iPathwayGuide User Guide v1 - Advaita Bioinformatics 5 Organism:)Choose)your)organism.)...

!

User Guide

v 1.1

Released January 27, 2014

Developed by:

Copyright (C) Advaita Corporation 2015

Page �2Contents

Background and Introduction 3 Dashboard 4

Data Upload 6 Affymetrix CEL file upload 6 Report Intake Form 7 Viewing, Sharing, and Purchasing Reports 9

Summary 11 Editing and Updating Projects 12 Multiple Contrasts 13

Genes 14 Predicted Micro RNAs (miRNAs) 16 Gene Ontology 19 Pathways 21

Pathway Plot 21 Pathway Map 22

Pathway Details 23 Two-Way Plot 23 Gene Table 23 Bootstrap Diagram 24 Related Drugs, miRNAs, SNPs 25 Pathway Legend 26

Diseases 27 Meta-Analysis 28

Existing Report 28 Constructing a New Meta-Report 31

Exporting/Downloading 33 Help and Video Tutorials 33 Citing iPathwayGuide 34


Page �3

Background and Introduc.on Advaita’s iPathwayGuide offers the most advanced pathway analysis among other analyses. iPathwayGuide is the only pathway analysis platform that considers how each gene is connected in the system to provide the most accurate results.

iPathwayGuide is designed to be easy and simple to use. There are eight main sections within iPathwayGuide:

Dashboard

Summary

Meta Analysis

Genes

Predicted Micro RNA

Gene Ontology

Pathways

Diseases

Each section is designed to provide speciEic insights to your data while preserving the relationships across the entire analysis. The analyses provided through iPathwayGuide have been built to the highest standards by experts in their respective Eields.


Page �4

Dashboard

The dashboard contains a list of all your pathway analysis reports and is the starting point for all new analyses.

There are 5 possible status indications for each report: • Pending -‐ a report that is currently being processed • Trial -‐ a report that has completed processing and is available for review for 72 hours 1• Expired -‐ a report that is completed, but the preview period has expired • Purchased -‐ a report that has been purchased • Error -‐ a report that has failed processing

To generate a new analysis:

Clicking on the � will take you to the data intake page:

Preview period timeframe is subject to change without notice.1


Page �5

Organism: Choose your organism. iPathwayGuide supports Human, Mouse, and Rat.

File Format: Choose the type of Eile you are working with. iPathwayGuide accepts the following data Eile formats:

• CuffDiff • DESeq(2) • JMP Genomics (a division of SAS) • Custom tab-‐delimited differential expression Eile • GEO2R/Limma (R/Bioconductor) • Affymetrix CEL Eiles

CuffDiff: Select the Eile from your CuffLinks output that ends with “_gene_exp.diff” or .txt DESeq: Choose a DESeq output Eile -‐ typically a txt Eile JMP Genomics: Select the iPathwayGuide output from JMP Genomics *Custom tab-‐delimited File: Choose a generic tab-‐delimited .txt Eile that contains gene symbol, log2FC, and adjusted p-‐value as in the example below.

* It is recommended that you use a list of all the genes that were measured. Cutoff thresholding will be performed in the next step.

GEO2R/Limma: Choose the resulting Eile from processing through GEO2R or Limma Affymetrix CEL Files: Raw Affymetrix CEL Eiles from the following platforms

Human Human Genome U133 Human Genome U133A 2.0 Human Genome U133 Plus 2.0 Human Genome U95 Human Genome U35K

Mouse Mouse Expression Set 430 Mouse Expression Set 430 2.0 Mouse Genome 430A 2.0 Rat Rat Expression Set 230 Rat Genome 230 2.0 Rat Genome U34

See below for speciEic upload dialogs for Affymetrix CEL Eiles

Note for BaseSpace Users: Select the app result from the CuffLinks, RNA Express, or the Protein Assembler workElow.

Gene Symbol Log2 Fold Change Adjusted p-value

-4.74343 0.01789

E"d2 2.15787 0.23478

Myd88 0.44178 0.78945

� Erdr1


Page �6Data Upload

Input data File: Click on the � to choose your input Eile and begin the uploading process.

