Data Preprocessing

Input

The input files can be upload by clicking the "Open" button. You can select one or multiple files at one time. If you select multiple files, you should guarantee the files are from the technical replicates. The output will be only one file which combine the quantitative information of proteins.

After selecting the input files, it will ask you to choose the columns which contain the protein abundances. Make sure that the columns you select only include numeric values. All the other values will be regarded as zero.

If you are processing TPP datasets including temperture information, you should input the corresponding tempertures in the "Temperture Setting" table. It is essential for the subsequent procedures. If you are processing iTSA/NPA datasets, you can ignore this step.

Parameters

  • PSM column name: You can select the column which indicate the number of PSM. This will be used for PSM filter. If you do not need the filter, keep it as None.
  • PSM filter threshold: The thershold for PSM filter. Proteins with PSM number bellow this number will be removed.
  • Merging metrics: If more than one files are input, how to combine the abundances of the same protein in different files. If onle one file is input, ignore this.
  • Normalization: Method for normalization. Reference means all the quantitative values are divided by the reference sample. This is usually used for the TPP analysis; Median means normalize all the samples to the same median value. None means no normalization is applied.
  • Missing value imputation: Method for impute missing values. KNN means the iterative KNN method for the imputation; Zero means all the missing values to zero are replaced to zero; None means no imputation are applied, which the protein group with missing values will be discarded in the following steps. If TMT-labeled quantitative method is used in your experiment, None is recommended. If label-free quantitative method is used, KNN is recommended.
  • Missing value filter ratio: The cut-off threshold for missing value. If the ratio of samples with missing value are more than this value, the protein group will be discarded.
  • RSD filter threshold: If you have technical replicates, protein group with RSD over this threshold will be discarded.
  • Reference column: If you are using Reference as normalization method, this is which column you are used as the reference.

Display

  • File list: The input files for preprocessing. If you want to remove the files, please click the clear button.
  • Temperture: The table used for input the tempertures corresponding to the samples.
  • RSD distribution: Density plot of the RSDs of the technical replicates.
  • Data Viewer: The output of the preprocessed data. Click Save to write to csv file.