in high-throughput sequencing make it possible to study how epigenetic
changes mediate the effect of environmental risk factors on
diseases. Epigenetic changes are said to be mediators when they
intervene in the causal pathway between environmental exposures and
diseases. As an example, the figure above represents a mediation model
where epigenetic changes partly mediate the effect of air pollution on
analyses for large-scale omics data raise several challenges. The
Epigenetic & High Dimension Mediation Data Challenge seeks to
introduce and apply key statistical methods for multidimensional
mediation analysis in epidemiology.
We will provide the necessary
background to participate to the Data Challenge, which includes introductory lectures
and demo in R about multiple hypotheses testing,
epigenome-wide association studies (EWAS), and statistical mediation analyses.
To promote collaboration between participants during the Data Challenge, we will show how to share source codes using R notebooks and github.
Participants will have the opportunity to present their own research during poster sessions.
Participants should bring their own laptop in order to be able to participate to the Data Challenge.
Who can attend?
Researchers or students interested in mediation analysis in a
multi-dimensional setting are welcomed. The data challenge is open to
researchers from various backgrounds (statisticians, epidemiologists,
biologists, physicists). The data challenge concerns mediation analysis
in epigenomics but the statistical concepts we introduced are also
relevant for other multidimensional data.
We expect that participants are familiar with the R software and with
standard statistical notions such as linear regression, and P-values.
We are a group of researchers
interested in data science methodologies for analyzing large biomedical
data. Our group of organizers includes Michael Blum (TIMC-IMAG), Olivier Francois (TIMC-IMAG),
Johanna Lepeule (IAB), Magali Richard (TIMC-IMAG), Adeline Leclercq Samson (LJK), and Remy Slama (IAB).
Liu, Y et al. "Epigenome-wide
association data implicate DNA methylation as an intermediary of
genetic risk in rheumatoid arthritis." Nature biotechnology 31.2
Kupers, L et al. "DNA methylation mediates the effect of maternal
smoking during pregnancy on birthweight of the offspring."
International journal of epidemiology 44.4 (2015): 1224-1237.
VanderWeele, TJ "Mediation Analysis: A Practitioner's Guide." Annual review of public health 37 (2016): 17-32.
Zhang, H et al. "Estimating and testing high-dimensional mediation effects in epigenetic studies." Bioinformatics (2016)