ENViz - Agilent Technologies
Contents
1. shared shared name interacti PivotTer WikiPat Enrich MHG c mHG n MHG c mHG b Hs_Focal enrichme Hs_Focal enrichme hsa miR Hs Focal 17 113256 13639 2377 182 85 Ci eat eric te ll rice ben eit tie Palle an enenne T E maa ea rn Node Table Edge Table Network Table A This network is much smaller than pathway enrichment network generated without the filtering by permutation test This network has 114 nodes and 142 edges at the display cutoff of 10 compare to 340 nodes and 537 edges in the network obtained without the permutation test filter 3 7 Visualization of generic enrichment data When generic annotation data is used for enrichment analysis ENViz only displays the enrichment network as a bi partite graph with pivot and annotation nodes As before edges connect pivot and annotation nodes corresponding to pivot annotation pairs with annotation enrichment scores better than the Enrichment Display Cutoff value These edges are color coded by direction of gene pivot correlation Red corresponds to the enrichment among genes positively correlated to the pivot and blue corresponds to the enrichment among genes negatively correlated to the pivot Thickness of the edge is proportional to the enrichment score Page 29 of 40 Tutorial ENViz fhe Agilent Technologies 4 ENViz Operational Notes For details on recommended computer configuration ENV2. Enerly_PathwayAnalysisRest340 0 537 1 sta Cell Cycle hsa miR 301b Cell Cycle 190 0 77 0 sta Cell Cycle hsa miR 130b Cell Cycle 190 0 77 0 Table Panel pe D gg OO 5 wW Eo f z Enely_PathwayAnalysisResuts 04112014 enrichments Homo sapiens Hs_Cell_c enrichme Hs_Cell_c enrichments positive hsa miR 130b Hs Cell cycle 31 02638 13639 331 89 35 Node Table Edge Table Network Table Memory OK Here is a slightly larger view of the Cell cycle pathway with genes color coded by correlations to hsa miR 130b Ubiquitin Mediated Proteolysis S Displaying a pathway colored by several connecting edges Double clicking on a pathway node displays the pathway multiple times up to a configurable limit with each pathway instance colored by the data associated with a different coloring edge connected to this pathway The coloring edge is used to color the pathway through the use of the correlation to the pivot node connected to Page 22 of 40 Tutorial ENViz i Agilent Technologies that co oring edge If the number of edges connected to the pathway node is bigger than the limit pathway instances are generated for the edges with the highest enrichment scores Network views are tiled in a small multiples view that accentuates contrasts between correlations for different pivot data In the Figure below 11 different views of cell
3. d Generate pathway annotations file Generate GO annotations file gt Instructions gt Ir Interactive Legend v a Analyze 4 Visualize Page 17 of 40 Tutorial ENViz Agilent Technologies To generate a GO annotations file right click on the Annotations box or click the little triangle next to the annotations box and select Generate GO annotations file and then select one of the options Use biological_process GO terms Use molecular_function GO terms or Use cellular_component GO terms Control Panel Ox fe Network Style Select 257 Agilent ENViz Instructions gt Interactive Legend T Generate pathway annotations file Generate GO annotations file b Use biological_process GO Terms Use molecular_function GO Terms Use cellular_component GO Terms 4 Analyze 4 Visualize When GO annotation is generated for the first time gene_info gz gene2go gz and gene_ontology_edit obo are downloaded from ftp ftp ncbi nlm nih gov gene DATA _ website and copied to your ENViz app s data directory see the Cy2 User Readme or the Cy3 User Readme documents for details These files are used to generate GO annotation matrices and visualize the GO hierarchy These files are updated monthly 2 6 Good analysis practices 1 Make sure that samples are ordered in the same way in primary and pivot data files ENViz checks that samp
4. Analyze Enrichment Display Cutoff visualize Node Table Edge Table Network Table Page 11 of 40 Tutorial ENViz aS Agilent Technologies The Control Panel contains Instructions an Interactive Legend Analysis and Visualization controls Interactive legend provides a graphical overview of the workflow Under the Interactive Legend subpanel you can click on the labeled boxes and be prompted for the appropriate files You can also drag and drop a file reference onto a labeled box Control Panel Ox fe Network Style Select 77 Agilent ENViz Instructions gt Interactive Legend v Primary Annotations Data Pivot Data 4 Analyze Analysis Results Analysis Results Correlation Type Pearson Organism Homo sapiens Filter by Permutation Test F Visualization Y Analysis Results Enrichment Display Cutoff Separate sub panels can be collapsed or expanded by clicking on their handles This is based upon collapsible subpanels Bader Lab University of Toronto 2 2 Setting up input and results files To load primary data click on the box labeled Primary Data on the interactive legend Navigate to the folder with your input data and select your primary data file For the sample dataset select Enerly_GeneExpression tab and click Import or simply double click on Enerly_GeneExpression tab
5. enrichme Hs_Cell_c enrichments positive hsa miR 18a Hs Cell cyde 28 221792 13639 214 89 Hs_Cell_c enrichme Hs_Cell_c enrichments positive hsa miR 106a Hs Cell cycle 26 669687 13639 565 89 Hs_Cell_c enrichme Hs_Cell_c enrichments positive hsa miR 30 1a Hs Cell cycle 26 476448 13639 528 89 7 Hs_Cel r enrichme Hs_Cel F3 enrichments positive hsa miR 17 Hs Cell cyde 26 376232 13639 824 89 E 1 Hs_Cell_c enrichme Hs_Cell_c enrichments negative hsa miR 493 Hs Cell cycle 26 153448 13639 380 89 Hs_Cell_c enrichme Hs_Cell_c enrichments negative hsa miR 133b Hs Cell cyde 25 330515 13639 186 89 E Hs_Cell_c enrichme Hs_Cell_c enrichments positive hsa miR 93 Hs Cell cycle 25 177492 13639 620 89 Hs_Cell_c enrichme Hs_Cell_c enrichments positive hsa miR 20b Hs Cell cycle 25 145546 13639 941 89 Hs_Cell_c enrichme Hs_Cell_c enrichments positive hsa miR 20a Hs Cell cycle 24 700878 13639 848 89 _ 1 m Node Table Edge Table Network Table For each visualized enrichment network False Discovery rate FDR rate is calculated and reported in the Network Table FDR is calculated as the following ratio expected number of significant edges observed number of significant edges Where expected number of significant edges is num annotation nodes A x num pivots M x 10 Enrichment
