… in the Results Section), we found that K − 1 components, where K is the number of classes,
was … However, assessing the performance of a non sparse model with K to K + 2 components …

Chemical ecology has strong links with metabolomics, the large-scale study of all metabolites
detectable in a biological sample. Consequently, chemical ecologists are often challenged …

… k-medoids clustering [53], whereby data are allocated into k clusters (similar to … Kim-Anh
Lê Cao is associate professor in Statistical Genomics. Her laboratory focuses on the …

… K - 1 (K - 2) dimensions is sufficient to explain the covariance structure of the microarray (SNP)
data. In the following, we only record the classification error rate obtained after K - 1 (K - 2…

… The first and rather straightforward way is to use k-fold cross validation or leave-one-out
cross validation and compute the prediction error (“RMSEP” see section 1.6.3) in order to …

… As the class prediction relies on individual vote from each omics set, DIABLO allows for
some missing datasets X k during the prediction step, as illustrated in the Breast Cancer case …

Motivation: With the availability of many ‘omics’ data, such as transcriptomics, proteomics or
metabolomics, the integrative or joint analysis of multiple datasets from different technology …

In animals, immunomodulatory dendritic cells (DCs) exposed to autoantigen can suppress
experimental arthritis in an antigen-specific manner. In rheumatoid arthritis (RA), disease-…

Evolutionary change in gene expression is generally considered to be a major driver of
phenotypic differences between species. We investigated innate immune diversification by …

… We focus on two-block data matrices denoted X(n × p) and Y (n × q), where the p variables
x j and q variables y k are of two types and measured on the same samples or individuals n, …