General Features

The main interface of ProSAP includes four buttons corresponding to the function modules. Generally the export file of Proteome Discoverer or MaxQuant should be preprocessed first. For different experimental purpose, you may need PISA/iTSA analysis, TPP analysis, 2D-TPP analysis and/or TPCA analysis, just push the corresponding buttons.

Data Preprocessing

Data preprocessing module aims to transform the proteomics quantitative table given by mass-spectrometric analyzing software into the acceptable data matrix of ProSAP. Video of example

Usage:

  1. Click 'Open' button, select raw data files output by PD software. ProSAP can accept single or multiple files. If there is no technical replicates in this step, it is alright to choose a single file. Note that technical replicates will be merged to one data matrix, which is not equal to sample replicates used for TPP analysis. Example files

  2. If you need remove proteins with low PSM values, please choose the column correspending to the PSM values at 'PSM column name' combo box.

  3. Choose how to merge the values of multiple technical replicates.

  4. Choose the normlization method. For TPP analysis, usually use 'reference' method, i.e. all the columns are divided to the reference column; For iTSA method, usually use median method, which is similar to the standard normalization in proteomics analysis. If 'reference' method is used, please choose which column is set as reference in the 'Reference column' combo box.

  5. Choose the missing value imputation method. 'KNN' means use KNN based missing value impuataion; 'Zero' means replace missing values as 0.

  6. Input the RSD threshold for filtering proteins with high RSD in technical replicates. If only one file is input, ignore it.

  7. Click 'confirm' button, and wait for finishing.

  8. Save the results.

iTSA/PISA Analysis

Although the experimental protocol of PISA and iTSA are different. PISA mixes samples under different tempertures, while iTSA uses single experimental temperture. However, their statistical analysis can be similar. The core is to determine whether a protein is significantly different between case and control groups. ProSAP implements four different methods for this purpose, which are t-test, Limma, edgeR, and DESeq2. Video of example

Note:
Both edgeR and DESeq2 are designed for designed for count data of transcriptomics with negative binomial distribution assumption. They were also applied in proteomics in some recent works. So, they are embedded in ProSAP as options. However, the applicability may need more experimental validation.

Usage:

  1. Click 'Data' button, and select a quantitative data file of proteomics. The first column of the data table should be 'Accession', which could be gene name, uniprot id, or any other index. Example files.

  2. Select columns representing protein abundences. If the abundances have been log2 transformed, you should set 'log2 transformed' combo box as 'True'.

  3. Select statistical method and parameters, then click 'Confirm' button

TPP/NPA Analysis

The workflow of TPP analysis is implemented under the Savitski protocol. Nonparametric analysis (NPA) is an alternative method of TPP analysis. It aims to avoid the arbitrary of parameter setting in TPP analysis. NPA includes two different strategies: fitness-based model and distance-based model. Fitness-based model is mainly followed the NPARC method Distance based model is more straightforward, the data of case and control groups are fitted into melting curve, respectively. Various kinds of distance functions, such as Euclidean, Cityblock, Chebychev or cosine function. Video of example

Usage:

  1. Click 'Data' menu and load proteomics files. Please select multiple preprocessed data files. one or two replicates of control group, and one or two replicates of case group. The first column of each file should be 'Accession'. Example files.

  2. Take turns to select data files and click 'Rep 1 Control', 'Rep 1 Case', 'Rep 2 Control' and 'Rep 2 Case'. In the meantime, the data are browsed.

  3. Click 'Params' button and set the parameters, then click 'Analysis' menu and choose the statistical method. The process will take some minutes.

  4. View and save the results. The melting curve of proteins can be visulized by clicking the 'Show Curve' button.

TPCA Analysis

TPCA is an analysis method of coaggregation of proteins based on TPP experiments. ProSAP implements the TPCA procedure for visualizing protein complex dynamic changing and protein pair prediction.Moreover ProSAP also calculate p value for sorting the protein complex in database, which assists users to find which protein complex changes most significantly. Video of example

Usage:

  1. Click 'Data' menu and load proteomics files. Please select multiple preprocessed data files. one for control group, and one for of case group. The first column of each file should be 'Accession', and the 'Accession' must be uniprot ID. Example files.

  2. Take turns to select data files and click 'Set as G1', 'Set as G2'. In the meantime, the data are browsed.

  3. Click 'Data' menu and load protein complex database. The database file should include 'Complex_ID' and 'Subunits_UniProt_IDs'. The 'Subunits_UniProt_IDs' must represented by uniprot ID

  4. Select the database file to be used, and click 'Confirm' button. Example database is available at here

  5. Click 'Calc Change' button. The processing will take some minutes.

  6. The protein complex will be sorted by the changing significance, using p-value as numerical description

  7. View and save the results. The melting curves of proteins can be visulized by clicking the 'Show Curve' button.

  8. If you want to calculate protein pair dynamic changing and/or evaluate the prediction performance of PPI, just click 'Analysis' menu, and select 'Protein pair analysis'

2D-TPP analysis

2D-TPP employs a multiplexed MS analysis of samples in the presence of n ligand concentrations (including a vehicle control) at m temperatures. The additional drug concentration dimension of 2D-TPP enables more sensitive detection of off-target targets. The 2D-TPP analysis of ProSAP is implemented based on the R package proposed by Savitski et al. Video of example

Usage:

  1. Click 'Data' menu and load proteomics files. Please refer the column names of the example file. Example files.

  2. Set the parameters, then click 'Confirm' button. The processing will take quite a long time, wait patiently.

  3. View and save the results. The heatmaps of proteins can be visulized by clicking the 'Show Result' button.