6. Display Cutoff And observed number of significant edges is the number of edges in the enrichment network with enrichment scores above Enrichment Display Cutoff 3 6 Visualization of saved results Any analysis results archive file generated by ENViz can be visualized as an enrichment network independent of running enrichment analysis To visualize a previously generated analysis results file select it using the Visualization subpanel and then click Visualize The figure below shows the visualization of our sample pathway analysis results when filtering by permutation test sample file Enerly_PathwayAnalysisResults WithSimulations 1K mhg Page 28 of 40 Tutorial ENViz o e Agilent Technologies 2 0 e g Session New Session Lo E File Edit View Select Layout Apps Tools Help 4 gt f amp bi P mm ot Ot a Ga BQ QAO S ters He hsasmiR 1300 Control Panel o x 4 Enerly_PathwayAnalysisResults WithSimulations 1K enrichments Homo sapiens ee fers fg Network Style Select 57 Agilent ENViz Network si Enerly_PathwayAnalysisResults 04112014 enrichments i Enerly_PathwayAnalysisResults 04112014 enrichmer cs Enerly_PathwayAnalysisResults WithSimulations 1K enric Enerly_PathwayAnalysisResults With Simulations 1K Table Panel Ox gg OO 0 f z Enerly_PathwayAnalysisResults WithSimulations 1K enrichments Homo sapi w JI
7. GO term cumulative enrichment value Cumulative enrichment score for each GO term is the sum of enrichment scores for all edges connected to the GO term node e Nodes for included parents of these GO terms even if they don t have edges connecting them to a pivot Unconnected parents are colored gray e Pivot nodes miRNAs colored gray e Edges between GO and miRNA nodes corresponding to pivot annotation pairs with annotation enrichment scores better than the Enrichment Display Cutoff value These edges are color coded by direction of gene pivot correlation Red corresponds to the enrichment among genes positively correlated to the pivot miRNA and blue corresponds to the enrichment among genes negatively correlated to the pivot miRNA Thickness of the edge is proportional to the enrichment score e Edges between GO terms represent parent child relationship between GO terms The right panel shows a GO summary network part of GO DAG for GO terms from the left enrichment network 3 Each GO node is color coded by the cumulative enrichment score for its set of pivot miRNA nodes Parent terms are added to complete the GO hierarchy Parent GO terms that do not have individual miRNA GO term edges above the enrichment score cutoff are colored gray Page 24 of 40 Tutorial ENViz fhe Agilent Technologies j Enerly_ GOAnalysisResults 1K GO enrichments amp amp Enerly_GOAnalysisResults 1K GO Hierarchy Lota fae Di
8. Pearson Organism Homo sapiens Filter by Permutation Test Analyze Reset Visualization Y Analysis Results Enerly_PathwayAnalysisResults mh mhg Enrichment Display Cutoff i Page 15 of 40 Tutorial ENViz lt 3 Agilent Technologies The analysis results file is a zipped archive containing all the analysis data files needed for visualizing your enrichment results as an enrichment network 2 3 Running analysis Once the input and results files are set up the Analyze button becomes enabled Select Correlation Type Pearson or Spearman Organism Homo Sapiens for sample dataset Filter by Permutation Test unchecked for this example and click the Analyze button to run the analysis You will see a task monitor dialog that displays the status of the analysis as it executes along with the total and remaining time estimates Sa Cytoscape Performing Enrichment Analysis Performing Enrichment Analysis x i Calculating enrichment for pivot 109 489 Total time 0 hours 0 minutes 35 seconds Time left Oh When the analysis is completed the analysis results file name becomes black meaning the results file now exists and you can now visualize the results as a Cytoscape network as described in Section 3 To run an analysis using GO annotations perform the same operations above but select Enerly_GO_Annotation tab gz as the annotation file and change the name of the analysis re
9. based miRNA profiles of 489 microRNAs measured for the same 100 samples The sample dataset includes pathway annotation files generated using WikiPathways and GO annotation file Primary data file Ener y_GeneExpression tab a tab delimited file containing gene expression data for 13 639 genes and 100 samples Gene expression data is taken from gProcessedSignal Agilent Feature Extraction files The expression levels were log2 transformed and quantile normalized 5 Pivot data file Ener y_miRNAExpression tab a tab delimited file containing data for 489 miRNAs and 100 samples miRNA expression data is log2 transformed and normalized to the 75 percentile 5 Annotation files GO annotation file Enerly_GO_Annotation tab and WikiPathway annotation file Enerly_WikiPathway_Annotation tab The GO annotation file contains gene annotations for 11 177 GO terms GO annotation is based on GO database release downloaded on 08 22 2013 The pathway annotation file contains gene annotations for 204 pathways In all examples and descriptions below we will refer to primary data as gene expression data from the sample dataset pivot data as miRNA expression data from pivot dataset and annotation data as either GO annotation or pathway annotation from the sample dataset Page 9 of 40 Tutorial ENViz Agg Agilent Technologies 2 ENViz Analysis 2 1 Starting ENViz Launch Cytoscape You should see a window that looks something like ai ess
10. correlated to the pivot miRNA and blue corresponds to the enrichment among genes negatively correlated to the pivot miRNA Thickness of the edge is proportional to the enrichment score Page 19 of 40 Tutorial ENViz Agilent Technologies The network below is generated for the enrichment score cutoff of 10 and contains 340 nodes miRNAs and pathways and 537 edges Session New Session oll es File Edit View Select Layout Apps Tools Help to ee we e GE ae Ge aA ag ich Lo EY Control Panel ox Enerly_ PathwayAn ale Tg Network Style Select 57 Agilent ENviz Instructions gt __ Be _ o e i p a mm 2NA EnerlyData SetiEnerly_Gene _ JA EnerlyData Set Enerly_miRNAE AEE oF ae SS G4 1 Loe yData Set Enerly_WikiPathwayAn se ty WUE N Wek Analysis Results nerly_PathwayAnal sisResults f H l l MARSS y E y yAnaly es sy fel ate SS Correlation Type Pearson aa jae TAN Organism Homo sapiens X Filter by Permutation Test Table Panel Ox Anal Reset ai v a oO zg oO D wy x f z Enerly_PathwayAnalysisResults 04112014 enrichments Homo sapiens X SUID selected name Analysis Results Results Enerly_PathwayAnalysisResults lt 62 false Hs_Cyto Hs_Cytoplasmic_Ribosomal_Proteins_WP477 Hs Cytopl
