez and R. Wennberg (Ed.), Coordinated Activity in the Brain (pp. 225-239). New York: Springer.

Publications in preparation

Mulder, K., R. Schwarz, J. Gillis, G. Donati, P. Pavlidis, F. Markowetz, F. Watt. Functional genetic screening identifies a network of epigenetic strategies controlling adult stem cell fate.

Gillis, J., P. Pavlidis. High-throughput gene function prediction reveals gene networks in the human brain reflect anatomical connectivity and neuropsychiatric disease.


In this interval, we began a collaboration with Dr. Bernard Crespi (Simon Fraser University) on imprinted gene networks and the role they play in schizophrenia and autism. I have also begun working with Meeta Mistry to look at the network properties in her schizophrenia (differential) expression datasets. My supervisor (Dr. Paul Pavlidis) and I are collaborating with Dr. Evica Separovic (UBC) on the role some of our results may play in gene function prediction for genetic studies (particularly candidate genes in genome wide association studies). A PhD student, Eloi Mercer, will be joining the Pavlidis lab on a permanent basis to work on this project. During this interval I also supervised a rotation student (Fong Chun Chan) on related work.

International conferences presented at:

Gillis, J., P. Pavlidis. Influence of aging on candidate neuropsychiatric disease genes measured using differential coexpression. Poster at Society for Neuroscience 2009.

Gillis, J., P Pavlidis. Most functional information in yeast gene networks comes from variation in prevalence and not specific network associations. Poster at Yeast Genetics and Molecular Biology 2010.