Introduction

Welcome to the dfPPI analysis platform, a Shiny web application for the investigation of protein-protein interaction (PPI) networks. Dysfunctional Protein-Protein Interactome (dfPPI) is a newly developed chemoproteomic platform designed to capture context-dependent dynamic changes in protein-protein interaction (PPI) networks under stressor-induced cellular perturbations. dfPPI specifically interrogates disease-associated disruptions in native (unengineered) cells and tissues, focusing on how chaperone-based scaffolds – termed “epichaperomes” – rewire thousands of proteins into aberrant complexes. This reorganization of the interactome underpins diverse pathological states by linking stressors directly to a dysfunctional phenotype. Through dfPPI analyses, researchers can identify which proteins and pathways become pathologically connected, quantify these altered interactions in a range of disease contexts, and pinpoint critical targets for therapeutic intervention. dfPPI offers a systems-level lens for understanding, visualizing, and ultimately modulating PPI network dysfunctions in complex diseases, facilitating both fundamental discovery and the development of precision treatments

Herein, we make available an open resource with a bioinformatics pipeline, ‘dfPPI analysis platform’ that members of the broad scientific community – with or without computational background – can access freely and without restriction.

The outcome is a dynamic site with interactive network plots and complex analysis input screens. This allows end users to change analysis parameters and interact with plots, individual proteins, and pathways, to analyze epichaperomics datasets to inform disease biology, as well as help in the development of precision medicine for treatment and prevention. The app aims to provide a platform for researchers with limited programming skills to facilitate rapid data processing and visualization.

About the tool

This analysis platform is developed by R Shiny (Version 1.10.0) and is free and open to all users with no login requirement. It can be readily accessed by all popular web browsers, including Microsoft Edge, Google Chrome, Mozilla Firefox, and Safari. In the app, the majority of analysis and plotting functions are written in R programming language with popular packages, to produce high-quality images, similar to running bona fide R codes minus the coding process. We expect feedback from active users to refine the app moving forward.

How to Use the Tool

Here is an overview of how to analyze your data using the dfPPI Platform. While each tab in the navbar has its own settings and instructions, the workflow typically follows the sequence below:

  1. Upload Data
    • Choose either sample data (preloaded demos) or upload your own file (e.g., CSV, Excel).
    • Confirm that you have selected the correct file format and measurement type (e.g., Raw MS1 intensities, MS2).
    • (Optional) Upload a custom contaminant list if needed.
    • Once validated, click the 'Upload' button to process data.
    • Review classified data by clicking the “Review Data Classification” button.
    • Submit changes to finalize.
  2. Data Transformation
    • Select a transformation (e.g., log, power) and optional normalization (e.g., Z-score).
    • Inspect summary statistics and histogram-based plots to confirm the chosen method.
    • Apply the transformation and review the updated data distribution.
  3. Data Imputation
    • (Optional) Filter out proteins/features with excessive missing values.
    • Choose an imputation method (e.g., kNN, MLE, PCA-based).
    • Check logs and diagnostic plots (heatmaps, evaluation metrics) to confirm missing data is handled properly.
  4. Differential Capture
    • Define contrasts or comparisons (e.g., disease vs. control).
    • Select the analysis method (t-test or linear model).
    • If there are less that 3 samples/replicates per contrast level, you can simply calculate the fold change.
    • For even lower sample size, you can skip the differential capture step altogether.
    • Run the analysis, and review volcano plots for significant protein changes.
  5. Enrichment Analysis
    • Filter results by fold-change and p-value cutoffs to select candidates for pathway enrichment.
    • Choose a pathway analysis method (e.g., g:Profiler) and relevant data sources (e.g., Reactome, GO).
    • Optionally you can also run interactome based enrichment analysis (iEA; formerly iGSEA) to get pathway enrichment for the direct interactors of chaperome proteins.
    • Inspect the pathway enrichment plot to pinpoint functional categories most affected by these PPI changes.
  6. Pathway Network Visualization
    • Visualize your PPI networks/pathways in interactive network plots.
    • Hover/click nodes to see details about altered proteins, fold changes, etc.
    • Use the search box to locate specific proteins or pathways of interest.
  7. Download & Further Analysis
    • Export tables or figures (e.g., transformed data, imputed dataset) for offline validation.
    • Integrate outputs in other tools (like Cytoscape) for advanced manipulation.

At each stage, consult the log panels for progress and quality checks. For more tips on best practices and advanced features, see the in-app tooltips or user documentation links provided.

Citations

Chakrabarty S, Wang S, Roychowdhury T, et al. Introducing dysfunctional Protein-Protein Interactome (dfPPI) - A platform for systems-level protein-protein interaction (PPI) dysfunction investigation in disease. Curr Opin Struct Biol 2024, 88:102886. https://doi.org/10.1016/j.sbi.2024.102886. PMCID: PMC11392609.

Rodina A, Xu C, Digwal CS, et al. Systems-level analyses of protein-protein interaction network dysfunctions via epichaperomics identify cancer-specific mechanisms of stress adaptation. Nat Commun. 2023 Jun;14(1):3742. https://doi.org/10.1038/s41467-023-39241-7. PMCID: PMC10290137.

Ginsberg SD, Sharma S, Norton L, et al. Targeting stressor-induced dysfunctions in protein-protein interaction networks via epichaperomes. Trends Pharmacol Sci. 2023 Jan;44(1):20-33. https://doi.org/10.1016/j.tips.2022.10.006. PMCID: PMC9789192.

Ginsberg SD, Neubert TA, Sharma S, et al. Disease-specific interactome alterations via epichaperomics: the case for Alzheimer's disease. FEBS J. 2022 Apr;289(8):2047-2066. https://doi.org/10.1111/febs.16031. PMCID: PMC8611103.

Joshi S, Gomes ED, Wang T, et al. Pharmacologically controlling protein-protein interactions through epichaperomes for therapeutic vulnerability in cancer. Commun Biol. 2021 Nov;4(1):1333. https://doi.org/10.1038/s42003-021-02842-3. PMCID: PMC8617294.

Inda MC, Joshi S, Wang T, et al. The epichaperome is a mediator of toxic hippocampal stress and leads to protein connectivity-based dysfunction. Nat Commun. 2020 Jan;11(1):319. https://doi.org/10.1038/s41467-019-14082-5. PMCID: PMC6965647.

Contact

For questions and comments, contact: Souparna Chakrabarty (chakras6@mskcc.org), Gabriela Chiosis (chiosisg@mskcc.org)

Please provide detailed feedback here

For more information, visit the Chiosis Lab website, Institutional website