11. cycle 2 pathway are shown genes in each view are colored by the correlation to the corresponding miRNA The title of each corresponding panel has the name of the pathway and the name of the miRNA f S eN i File Edit View Select Layout Apps Tools Help w ot GE ae a Hh 9 QQ QA GS as He ere le Control Panel Ox l Tg Network style Select 5 57 Agilent ENvia Network No Ed sa Enerly_PathwayAnalysisResult a Enerly_PathwayAnalysisR 340 1 537 0 lt a DNA Replication hsa miR 130b DNA Replicz203 0 83 0 sta G1 to S cell cycle control hsa miR 130b G1 to S cell 161 0 340 lt a Cell Cycle hsa miR 130b Cell Cycle 190 0 77 0 cea DNA Replication hsa miR 130b DNA Replicz203 0 83 0 ta G1 to S cell cycle control hsa miR 130b G1 to S cell 161 0 34 0 X X ta Cell Cycle hsa miR 130b Cell Cycle 2 190 0 77 0 a Cell Cycle hsa miR 301b Cell Cycle 190 0 77 0 ct Cell Cycle hsa miR 130b Cell Cycle 3 190 0 77 0 gt Cell Cycle hsa miR 18b Cell Cycle 190 0 77 0 a Cell Cycle hsa miR 18a Cell Cycle 190 0 77 0 i Cell Cycle hsa miR 106a Cell Cycle 190 0 77 0 gt Cell Cycle hsa miR 301a Cell Cycle 190 0 77 0 sta Cell Cycle X a oD gg oO D WwW 0 f z Enerly_PathwayAnalysisResults 04112014 enrichments Homo sapiens X Table Panel Ox Node Table Edge Table Network Table Se E Displaying multiple pathw
12. measurement results No such tools provide the visualization strength of Cytoscape In ENViz we aimed to provide a tool that enables joint analysis and harnesses Cytoscape to support and provide a visualization interface ENViz approach to integrated data analysis uses the power of enrichment statistics and knowledge of genomic annotation databases to assign relevant function annotations to profiled elements and to gain better understating of the relationship between different molecular levels in the cells The basic idea is to first rank the informative elements according to a particular aspect of the non informative dataset and then search for enrichment of interesting annotations in the ranked list of elements The enrichment results are therefore directly assigned to the un annotated profiled elements Even though development of ENViz was motivated by available modern biological measurements joint analysis of two sample matched datasets and systematic annotations may be applied to other measurement that fit in the setup described below Page 4 of 40 Tutorial ENViz fee Agilent Technologies ENViz follows an enrichment analysis approach driven by three input matrices samples Pathways GO other Primary Annotation genes Data samples Jb Enrichments miRNAs other Correlations ENViz input consists of primary data matrix e g a set of genes with expression measured across a set of samples pivot data mat
13. row for each primary data element e g gene and one column for each sample First column should have primary data element identifier If primary data contains gene based measurements first column should contain a gene identifier which should be an EntrezID or a GeneSymbol The first row should have sample identifiers Other entries in this file have measurements for each primary data element in each sample Missing values Should be denoted by In the sample dataset primary data file is Enerly_GeneExpression tab Pivot data file tab delimited file with one row for each pivot element and one column for each sample The first column should have pivot identifiers The first row Should have sample identifiers Samples should be ordered in the same way in primary and pivot data files and use the same identifiers i e the first row in primary data file and the first row in the pivot data file should be the same Other entries in this file have measurements for each pivot element in each sample Missing values Should be denoted by In the sample dataset pivot data file is Enerly_miRNAExpression tab Note that ENViz is not performing any pre processing or filtering of the input data files You should apply the normalization filtering processes best suited for your data Primary and pivot data should not have any duplicate entries Annotation file optional tab delimited file containing systematic annotations of primary data elements t
14. sa miRA a MIRINDAN DLA a EE a ee e mann A a miR hsa miR 381 o ag Ma insa miR 376a 3763 P thsa miR 377 Mr A a p hsa miR 1 t isa miR 299 5p 369 5p J 499b 5p TT jm i sa miR 337 3p i miR 154 il aa s Table Panel Ox a D zg 00 B w Eo f z Enerly_PathwayAnalysisResults 04112014 enrichments Homo sapiens v SUID selected shared name Label Type Cumen Normali 62 false Hs_Cyto Hs_Cytoplasmic_Ribosomal_Proteins_WP477 Hs Cytoplasmic Riboso Pathway 2174 576 1 0 false hsa miR hsa miR 605 hsa miR 605 BioEntity false hsa miR hsa miR 602 hsa miR 602 BioEntity false Hs_Cell_c Hs_Cell_cycle_WP179 Hs Cell cycle Pathway 2513 288 V 1 0 false hsa miR hsa miR 575 hsa miR 575 BioEntity false hsa miR hsa miR 145 hsa miR 145 BioEntity false Hs_Focal Hs_Focal_Adhesion_WP306 Hs Focal Adhesion Pathway 1544 910 0 999941 AIA a i m fii Node Table Edge Table Network Table Visually overlaying enrichment information on pathways For analysis of enrichment in biological pathways correlations of primary data and corresponding pivot data are visually overlaid on biological pathways for each significant pivot annotation pair using WikiPathways 4 You can overlay enrichment information on pathways
15. BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM DAMAGES OR OTHER LIABILITY WHETHER IN AN ACTION OF CONTRACT TORT OR OTHERWISE ARISING FROM OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE We are very grateful to Robert Kincaid and Melanie Tory for multiple suggestions on the software and UI improvements We thank Josh Spin and Sandra Nyberg for valuable user feedback Also many thanks to Alexander Pico for help with Cytoscape plugin and WikiPathways issues Page 34 of 40 Tutorial ENViz Agilent Technologies A Appendix Overview of ENViz Enrichment Analysis Consider a primary data matrix with N rows and S columns In the sample data set this matrix has gene expression data primary data genes samples Next consider pivot data matrix with M rows and S columns In the sample data set this matrix has miRNA expression data primary data genes samples pivot data miRNAs samples Page 35 of 40 Tutorial ENViz fhe Agilent Technologies Samples in the primary and pivot data matrices should be ordered in the same way i e sample s in primary data is the same as sample s in pivot data primary data genes samples pivot data miRNAs samples Now consider the first row in the pivot matrix first pivot Page 36 of 40 T
