DESnowball: Bagging with Distance-based Regression for Differential Gene Expression Analyses


The DESnowball package implements a statistical data mining method that compares the whole genome gene expression profiles with respect to the presence of a recurrent genetic disturbance event ( e.g. a recurrent driver mutation) to identify the affected target genes.

The input data for the snowball analysis are the whole genome gene expression profiles and the mutation status of a recurrent genetic event on a group of samples. The analysis has been tested on the TCGA melanoma primary tumor samples. The minimum sample size required per group is three.


From R:

install_github("DESnowball", user="snowball-project")



    # snowball analysis on the demo dataset included in the package 
    # A test run
    Bn <- 10000
    ncore <-4
    # call Snowball
    sb <- snowball(y=sb.mutation,
    # process the gene ranking and selection
    sb.sel <- select.features(sb)
    # plot the Jn values
    plotJn(sb, sb.sel)
    # get the significant gene list
    top.genes <- toplist(sb.sel)


Xu, Y. and Sun, J. (2005) PfCluster: a new cluster analysis procedure for gene expression profiles. Presented at a conference on Nonparametric Inference and Probability With Applications to Science honoring Michael Woodroofe; September 24-25, 2005; Ann Arbor, Mich, 2005.

McArdlei, B.H. and Anderson, M.J. (2001) Fitting multivariate models to community data: A comment on distance-based redundancy analysis. Ecology 82(1): 290-297.

Xu, Y., Guo, X., Sun, J. and Zhao. Z. Snowball: resampling combined with distance-based regression to discover transcriptional consequences of driver mutation, manuscript.

Guo, X., Xu, Y. and Zhao, Z.. Driver mutation BRAF regulates cell proliferation and apoptosis via MITF in the pathogenesis of melanoma, manuscript.