PEGASAS

The PEGASAS Server is an interactive web app for the user to perform analysis using PEGASAS.

General Use of PEGASAS Server

Required Inputs

  • Alternative Splicing Measurements: TXT file of alternative splicing measurements. First eight columns should be in the following order: Accession, Gene Name, chromosome number, strand (-/+), exon start site, exon end site, upstream exon end site, downstream exon start site. The following columns should be patient ID’s as the column names and alternative splicing measurements (PSI values) corresponding to each patient ID as the values.
  • Gene Expression Values: TSV file of a matrix of gene expression values where each column is one gene and each row is one sample. For a given row, the first column should be the patient ID and the following columns should be gene expression levels for the patient sample.
  • Group Information: TSV file of patient ID and phenotype/disease stage in each row (two columns total).
  • Gene Set: Gene set of pathway of interest from a given list of gene sets.

Creating Table of Events

After uploading all files and selecting the desired gene set in the sidebar panel, a table of events can be created. This table will show all alternative splicing events in the file of alternative splicing measurements along with the coorelation coefficient and p-value of each event calculated with the PSI Values and patient pathway scores.

Note that the p-value of the coorelation coefficient is different from the stand-alone PEGASAS tool which uses a permutation-based procedure.

After all inputs are provided, click the ‘Make Table’ button to create the table of events. The table of events can be sorted by coorelation coefficient and filtered by both correlation coefficient and p-value. The table can be downloaded by clicking the ‘Download Table’ button.

The density plot at the bottom of the sidebar panel shows the coorelation coefficients across the filtered events.

To plot an event, select the row with the desired event in the table of events and click the ‘Plot’ button.

Customizing the Plot

There are different elements can be displayed on the plot:

  • Regression Line
  • Density Contours
  • Color by Phenotype
  • Density Plots (for each axis)

Additionally, to view specific phenotypes on the scatter plot, the user can check and un-check the boxes next to the phenotype names. All phenotypes are automatically checked when the plot is first made.

The confidence level of the regression line can also be changed with the ‘Set Confidence level’ slider.

Interacting with the Plot

  • Zoom: Zoom in by dragging your mouse across the plot and double clicking the window. Zoom out by double clicking the window without dragging.

  • View point data: The data of a point can be viewed by clicking on the point. The point information will be displayed in the table at the bottom of the page. If there are multiple points near the mouse click, these points will all be displayed in the table.

Downloading the Plot

The current plot can be downloaded by clicking the ‘Download Plot’ button. The recommended dimensions are 6.5in by 6.5in. Note that some plots with density plots will require a wider window, or plot title will be cut off.

Summary Plot of PEGASAS

Required Inputs

  • Pathway Exon Correlation Matrix: TXT file correlation coefficients between each signaling pathway and alternative splicing event. The first column contains the names of alternative splicing events, and the following columns contain the corresponding correlation coefficient for each signaling pathway. If the pathway is not correlated with the event (decided by the p-value cutoff and effect size in PEGASAS), there will be an ‘NA’ in the corresponding position.
  • Gene Ontology results: TXT file of the top selected enriched GO terms and their p-values for each signaling pathway.The first column contains the names of the signaling pathways, the second column contains the top enriched GO terms separated by ‘;’, and the third column contains the corresponding adjusted p-value for each GO term separated by ‘;’.

Selecting Signaling pathways

Up to ten signaling pathways can be selected for the hive plot. A pathway can be added by selecting the desired pathway from the dropdown menu and clicking the ‘Add Pathway’ button. The order of the pathways in the hive plot can then be rearranged in the ‘Drag to sort’ box. To clear all selected pathways, press the ‘Clear List’ button.

References

Phillips J.W. , Pan Y. , Tsai B.L., Xie Z., Demirdjian L., Xiao W., Yang H.T., Zhang Y., Lin C.H., Cheng D., Hu Q., Liu S., Black D.L., Witte O.N.^+ , Xing Y.^+ (2020) Pathway-guided analysis reveals Myc-dependent alternative pre-mRNA splicing in aggressive prostate cancers. Proc. Natl. Acad. Sci. U.S.A., In Press (^+ joint corresponding authors; * joint first authors)

Contacts

Beatrice Zhang beazhang@sas.upenn.edu
Yang Pan panyang@ucla.edu
Yi Xing Yi.Xing@pennmedicine.upenn.edu