16. Cell cycle pathway node and perform Se ect gt Edges gt Select Adjacent Edges 3 Now select all the nodes connected to your selected edges by choosing Select gt Nodes gt Nodes Connected by Selected Edges Your enrichment network will look similar to Page 30 of 40 Tutorial ENViz k Agilent Technologies l Enerly pathway enrichments Homo sapiens Wi ag AN yi d Foy ET P 4 Now create a new sub network by performing File gt New gt Network gt From Selected Nodes Selected Edges You should now see a new sub network like L Enerly pathway enrichments Homo oe h LET haar Al itp Y har Fa i IRA p Se eee Hn R E haan ARAN REF T hara iii han Aa R hjn AT This new sub network should behave just like the original enrichment network for ENViz operations ENViz uses a heuristic to determine the originating enrichment network so that it can associate the appropriate ENViz state with the sub network This heuristic can fail when there isn t a clear originating network as in creating a network from ten different enrichment networks When there is more than one possible originating enrichment network ENViz currently arbitrarily chooses one a future release of ENViz will allow you to choose the originating network When an originating enrichment network is not found the sub network will not behave as an ENViz network Page 31 of 40 Tutorial
17. E ABOVE EXCLUSION MAY NOT APPLY TO YOU YOU MAY HAVE OTHER RIGHTS THAT VARY ACCORDING TO LOCAL LAW LIMITATION OF LIABILITY TO THE EXTENT ALLOWED BY LOCAL LAW IN NO EVENT WILL AGILENT OR ITS SUBSIDIARIES AFFILIATES OR SUPPLIERS BE LIABLE FOR DIRECT SPECIAL INCIDENTAL CONSEQUENTIAL OR OTHER DAMAGES INCLUDING LOST PROFIT LOST DATA OR DOWNTIME COSTS ARISING OUT OF THE USE INABILITY TO USE OR THE RESULTS OF USE OF THE SOFTWARE WHETHER BASED IN WARRANTY CONTRACT TORT OR OTHER LEGAL THEORY AND WHETHER OR NOT ADVISED OF THE POSSIBILITY OF SUCH DAMAGES YOUR USE OF THE SOFTWARE IS ENTIRELY AT YOUR OWN RISK SOME JURISDICTIONS DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR DAMAGES SO THE ABOVE LIMITATION MAY NOT APPLY TO YOU Applicable Law Disputes arising in connection with this Agreement will be governed by the laws of the United States and of the State of New York without regard to choice of law provisions The United Nations Convention for Contracts for the International Sale of Goods will not apply to this Agreement Unenforceability To the extent that any provision of this Agreement is determined to be illegal or unenforceable the remainder of this Agreement will remain in full force and effect Entire Agreement This Agreement constitutes the entire agreement between you and Agilent and supersedes any previous communications representations or agreements between the parties whether oral or written regarding transact
18. ENViz 2 Agilent Technologies 4 3 Saving and restoring sessions Cytoscape networks created by ENViz can be saved and restored in Cytoscape sessions however the functionality of these networks is limited to displaying the networks at they appeared when saved and to having the correct attributes associated with the nodes and edges in each of these networks All ENViz specific operations like double clicking on nodes and edges are not Supported 5 Software license AGILENT TECHNOLOGIES INC SOFTWARE LICENSE AGREEMENT ATTENTION DOWNLOADING COPYING PUBLICLY DISTRIBUTING OR USING THIS SOFTWARE IS SUBJECT TO THE AGREEMENT SET FORTH BELOW TO DOWNLOAD STORE INSTALL OR RUN THE SOFTWARE YOU MUST FIRST AGREE TO AGILENT S SOFTWARE LICENSE AGREEMENT BELOW IF YOU HAVE READ UNDERSTAND AND AGREE TO BE BOUND BY THE SOFTWARE LICENSE AGREEMENT BELOW YOU SHOULD CLICK ON THE AGREE BOX AT THE BOTTOM OF THIS PAGE THE SOFTWARE WILL THEN BE DOWNLOADED TO OR INSTALLED ON YOUR COMPUTER IF YOU DO NOT AGREE TO BE BOUND BY THE SOFTWARE LICENSE AGREEMENT BELOW YOU SHOULD CLICK ON THE DO NOT AGREE BOX AT THE BOTTOM OF THIS PAGE AND CANCEL THE DOWNLOAD OR INSTALLATION OF THE SOFTWARE Software Software means one or more computer programs in object code format whether stand alone or bundled with other products and related documentation It does NOT include programs in source code format License Grant Agilent grants you a non exclusi
19. ENViz is a research prototype tool developed by the Computational Biology and Informatics project at Agilent Laboratories the central research organization of Agilent Technologies License ENViz is protected under the license agreement terms in Section 5 Page 6 of 40 Tutorial ENViz 4 2 Agilent Technologies ENViz also includes a number of other open source resources which are detailed in Section 6 References 1 Cline MS et al 2007 Integration of biological networks and gene expression data using Cytoscape Nat Protoc 2 10 2366 2382 2 Eden E Lipson D Yogev S and Yakhini Z 2007 Discovering Motifs in Ranked Lists of DNA sequences PLoS Computational Biology 3 3 e39 3 Eden E Navon R et al 2009 GOrilla A Tool For Discovery And Visualization of Enriched GO Terms in Ranked Gene Lists BMC Bioinformatics 10 48 4 Kelder T et al 2011 WikiPathways building research communities on biological pathways NAR doi 10 1093 nar gkr1074 5 Enerly E et al 2011 miRNA mRNA Integrated Analysis Reveals Roles for miRNAs in Primary Breast Tumors PLoS One Feb 22 6 2 e16915 Page 7 of 40 Tutorial ENViz fhe Agilent Technologies 1 2 Installation For installation details see the Cy2 User Readme or the Cy3 User Readme documents 1 3 Input files ENViz requires two input files the Primary data file and Pivot data file Primary data file tab delimited text file with one