Affymetrix CEL file upload

If you choose Affymetrix CEL File as your Eile type, a dialogue will prompt you to choose the CEL Eiles that belong to the condition and the control groups. You must select at least 3 Eiles for each group.

Once you have chosen the appropriate Eile(s), the data will be uploaded to the Advaita servers for initial processing. Some of these Eiles can be quite large and upload times can take several minutes depending on the quality of your data connection.


Page �7Report Intake Form

Once the data has been uploaded, the Report Intake form is displayed to gather additional information.

Report title: Provide your report a descriptive title that you and your colleagues will understand. For example: “Tumor vs Normal Breast cancer group 2”

Report description: Provide a detailed description of your experiment.

Contrasts: Provide a contrast label (e.g.: Tumor vs. Normal). Some Eiles will auto-‐detect the contrast labels. Set the thresholds to select the signiEicant DE genes. Default values are:

Log FC: 0.6 p-‐value: 0.05

Clicking on “View/Apply” of a contrast will highlight that row and display the volcano plot associated with that contrast.

If your input Eile has multiple contrasts, you may select up to 5 contrasts to be analyzed in a single report.

Once the report title and report description are entered and an input data Eile has been chosen, the button in the bottom right corner will turn green:


Page �8

Pressing the � button will submit the Eile for analysis. Each data set takes approximately 10-‐15 minutes to analyze. If the server is busy with other reports, it may take up to 24 hours to complete. You will receive an automated email with a link to the analysis once the analysis is complete and ready to view.

A note on multiple contrasts. Not all Eile types support multiple contrasts in a single Eile. Please refer to the table below for more information.


Page �9Viewing, Sharing, and Purchasing Reports

Viewing a Report: To view a completed report from your Dashboard, click on the title of the report and the analysis will load. Expired reports must be purchased before further viewing. Pending reports must Einish processing before they can be viewed.

Sharing a Report: To share a report, click on the “Share” link next to the report title in your Dashboard or in the report. A window will pop up where you may add as many emails as you need to share the report. You may also set whether a shared report may continue to be shared to others.

If you see “Share” that means the report either has not Einished or the person who shared it with you did not give you permission to share it onward.

Note for BaseSpace Users: The share function is not available in the app because result sharing is directly available from within BaseSpace.

Purchasing a Report: To purchase a report, click the shopping cart icon next to the status of the report.

You may purchase the report either with a credit card, or you may call for a subscription to iPathwayGuide.


Page �10

When a report has been purchased successfully, you will get the following message:


Page �11

Summary The Summary tab is designed to provide a very brief high-‐level overview of your data. The summary page consists of:

• The report title and description as entered by the person who submitted the project. • Volcano plot of DE genes • Two-‐way evidence plot of pathways: Perturbation vs. Over-‐representation

(enrichment) • Top scoring pathways • Top Predicted micro RNAs (miRNA’s) • Top Over-‐represented Biological Processes • Top Over-‐represented Molecular Functions • Top Over-‐represented Cellular Components • Top Over-‐represented Diseases

Clicking on the sections will take you to the respective detailed analysis.


Page �12Editing and Updating Projects

Editing Description/Title You may edit the project title or the description by clicking the “+” next to the Project title. From the dialogue box, you may change any of the text. Be sure to click “Save Changes”. If you are not the project owner, you cannot edit the project, but you may send an automated email to the project owner to make an update.

Updating a Project If the background data or a new analysis is available, you may run the report again and get the latest report. Your previous report will be preserved and a history of each report will be available from your dashboard.


Page �13Multiple Contrasts

iPathwayGuide supports multiple contrasts in a single report, because both the CuffLinks and DESeq(2) workElows support multiple contrasts in a single output Eile. These will be auto-‐detected when uploading your data. This unique feature allows for quick switching between various contrasts within the same experiment. For example, you may have the following comparisons:

• UNTR vs. TNF • UNTR vs. GMCSF • TNF vs. GMCSF • Meta-‐Analysis -‐ Please see page 25 for this topic

From the menu bar, select which contrast you would like to view.


Page �14

Genes The Genes tab is designed to provide detailed information about each gene that was considered in the analysis (signiEicant differentially expressed genes.) Genes may be selected either by clicking on the gene in the volcano plot or in the gene table below. You may also search for a gene by its ofEicial gene symbol or Entrez gene ID.

