[BIC-announce] Special Seminar - “Imaging Genetics: Towards a new paradigm” - September 25, 2014 (Jean-Baptiste Poline)

[Grace Flynn, Ms.]

SPECIAL SEMINAR ANNOUNCEMENT:



Thursday, September 25, 2014 at 10:00am



de Grandpré Communications Centre, MNI



“Imaging Genetics: Towards a new paradigm”



Jean-Baptiste Poline, PhD

Research scientist,

Helen Wills Neuroscience Institute,

Brain Imaging Center,

University of California, Berkeley




Dr. Poline is a potential faculty hire in the area of Neuroinformatics. His expertise lies in the area of data analysis and neuroinformatics. He is a well-established scientist in this area and would take a lead role in many of our large-scale international neuroinformatics initiatives.



Imaging genetics aims at studying the relation of brain structural or functional phenotypes variability with genetic variability. As such, it will help understand the mechanisms that contribute to variations in both normal and pathological behaviour, holding great promises in fundamental neuroscience, neurology, psychiatry and personalized medicine.

Early studies have shown association of brain phenotypes with specific gene polymorphisms, for instance Hariri et al, 2002, showed that amigdala Blood Oxygen Level Dependent (BOLD) signal activity is linked to the allele of the promoter of the serotonin transporter gene (5-HTT). Pezawas et al., 2004, showed that the amount of hypocampal grey matter depends on the variation of the Brain-derived neurotrophic factor (BDNF) gene.  These genetic variations also predict variations in behavioural phenotypes (memory, anxiety-related behaviours) and are associated with pathologies.

Later, family or twin studies were used to establish the heritability of brain phenotypes such as resting state networks (Glahn et al., 2010), diffusion tensor imaging and reconstructed white matter tracts (Kochunov et al., 2010), or cortical thickness (Rimol et al., 2010). Recently, whole genome (GWAS) whole brain analyses have been attempted using customized statistical analyses, but the very large number of multiple comparisons leads to poor sensitivity of these analyses (Stein et al, 2010, Hibar et al., 2011).  Currently, a number of studies are using meta analyses across a large number of imaging genetics consortia and databases through the ENIGMA project to obtain well powered studies (eg Stein et al., 2012, Thompson et al., 2013).

While imaging genetics has great potential, it is also facing great challenges.  Early results were obtained with small or medium sized cohorts with limited power, making the reproducibility of some results questionable (Ioannidis, 2005, Button et al., 2013). The size and complexity of the data are factors that limit power and interpretation. The flexibility of the analyses is also challenging the reproducibility of results in this field. There is no standard solution to managing these data and the computer infrastructure necessary to conduct large scale analyses. As far as datasharing is concerned, the genetics and the imaging communities may have different cultures but they share many common ethical and informatics questions.

In Dr. Poline’s talk, he will first review these analytical and informatics challenges and put the emphasis on the power and reproducibility issues. He will then describe a set of statistical techniques designed to handle these high dimensionality data, which should yield better sensitivity and interpretability, as well as describe various strategies to reduce the multiple comparison issue. To fulfill its promises, and yield robust and relevant results, imaging genetics has to move from the current framework and be inspired by systems biology techniques, integrating information originating from different sources to address biological or clinical questions. Dr. Poline will present an example of this approach and demonstrate how results can be validated by using public genomics resources. He will describe how the research community could adopt new technologies, and develop a culture of data sharing and collaborations between research communities which could lead to new analysis and meta analysis paradigms. Genomics and imaging are two rapidly advancing fields of fundamental importance for neuroscience. Combining these fields with a solid methodology and a culture of open and reproducible science is bound to lead to critical results in the neuroscience and clinical research.


Alan C. Evans, PhD
McConnell Brain Imaging Centre
MNI, McGill University










-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/bic-announce/attachments/20140916/2b465e33/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Special Seminar Jean-Baptiste Poline Sept 25 2014.pdf
Type: application/pdf
Size: 71433 bytes
Desc: Special Seminar Jean-Baptiste Poline Sept 25 2014.pdf
URL: <http://www.bic.mni.mcgill.ca/pipermail/bic-announce/attachments/20140916/2b465e33/attachment-0001.pdf>