20. Page 12 of 40 Tutorial ENViz fe Agilent Technologies Import Primary Data File Lookin EnerlyDataSet Enerly_GeneExpression tab Eil Enerly_GO_Annotation tab sees Items Z Enerly_miRNAExpression tab Z Enerly_WikiPathwayAnnotation tab Desktop Documents im File name Enerly_GeneExpression tab Files of type All Files Notice the checkmark that now appears on the Primary Data box These checkboxes inform you that the given box has been filled in Also notice that the Primary Data textbox in the Analysis subpanel is filled in with your choice S Session New Session File Edit View Select Layout Apps Tools Help pat eae A k LLLI IEA Control Panel 4 to e egag d ci ers 357 Agilent ENViz Instructions gt Interactive Legend v Pivot Data 4 Analyze Analysis Results RNA EnerlyData SetiEnerly_GeneExpression tab B i5 B Table Panel Ox Pearson _ D se o0 We f z Nonetwork Filter by Permutation Test F Enrichment Display Cutoff Visualize Node Table Edge Table Network Table Similar to Primary Data to load pivot data click on the box labeled Pivot Data and navigate to the pivot data file For the sample dataset double click on Enerly_miRNAExpression tab Page 13 of 40 Tutorial ENViz es Agilent
21. Technologies o e E E KFN Sah Se hg BAH wR PB meaaay Control Panel 357 Agilent ENViz Instructions gt Interactive Legend v f 4 Analyze 4 Visualize 2NA EnerlyData Set Enerly_GeneExpression tab i 1A EnerlyData Set Enerly_miRNAExpression tab les L3 Table Panel ox Pearson S o Dee co D fo No Network v Filter by Permutation Test F Analyze Reset Enrichment Display Cutoff _ Visualize Node Table ork Table Memory OK Click on the Annotations box and navigate to the annotation file Enerly_WikiPathway_Annotation tab gz and click New or Import ession New Session File Edit View Select Layout Apps Tools Help palteaeaia Control Panel Ox 357 Agilent ENViz eaase Instructions gt Interactive Legend v J Analyze Analysis Results INA EnerlyData Set Enerly_GeneExpression tab m 1A EnerlyData Set Enerly_miRNAExpression tab m yData Set Enerly_WikiPathwayAnnotation tab gz m Table Panel Ox Pearson Lane D gg OO B w o f z No Network Filter by Permutation Test Enrichment Display Cutoff Node Tal Memory OK You ve now set up the input for a pathway oriented analysis Page 14 of 40 Tutorial ENViz ee Ag
22. User Tutorial ENViz is software for extracting biological insights from multiple types of measurements via integrated statistical analysis of these measurements and available systematic annotations such as ontology or pathway annotations Visualization of the analyses is provided by an app of the Cytoscape network biology software platform Fa Edt View Select Layout Plugins Help PH AQQA T 8 BA 4 kas Anya Tsalenko Roy Navon Israel Steinfeld Michael L Creech Zohar Yakhini Allan Kuchinsky Agilent Laboratories Technion Israel Institute of Technology Blue Oak Software fe Agilent Technologies In memory Allan Kuchinsky a colleague a mentor a fellow scientist a dedicated friend Allan was the living spirit behind ENViz He passed away not long before we finished working on this project leaving us all in shock and agony Allan identified the potential for weaving a joint data analysis approach into Cytoscape and led us all through executing and completing this task Even though Allan was constantly fighting cancer and its complications he led our team with great enthusiasm to cross countless obstacles to make ENViz a reality Allan also worked with great excitement with collaborators to apply ENViz to scientific studies and to learn how we could improve provide more information and add the correct touch of extra functionality Allan s work and enthusiasm brought the software to its current sta
23. anking by correlation primary data elements genes and compute enrichment of this annotation element in the top k genes based on the hypergeometric statistics We repeat this process for all threshold values k and choose k that optimizes the hypergeometric significance A bound on the corresponding p value is calculated to correct for multiple testing 2 3 and an mHG score lt og10 MHG p value is reported large values of the mHG score represent significant enrichments The correction is valid for every individual pivot element but all enrichment results are not corrected for the number of pivot elements and the number of annotation elements For guidance on this next level of correction please see Section 2 6 below Significant results are represented in Cytoscape as an enrichment network a bipartite graph with nodes corresponding to pivot and annotation elements and edges corresponding to pivot annotation entry pairs with enrichment scores better than the user defined threshold In addition e for analysis of enrichment in biological pathways correlations of primary data and corresponding pivot data are visually overlaid on biological pathways for each Significant pivot annotation pair using the WikiPathways 4 resource e enrichments of GO categories are overlaid on top of the Gene Ontology DAG Edges of the enrichment network representing significant associations may point to functionally relevant mechanisms Development
24. asmic Riboso Pathway 2174 576 1 0 100 0 a Enrichment Display Cutoff 63 false hsa miR hsa miR 605 hsa miR 605 BioEntity 80 0 i 65 false hsa miR hsa miR 602 hsa miR 602 BioEntity 80 0 Cumulative Enrichment 67 false Hs_Cell_c Hs_Cell_cycle_WP179 Hs Cell cycle Pathway 2513 288 1 0 100 0 68 false hsa miR hsa miR 575 hsa miR 575 BioEntity 80 0 70 false hsa miR hsa miR 145 hsa miR 145 BioEntity 85 0 72 false Hs_Focal Hs_Focal_Adhesion_WP306 Hs Focal Adhesion Pathway 1544 910 P 0 999941 100 0 ms m i gt Node Table Edge Table Network Table Memory OK To see specific areas of the network in more detail use the Cytoscape pan and zoom operations By clicking on the Control Panel Network tab you can also pan around the network and see the zoomed in network in the context of the whole network on the bottom left Page 20 of 40 Tutorial ENViz 3 6 Agilent Technologies Session New Session File Edit View Select Layout Apps Tools Help af z P ts w e Gt e aea a ice oe Y Control Panel ox amp Enerly_PathwayAnalysisResults 04112014 enrichments Homo sapiens Tg Network style Select agilentenvi i i Sica Network Nodes Edges ss Enerly_PathwayAnalysisResults 04112014 er Enerly_PathwayAnalysisResults 041120 340 0 537 0 ihsa miR 454 OSS Y hsa miR 539