When a gene is selected, information about that gene is displayed, along with literature references and relationships to: biological processes, molecular functions, cellular components, pathways, drugs, miRNAs, SNPs, and diseases. Clicking on any of these links will take you to the respective analysis.

If you do not see your gene listed, you may choose to “show all genes” at the top right corner of the volcano plot. It may take a few seconds to load all the genes from your input Eile. When this option is selected, both black and red dots are displayed. Red dots identify signiEicant genes, while black dots identify non-‐signiEicant genes.

When a black dot is selected, relationships to other elements like GO Terms and pathways will lead to external sources because that gene was not considered in the analysis.


Page �15

� �


Page �16

Predicted Micro RNAs (miRNAs) iPathwayGuide has the unique ability to predict if microRNAs (miRNAs) are active in your samples based on the gene expression signature of the target genes for each miRNA.

The prediction of active miRNAs is based on enrichment of the target genes. Because miRNAs are generally considered to have an inhibitory effect, the method looks at the number of downwardly expressed DE targets versus all DE targets, and compares that to the total number of downwardly expressed targets vs all targets. From these relationships the method calculates a p-‐value for each miRNA based on the likelihood of observing the number of DE targets by random chance alone.

The layout for the miRNA analysis has 4 key sections: • miRNAs summary • miRNA details • Differentially expressed target genes • Literature references


Page �17

In the miRNAs summary section, you may choose between two different views by clicking on the label in the top-‐right corner.

Counts View: The “Counts View “is a bar graph that shows the count of DE targets that are downwardly expressed (blue bars) and the number of DE target genes that are upwardly expressed (red bars) for each miRNA. The selected miRNA bars change to green and orange to indicate its selection. By default the miRNAs are ordered by signiEicance (most to least) without correction. You may click on any set of bars to select that particular miRNA. p-‐value View: The p-‐values view plots two p-‐values. On the horizontal axis (-‐log10(pv)), the p-‐value based the total number of DE target genes versus the total number of target genes. (ignoring direction of change This p-‐value is not used for selecting signiEicance. On the vertical axis (-‐log10(pvn)), the p-‐value based on the number of downwardly expressed DE targets versus the total number of DE targets. This p-‐value is the basis for selecting for signiEicance. The size of each dot represents the relative number of target genes for that miRNA with a larger size indicating a larger number of target genes.

Below the plots, is a table with the relative counts and a global p-‐values for each miRNA. By default, no correction is applied, but you may choose to apply FDR or Bonferroni correction factors.


Counts View p-‐value View

Page �18miRNA Details provide the description of the miRNA and a link to miRBase for additional information about the selected miRNA

Differentially expressed target genes provides a plot of the target DE Genes for the selected miRNA with the relative measured fold change from the experiment and an overall box plot. The DE targets are ordered from greatest negative FC to greatest positive FC. Depending on the number of DE targets associated to the selected miRNA, you may need to scroll the table to the right or left to view all the targets. Clicking on any one of the bar plots for the genes will take you to the genes tab for additional information about that gene.

Literature references are provided for the selected miRNA.


Page �19

Gene Ontology From the Gene Ontology (GO) analysis menu, select one of three domains:

• Biological Processes • Molecular Functions • Cellular Components

The layout and navigation is identical for all three GO analyses.

On the left is the GO graph with the table directly below it. On the right is the detailed information about the selected GO terms and below that is the ancestor chart. More information about GO Analysis can be found at http://www.ebi.ac.uk/QuickGO/

GO Graph

The GO graph is a ranked graphical representation of the GO terms over-‐represented in the dataset that was analyzed. Magenta blocks are more signiEicant than cyan blocks. By default, only signiEicant terms are displayed. Various correction factors (FDR, Bonferroni,


http://www.ebi.ac.uk/QuickGO/

Page �20Elim Pruning, Weight Pruning) and thresholds of minimum DE genes can be applied, and the table will automatically update. More information about Elim and Weight Pruning can be found at Alexa et al (2006).

Differentially Expressed Annotated Genes

For Each GO term you will see the differentially expressed genes that are directly annotated to that term with the measured expression change and a box-‐plot for the overall expression. Clicking on any one of the genes bars will take you directly to the Genes tab with that gene selected.

