Web Application Interface

• Dataset
  Input expression dataset. can be selected from preprocessed demo datasets or the uploaded data.

• or Upload the expression (or CT) file
  a Table in which rows are genes and columns are samples. qPCR data must be in CT values and RNASeq data must be already normalized (e.g. in count per million). Supported file formats are tab delimited .txt, Excel .xlsx and .csv . you can download template examples from "Sample input file" tab.

• Input dataset is in CT values
  The input expression dataset is in raw CT values. Uncheck this for RNASeq datasets.

• Normalization method
  This defines if the data should be normalized by the tool before calculating the aggregation weights. If your data contains less than 4 reference genes, select "None". If your data contains 4 or more low variation reference genes, select mean_all (average of all genes is used for normalization). If your data is from a qPCR array experiment with lots of genes, select high_exp (only CT values less than 35 are used for normalization).

• Number of genes to be aggregated
  The number of genes in each combination

• Subset of genes for combination
  Here you can select just a subset of genes to be used in the combinations. Note if you have a qPCR array dataset and only want to analyze a subset of genes, you should upload the ENTIRE data with all expressed genes and use this option to limit the combinations. The tool needs the entire dataset for the normalization step.

• Weighting method
  The method used to calculate the aggregation weights. For now, the only recommended method is geom_sd_hybrid. Please check out the paper for more detailed explanation.

• Show input dataset
  View the input dataset table in the interface.

• Show weights result
  View the combination weights result table in the interface.

• Show aggregated genes result
  View the expression of the aggregated reference genes after applying the weights.