25. ays colored by one pivot Double clicking on a pivot node displays each of the pathways connected to that pivot up to a configurable limit Each pathway is colored by its correlation with the selected pivot node Page 23 of 40 Tutorial ENViz 3 Agilent Technologies ec e e g Session New Session o e ka a File Edit View Select Layout Apps Tools Help w E ete ae SY QQ aA ahg Sei So pi hsa miR 130b On na Control Panel O x 4 Enerly_PathwayAnalysisResults 04112014 enrichments Homo sapiens SS ie S hsamiR 20 WR 199a dp te Network Style Select 7 Agilent ENViz Network No Edges ssa Enerly_PathwayAnalysisResults C Enerly_PathwayAnalysisRest340 1 537 0 lca DNA Replication hsa miR 1300 DNA Replicatic203 0 83 0 sta G1 to S cell cycle contro hsa miR 130b G1 to S cell cyc161 0 340 st Cell Cycle hsa miR 130b Cell Cycle 190 0 77 0 Table Panel Ox a D gg oO B Ww o f Enerly_PathwayAnalysisResults 04112014 enrichments Homo sapiens v Node Table Edge Table Network Table 3 3 Gene Ontology enrichment and visualization Enrichments of GO categories are overlaid on top of the Gene Ontology DAG in addition to the bi partite graph representing the enrichment network In the Figure below the left panel shows the pi partite GO enrichment network with e GO term nodes colored on a yellow to red scale according to the
26. by double clicking on edges pathways and pivot nodes in the enrichment network Besides selecting the node or edge you clicked on and displaying it in green one of the following pathway visual overlays takes place Displaying a pathway colored by one connecting edge Double clicking on an edge in the enrichment network displays the WikiPathway corresponding to the pathway node connected to the edge In this example the selected pathway is Cell cycle Genes in the selected pathway are colored by correlation to the pivot node at the other end of the selected edge hsa miR 130b In the right part of the figure below all gene nodes in the Cell cycle pathway that map to primary data elements are color coded blue gt red for the correlation score between the primary data element gene and the pivot data element hsa miR 130b Solid borders and high opacity show genes above correlation threshold that were included in the gene set used for enrichment analysis Genes colored red have positive correlations to hsa miR 130b genes colored blue have negative correlations to hsa miR 130b Page 21 of 40 Tutorial ENViz ee ae 2 a Agilent Technologies ie bd ra File Edit View Select Layout Apps Tools Help amp SS t l w og Be He fg nsa miR 130b Control Panel O x 4 Enerly Pathway fg Network style Select 7 Agilent ENviz Network No Edges aa Enerly_PathwayAnalysisResults C
27. hat will be used for enrichment analysis First column should have primary data element identifiers Primary data elements in the annotation file and in the primary data file should be ordered in the same way and use the same identifiers i e the first columns in primary data file and in annotation file should be the same First row should have annotation categories Other entries of this file are Os and is 1 in row column j indicates that primary element from row belongs to annotation category in column j If an annotation file is not provided ENViz can generate one for GO Gene Ontology and for pathway annotations using WikiPathways EntrezIDs or GeneSymbols are used to assign genes to pathway or GO categories In the sample dataset the GO annotation file is Enerly_GO_Annotation tab gz and the pathway annotation file is Enerly_WikiPathway_Annotation tab gz you may notice the different gz extension that specifies the files are compressed because the annotation files can be quite large 1 4 Example dataset Page 8 of 40 Tutorial ENViz i Agilent Technologies For this tutorial we use the dataset published in 5 Data formatted for ENViz can be downloaded from http bioinfo cs technion ac il people zohar ENViz data html This dataset consists of 100 breast tumor samples with various characteristics Primary data is gene expression profiles of 13 639 genes for 100 samples from microarray experiments Pivot data is microarray
28. hsa miR 1300 DNA Replication 203 0 83 0 N K NA hsa miR 483 3p ca G1 to S cell cycle control X b AA AN gt hsa miR 130b G1 to S cell cycle control 161 0 340 D _ cea Cell Cycle hsa miR 34c 5p hsa miR 3 hsa miR 130b Cell Cycle 190 0 77 0 ta DNA Replication hsa miR 130b DNA Replication 2 203 0 83 0 sta G1 to S cell cycle contro S EAF hsa miR 1300 G1 to S cell cycle control 2161 0 340 a A oai sta Cell Cycle hsa miR 130b Cell Cycle 2 190 0 77 0 hsa miR 629 da Cell Cycle 2 sT oe Gnae hsa miR 301b Cell Cycle 190 0 77 0 g E ce Cell Cycle 4 na a hsa miR 130b Cell Cycle 3 190 0 77 0 p A sa miR 5025p sta Cell Cycle re hsa miR 18b Cell Cycle 190 0 77 0 st Cell Cycle hsa miR 18a Cell Cycle 190 0 77 0 sea Cell Cycle hsa miR 106a Cell Cycle 1900 770 Table Panel Ox sta Cell Cycle 7 hsa miR 301a Cell Cycle 190 0 77 0 D gg OO E wW 6 f z Enerly_PathwayAnalysisResults 04112014 enrichments Homo sapiens v saa Cell Cycle v shared shared name interaction PivotTerm WikiPathway A EnrichmentValue MHG c mHG n MHG c mH Hs_Cell_c enrichme Hs_Cell_c m enrichments positive hsa miR 30 1b Hs Cell cyde 31 122416 13639 361 89 Hs_Cell_c enrichme Hs_Cell_c enrichments positive hsa miR 130b Hs Cell cycle 31 02638 13639 331 89 Hs_Cell_c enrichme Hs_Cell_c enrichments positive hsa miR 18b Hs Cell cycle 28 588976 13639 261 89 Hs_Cell_c
29. ichment score S we observe Sana gt S more than Stringency times across all permutations this pivot annotation element pair is considered not significant and its enrichment score is set to 0 Filter by Permutation Test Num Permutations 1 000 Enrichment Score Threshold 15 Stringency oF For pivot annotation pairs that survive this permutation test filtering the original MHG score is reported as the enrichment score 2 5 Generating annotation files To generate a pathway based annotation file based on WikiPathways provide an annotation file name in the Annotations box Remember that you can do this by clicking on the Annotations box in the interactive legend or by clicking the button next to the Annotations field and changing an existing file name or by directly typing the full pathname of file in the Annotations field After selecting a name right click on the Annotations box in the Interactive Legend or alternatively click on the triangle next to the top of the Annotations box Select Generate pathway annotations file This will generate a tab delimited file with one row for each gene and one column for each pathway will be generated Entry in row column j of annotation file will be 1 if gene belongs to pathway j O otherwise If the annotation file name has the extension gz the file will be automatically compressed zipped Control Panel Oox fg Network Style Select 57 Agilent ENviz