Ancestor Chart

For each GO term, an ancestor chart is provided for further analysis. The chart is interactive and can be dragged, zoomed, and manipulated. Clicking on a GO ID will take you to the chart with that node as the root node.


http://bioinformatics.oxfordjournals.org/content/22/13/1600.short

Page �21

Pathways

The Pathways tab provides a Eirst view of the pathway analysis results. On the top left is the two-‐way pathway plot and table with pathways sorted by p-‐value. By default, no correction factor has been applied. Both FDR and Bonferroni correction factors can be applied by selecting the correction factor from the spinner between the plot and table. A description of the currently selected pathway can be displayed by clicking on the “show description” label on the top-‐right of the window..

Pathways can be selected by clicking on the dots in the two-‐way plot or by clicking on the p-‐value in the table.

Pathway Plot

The pathway plot is one of the unique attributes to iPathwayGuide’s Impact Analysis. Each dot represent a different pathway. The size of each dot denotes the number of genes on the pathway. The x-‐axis measures the p-‐value obtained using the classical over-‐representation or enrichment analysis (pORA). The y-‐axis represents the p-‐value obtained from total


Page �22perturbation accumulation (pAcc) in the pathway. The table directly below the plot shows the combined global p-‐value for each pathway. More information about this method can be obtained here. http://vortex.cs.wayne.edu/papers/Genome_Research_reprint1537.pdf

Pathway Map

The pathway map, by default, displays the measured fold change for each gene on the pathway (Log FC). Total Accumulation and Total Perturbation (Fold Change + Total Accumulation) may be selected with the radio buttons above the pathway diagram.

Total Accumulation is calculated by taking the connection between the genes (as described by the pathway) and developing a system of equations to solve how the observed fold change values in the genes on the pathway must be connected.

Total Perturbation is the combination of both the measured LogFC plus the Total Accumulation, thus describing the total perturbation within the selected pathway.

Coherent Cascades are sections of the pathway where the data and the pathway diagram are coherent (i.e. the pathway diagram shows gene A represses gene B and the data is consistent with that state). These Coherent Cascades can serve as putative mechanisms for all the observed and calculated perturbations in the systems. These may be toggled on or off.


http://vortex.cs.wayne.edu/papers/Genome_Research_reprint1537.pdf

Page �23

Pathway Details Detailed information about the pathway can be obtained by clicking any of the buttons on the right hand side of the pathway map. The Pathway Details view allows for the interrogation of details about the speciEic pathway of interest. Details include:

•Two-‐Way Plot •Gene Table •Bootstrap Distribution (for perturbation) •Associated Drugs •Associated miRNAs •Associated SNPs •Associated Diseases

Two-‐Way Plot

The Two-‐Way Plot is an alternative way to view the genes on the pathway. The y-‐axis plots the measured fold-‐change for each gene, while the x-‐axis plots the calculated accumulated perturbation. Clicking a gene on the pathway will also highlight it in the Two-‐Way Plot.

Gene Table

The gene table contains the list of genes represented in the selected pathway. The table displays the gene symbol, Entrez gene id, fold change, calculated perturbation accumulation, and total perturbation (measured + calculated).


Page �24Each column can be sorted by clicking on the headers. Individual genes may be searched by using the search bar at the top of the table. Selecting genes on the pathway diagram can also Eilter the table to that selection.

Bootstrap Diagram

The bootstrap diagram shows the distribution (mean=0, std dev=1) of expected accumulated perturbation values based on the observed data for the selected pathway. 2,000 iterations are used to construct the distribution. The red line indicates where the actual observed values lie in the distribution of expected results. The further away from the mean, the less likely the observed values are due to random chance. This is the value used for the p-‐value on the vertical axis in the Pathway Plot.


Page �25

Related Drugs, miRNAs, SNPs

The related tables for Drugs, miRNAs, and SNPs list those items known to be associated with the genes on the selected pathway. Selecting a gene will further Eilter the list to display those items only related to that gene. Clicking on a check box will highlight the target gene(s) in the pathway. Clicking on the name will bring up additional information about that item.

miRNAs and SNPs have Eiltering capabilities. For SNPs, the default Eilter only shows SNPS that are both validated and clinically signiEicant.


