GDASC is a GPU-parallel based server used for identifying batches and classifying samples into different batches in a high dimensional gene expression dataset. The batch information can be further used as a covariate in conjunction with other variables of interest among standard bioinformatics analysis like differential expression analysis. If you want to learn how to use our server, please click on the “HELP” button to learn the background and then read the instruction below.

Good in Performance
Our server is outstanding in its high speed. It is able to dispose of a dataset with more than one thousand samples and tens of thousands of genes in several minutes.


  • Upload your data and select the parameters (refer to the help page for detailed instructions)
  • Select data (upload the data matrix with the data.frame form, which is a txt document)

    Select vector (upload the vector containing the group labels, which is also a txt document)

    Email (where the result will be sent to)

    Countable (the parameter that means whether the data is countable)

    Lambda (the regularization parameter that contains the data shrinkage ratio)

    Rank (number of batches you set to detect in an experiment)

    Penalty function (choose one of the three functions you want to use for data shrinkage)

    1 (L1 norm) 2 (L2 norm) 3 (L22 norm)


  • Download the results with the linkages here
    • After disposing of your data, linkages of the results will be shown here as well as sent
      to your email, which include the clustermap, batch labels and an estimated batch-free
      matrix. Click on the "download" button and you will get them.

    • Download the clustermap

    • Download the labels

    • Download the batch-free-matrix
    • Compute the similarity heatmaps


  • H Yi, A Raman, H Zhang, G Allen, Z Liu, Detecting hidden batch factors through data-adaptive adjustment for biological effects, Bioinformatics, 34: 1141–1147 (2018).