Welcome to MultiOmicsXplorer!
This tool analyzes phosphoproteomic, proteomic, and transcriptomic data from various types of tumors at different stages, allowing comparisons between tumors and stages.
In the OncoXplorer-CPTAC Data Analysis section, you can compare tumor data from the CPTAC portal across different stages and tumor types.
The Extract Protein Activity from Your Data section allows you to use SignalingProfiler to extract protein activity uploading your data.
Click here for the tutorial to learn how to use the app.
CPTAC data
Data Processing and Harmonization
We use data that have been downloaded using the tutorial described in the following paper:
Simplified and Unified Access to Cancer Proteogenomic DataWe collected clinical data - used to obtain informations about the tumors stages, and raw transcriptomics proteomics and phosphoproteomics data.
Pre-processing Steps
The raw data underwent several filtering and transformation steps to ensure consistency and usability:
- Transcriptomics: Filtered out rows with more than 80% zero values and removed non-coding elements.
- Proteomics: Updated protein identifiers to match Uniprot ID.
- Phosphoproteomics: Updated Uniprot ID and sequence window based on the phosphorylated residue.
Harmonization Process
To standardize the data across different omics layers, we applied the following transformations:
- Missing values (NA) in proteomics and phosphoproteomics data were imputed.
- Z-score normalization was applied to all three omics datasets to facilitate comparisons.
If you need to harmonize your data, you can easily access the PatientProfiler package developed by Veronica Lombardi, Eleonora Meo and Veronica Venafra.
SignalingProfiler Tool
This application uses the SignalingProfiler pipeline to perform protein activity inference as described in the following paper:
SignalingProfiler: A Tool for the Quantitative Analysis of Proteomics Data (Venafra V., Sacco F., Perfetto L., 2024)Click here to access the complete tutorial
The following three sample dataframes are provided to illustrate the proper format for providing omics dataframes to input to SignalingProfiler.
To ensure analysis, the columns must be renamed as they are in the examples.
All three omics dataframes can be loaded to initiate the analysis, or if one or two are missing, the analysis will be performed in the most optimal way based on the loaded dataframe(s).
Transcriptomics Data Example
Proteomics Data Example
Phosphoproteomics Data Example
Upload data for one or multiple samples
Note: File names should follow the pattern below:
- P_<ID> for proteomics (e.g., P_C3L00010.tsv/.csv/.xlsx)
- Ph_<ID> for phosphoproteomics (e.g., Ph_C3L00010.tsv/.csv/.xlsx)
- T_<ID> for transcriptomics (e.g., T_C3L00010.tsv/.csv/.xlsx)
For example, if the sample ID is C3L00010 and you want to upload proteomics data, the file name should be: P_C3L00010.tsv/.csv/.xlsx.
Pipeline Parameters
TF Activity Footprint Analysis
Kinase-Phosphatase Activity Footprint Analysis
PhosphoScore Computation
View the top 15 up and down regulated proteins based on their predicted activity
Here you can upload your results from the protein activity inference to visualize your data in a barplot with a separation by molecular function.
(All Samples shows a mean for the predicted activity in your cohort)
Download Plot- tf: Transcription Factor
- kin: Kinase
- phos: Phosphatase
- other: Other
Help and Documentation
Common Issues
Q: I can't see the plots after selecting the analytes. What should I do?
A: Make sure you have selected at least one tumor and one stage. Also, ensure that the data for the selected analyte is available.
Q: The data download button is not working. What should I do?
A: Please check your internet connection and ensure that your browser allows downloads.
Contact Information
For further assistance or inquiries, feel free to contact us:
PerfettoLab of Bioinformatics, University of Rome "La Sapienza"
Head of the Lab: Dr. Livia Perfetto
Primary curator of MultiOmicsXplorer: Dr. Eleonora Meo
Email: eleonorameo.hp@gmail.com
About the Lab
PerfettoLab specializes in bioinformatics research and analysis, with a focus on cancer genomics and proteomics. We are dedicated to developing tools like MultiOmicsXplorer to facilitate data analysis for cancer research.