Professor Benjamin Haibe-Kains from the University of
Toronto, Canada, will give a CIL doctoral course on "Biomarker discovery
from large pharmacogenomics datasets".
The course will take place on Monday March 05, 2018
from 10.00 to 12.30 and 14.00 to 16.30 in Shannon Seminar
Room
Place du Levant 3, B-1348 Louvain-la-Neuve.
REGISTRATION
Free but *requested*. Please consult:
http://sites.uclouvain.be/doctoralschool-cil/index.php?
page=agenda
TITLE:
Biomarker discovery from large pharmacogenomics
datasets
ABSTRACT:
Lecture:
This course will focus on the challenges encountered when
applying machine learning techniques in complex,
high dimensional biological data.
In particular, we will focus on biomarker discovery from
pharmacogenomic data,
which consists of developing predictors of response of
cancer cell lines to chemical compounds based
on their genomic features.
From a methodological viewpoint, biomarker discovery is
strongly linked to variable selection,
through methods such as Supervised Learning with sparsity
inducing norms (LASSO, Elastic Net, ...)
or techniques accounting for the complex correlation
structure of biological features (groupLASSO or MRMR).
Yet, the main focus of this talk will be on the sound use of
such methods in a pharmacogenomics context,
their validation and correct interpretation of the produced
results.
I will first discuss how to assess the quality of both the
input and output data.
I will further present our recent findings illustrating the
importance of unified analytical platforms,
data and code sharing in bioinformatics and biomedical
research, as the data generation process becomes
increasingly complex and requires high level of replication
to achieve robust results.
This is particularly relevant as our portfolio of machine
learning techniques is ever enlarging,
with its set of hyperparameters that can be tuning in a
multitude of ways,
increasing the risk of overfitting when developing
multivariate predictors of drug response.
Practical session:
Bring your own laptop and install the R/Bioconductor
PharmacoGx package
(https://bioconductor.org/packages/PharmacoGx/) and its
dependencies.
Run the following commands to download a local version of
two pharmacogenomic datasets
> GDSC <- downloadPSet("GDSC")
> CCLE <- downloadPSet("CCLE")
VENUE:
Maxwell Building
Shannon Seminar Room (First Floor)
Place du Levant, 3
B - 1348 Louvain-la-Neuve
Belgium
Maps:
https://goo.gl/maps/SB5YfdM4zaB2
External visitors are welcome to park on the Rédimé
Parking (middle - right of the map)
https://cdn.uclouvain.be/public/Exports%20reddot/adpi/images/PlanparkingsLLN090607-sl.jpg