30. ilent Technologies To run the analysis specify an Analysis Results file for the results of analysis This is an archive file that will contain the correlations enrichments and thresholds that are produced by the analysis It also serves as input to Visualization Click on the box labeled Analysis Results You will be prompted for a file name Output File Name Analysis Results Save in fm EnerlyDataSet J Recent Items Desktop h Documents i A DDr s Computer SCB2063LRH Ta File name Enerly_PathwayAnalysisResults Network Files of type Analysis Results Archive mhg Enter a new file name of your choice You can do this by choosing an existing file and modifying the name in the File name field or by just typing a new name You don t need to enter the mhg extension it will be added automatically After entering a name click on Save The output file name will be shown in red text on the control panel when the file doesn t currently exist It will also be shown in red in the Analysis Results input box for the Visualization sub panel You can also directly enter or edit the pathname in the Analysis Results input box however the pathname must be a full pathname to the file Analysis Primary Data SetiEnerly_GeneExpression tab Pivot Data efiEnerly_miRNAExpression tab Annotations arly_WikiPathwayAnneotation tab Analysis Results y_PathwayAnalysisResul ts hg m Correlation Type
31. ion New Session File Edit View Select Layout Apps Tools Help a 9 ea t s Table Panel Ox D zg OO S WwW o f z No network Node Table Memory OK Start ENViz by selecting the Apps gt ENViz gt Start ENViz menu item Page 10 of 40 Tutorial ENViz ag i Agilent Technologies ession New essi File Edit View Select Layout Apps Tools Help gt ae App Manager sS or OENViz gt Start ENViz to Enter search tem lt Control Panel Ox Table Panel o D es D W f z No Network Node Table Edge Table Memory OK When ENViz starts its opens a new Agilent ENViz tab and places a control panel in the left subpanel of the Cytoscape window This includes controls for inputting the primary data pivot and annotations buttons for running analysis and visualization and controls for setting thresholds that control the size of the networks to visualize a Session Ne Ses on File Edit View Select Layout Apps Tools Help hi agate amp e oe kan gri Enter search tem QQ OQ Q s Control Panel Ox 357 Agilent ENViz Instructions gt Interactive Legend v Pranay Annotations Data Data J Analyze Analysis Results Ox Table Panel Pearson Homo ian gg OO D w o f z fo neter Filter by Permutation Test F
32. ions hereunder except for a specific warranty issued by Agilent with regard to this product Your additional or different terms and conditions will not apply This Agreement may not be changed except by an amendment signed by an authorized representative of each party 6 Acknowledgments ENViz software uses the following resources WikiPathways wikipathways org index php WikiPathways used according to the following terms of use and license http wikipathways org index php WikiPathways License Terms Gene Ontology http www geneontology or JSON JavaScript Object Notation lightweight data interchange format used according to the following license Copyright c 2002 JSON org Permission is hereby granted free of charge to any person obtaining a copy of this software and associated documentation files the Software to deal in the Software without restriction including without limitation the rights to use copy modify merge publish distribute sublicense and or sell copies of the Software and to Page 33 of 40 Tutorial ENViz ee Agilent Technologies permit persons to whom the Software is furnished to do so subject to the following conditions The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software The Software shall be used for Good not Evil THE SOFTWARE IS PROVIDED AS IS WITHOUT WARRANTY OF ANY KIND EXPRESS OR IMPLIED INCLUDING
33. iz configuration and migrating data to new versions of ENViz see the Cy2 User Readme or the Cy3 User Readme documents 4 1 Destroying and recreating network views ENViz fully supports destroying and creating Cytoscape views of ENViz generated CyNetworks Because ENViz can quickly generate many Cytoscape views you may wish to destroy some network views and then bring them back at a later time To remove views 1 Click on Cytoscape s Control Panel Network tab 2 Select the network views to remove 3 Bring up the context menu on any of the selected networks e g right click on Windows and perform Destroy View To recreate views 1 Select some or all of the networks for which you destroyed views 2 Bring up the context menu on any of these selected networks and perform Create View The networks for which you recreated views should behave normally and support all ENViz operations 4 2 Creating sub networks from ENViz enrichment networks For many cases you can use Cytoscape to create sub networks of ENViz networks and have these sub networks behave like the original enrichment network in that the various ENViz operations will work on edges pivots and annotations in the sub network Consider the example of pulling out one pathway from an enrichment network We will pull out the Hs Cell cycle pathway and place it in a separate CyNetwork 1 Visualize the Enerly pathway data see Visualizing Analysis Results above 2 Select the Hs
34. le and press the Visualize button Analysis Primary Data SetiEnerly_GeneExpression tab Pivot Data etiEnerly_miRNAExpression tab Annotations erly WikiPathwayAnnotation tab Analysis Results y_PathwayAnalysisResults mhg Correlation Type Pearson Organism Homo sapiens ill Filter by Permutation Test F Analyze Reset Visualization Y Analysis Results Set Enerly_PathwayAnalysisResults mhg Enrichment Display Cutoff 1 Visualize Reset 3 1 Enrichment Networks As you may recall ENViz represents significant analysis results as an enrichment network a bipartite graph with nodes corresponding to pivot and annotation elements and edges corresponding to pivot annotation pairs with annotation enrichment scores better than the user defined threshold cutoff value Once your enrichment network is visible in Cytoscape you can pan and zoom the network using standard Cytoscape commands and menu items see the Cytoscape User Manual 3 2 Pathway enrichment visualization In the figure below e Nodes are pathways color coded yellow gt red by cumulative enrichment score and miRNAs color coded grey The cumulative enrichment score for each pathway node is calculated as the sum of enrichment scores for all edges connected to this pathway node e Edges between pathway and miRNA nodes are color coded by direction of gene pivot correlation Red corresponds to the enrichment among genes positively