Page �26Pathway Legend

See below for a legend of symbols for KEGG pathway diagrams.

Additional information on how to read a KEGG pathway image can be found at http://www.genome.jp/kegg/document/help_pathway.html


http://www.genome.jp/kegg/document/help_pathway.html

Page �27

Diseases The Diseases tab displays the over-‐represented diseases based on the list of DE genes used in the analysis. The diseases are categorized by the International ClassiEication of Disease – 2010 (ICD-‐10). As with the GO analysis, magenta blocks are more signiEicant than cyan blocks.

Working from the interior of the graph to the exterior, users may narrow the list of diseases based on groups and subgroups of diseases. For example, clicking on the “Neoplasms” groups will display those disease subgroups over-‐represented in the data associated to Neoplasms.

For each disease term you will see the differentially expressed genes that are directly annotated to that term with the measured expression change and a box-‐plot for the overall expression. Clicking on any one of the genes bars will take you directly to the Genes tab with that gene selected.


Page �28

Meta-‐Analysis

iPathwayGuide has the unique capability to compare conditions or contrasts within an experiment, or build a new meta-‐analysis from existing contrasts across several experiments.

Currently, iPathwayGuide supports analysis of up to 5 contrasts in a single meta-‐analysis.

Existing ReportTo analyze contrasts in an existing report, from the contrasts menu, select Meta-‐Analysis.


Page �29The default view for the meta-‐analysis is genes. The Venn diagram shows the intersections of the contrasts with the number of entities that are in common in that region. Selecting the genes tab shows the number of genes for each region; selecting the miRNAs tab will show predicted miRNAs, etc.

Selecting one or more regions will quickly Eilter the list of results to the entities for that selection.


Page �30

When analyzing entities with multiple annotated genes, such as pathways, on the right hand side, information about the selected entity is provided including the relative expression values for the the annotated genes -‐ color coded to the respective contrast(s).


Page �31Constructing a New Meta-Report

To generate a new meta-‐analysis based on existing reports:

From the dashboard, clicking on the will take you to the meta-‐report intake page:

From the Meta-‐report Intake form, provide a title, description, and select which contrasts to include.


Page �32

Only contrasts from purchased reports can be included for building a custom meta-‐report. Select up to 5 contrasts to include by expanding each report and selecting the desired contrast.

Once you have made your selections, you may create the report.

New Meta-‐reports are built immediately and should appear in your dashboard.


Page �33

Expor.ng/Downloading Images and tables can be exported or downloaded by clicking the download button . During the free preview period, this capability is disabled.

Help and Video Tutorials In each section, there are a number of resources available to answer any questions you may have. Simply click on the “!” in the section you are interested in to get a pop up window with a linked video or additional information.


Page �34

Ci.ng iPathwayGuide Using Advaita Bio’s products or content for any form of publication (e.g. print, electronic, etc.) requires citation. Please use one of the options below for citations:

• “The Data (signiEicantly impacted pathways, biological processes, molecular interactions, miRNAs, SNPs, etc.) were analyzed using Advaita Bio’s iPathwayGuide (http://www.advaitabio.com/ipathwayguide)”.

• “This software analysis tool implements the ‘Impact Analysis’ 1 approach that takes into consideration the direction and type of all signals on a pathway, the position, role and type of every gene, etc., as described in (Draghici, 2007, Donato, 2013).”1

1 Sorin Draghici, Purvesh Khatri, Adi Laurentiu Tarca, Kashyap Amin, Arina Done, Calin Voichita, Constantin Georgescu, and Roberto Romero. A systems biology approach for pathway level analysis. Genome Research, 17:1537-‐1545, 2007.

• LaTeX users may use the following code in the bibtex Eile:

~\cite{advaita2014} @ONLINE{advaita2014, author = {Advaita, Corporation}, title = {Pathway Analysis with iPathwayGuide}, month = Jun, year = {2014}, url = {http://www.advaitabio.com/ipathwayguide.html} }

For additional guidance, please contact [email protected]


mailto:[email protected]

iPathwayGuide User Guide v1 - Advaita Bioinformatics 5 Organism:)Choose)your)organism.)...

Documents

Transcript of iPathwayGuide User Guide v1 - Advaita Bioinformatics 5 Organism:)Choose)your)organism.)...