35. le names in the first rows in primary and pivot data are the same and will generate an error otherwise 2 Since ENViz does not do any pre processing of the input data it is very important that input data does not have any duplicate elements Having duplicate primary data elements will compromise mHG statistical model and having duplicate pivot entries will generate duplicate enrichment network nodes and edges When primary data is gene based and has GeneSymbol as gene identifier check also that different GeneSymbols do not represent the same gene 3 In case of GO annotation remove primary data elements that do not belong to any annotation categories This will improve both the speed of the analysis and the accuracy of the statistics 4 To reduce false positive results it is preferred that the Enrichment Score Threshold T be set to correct for multiple testing For example according to Bonferroni multiple testing correction T og10 po M A where M A are the dimensions of Enrichments Matrix and pg is the desired significance level 5 Running analysis with filtering by permutation test option reduces the level of false positive results Recommended Enrichment Score Threshold for filtering by permutation test is the same 7 as above Page 18 of 40 Tutorial ENViz A Agilent Technologies 3 Visualizing Analysis Results To visualize the results of a completed analysis select your Enrichment Display Cutoff 10 in this examp
36. ment Analysis and Visualization performs joint enrichment analysis of two types of sample matched datasets and available systematic annotations Examples of such data sets may be gene expression miRNA or other non coding RNA expression or proteomics measurements collected in the same set of samples together with pathway gene ontology GO or any custom annotation of the data Enrichment analysis is based on minimum hypergeometric statistics 2 3 Results of enrichment analysis are visualized as an interactive Cytoscape network and could be visually overlaid on biological pathways or GO hierarchy 1 1 Introduction to ENViz Over recent years modern biology has undergone an information revolution which is evident in a shift of thinking and practice While typical biological studies are focused on specific pathways like the p53 signaling pathway the emergence of novel high throughput technologies now enables the quantification of biological features in a genome wide scale The rapid development of technology in particular enabled measurements of MRNA expression levels miRNA expression levels DNA methylation states DNA copy number etc Various methodologies have been developed in recent years to handle integrated analysis of functional genomics data mainly by studying the transcriptional programs and global organization of biological processes Still only a few tools support routine joint analysis of sample cohort with multiple genomic
37. n the bottom 10 are colored yellow Nodes with cumulative enrichment in the top 10 are colored red colors for all other nodes are scaled linearly These color assignments can be manually adjusted using the left and right slider thumbs to get the best visual representation of the network Cumulative enrichment values are used to color pathway enrichment networks GO cumulative enrichment networks and cumulative GO graphs Enrichment score values are used to color selected pivot GO graphs 3 5 Viewing and saving enrichment statistics To see enrichment scores corresponding to individual edges right click on an edge and select Enrichment Statistics A window will pop up showing the enrichment score together with the details of mHG statistics In the example below enrichment of Page 26 of 40 Tutorial ENViz fhe Agilent Technologies pathway Cell cycle for genes positively correlated with hsa miR 130b has mHG score of 31 In more detail out of a total of 13 639 genes in the dataset 89 belong to Cell cycle pathway Of the top 331 genes most correlated to hsa miR 130b 35 belong to Cell Cycle pathway amp Stats for Hs Cell cycle enrichments positive hsa miR 130b 2 Stats for edge between Hs Cell cycle and hsa miR 130b Statistic Value Enrichment value 31 N total primary genes n genes above correlation cutoff B Genes in Hs Cell cycle 89 b Genes above cutoff in Hs Cell cycle 35 The enrichment statis
38. of any violation of this Agreement No Disassembly You may not disassemble or otherwise modify the Software without written authorization from Agilent except as permitted by law Upon request you will provide Agilent with reasonably detailed information regarding any permitted disassembly or modification High Risk Activities The Software is not specifically written designed manufactured or intended for use in the planning construction maintenance or direct operation of a nuclear facility nor for use in on line control or fail safe operation of aircraft navigation control or communication systems weapon systems or direct life support systems Transfer You may transfer the license granted to you here provided that you deliver all copies of the Software to the transferee along with this Agreement The transferee must accept this Agreement as a condition to any transfer Your license to use the Software will terminate upon transfer Termination Agilent may terminate this license upon notice for breach of this Agreement Upon termination you must immediately destroy all copies of the Software Note that you may see some error messages when restoring your session about the CytoscapeSessionReader not being able to read files Null PointerException You can ignore these messages the networks referenced seem to be restored correctly Page 32 of 40 Tutorial ENViz ohh Agilent Technologies Export Requirements If you export re export o
39. r import Software technology or technical data licensed hereunder you assume responsibility for complying with applicable laws and regulations and for obtaining required export and import authorizations Agilent may terminate this license immediately if you are in violation of any applicable laws or regulations U S Government Restricted Rights Software and technical data rights granted to the federal government include only those rights customarily provided to end user customers Agilent provides this customary commercial license in Software and technical data pursuant to FAR 12 211 Technical Data and 12 212 Computer Software and for the Department of Defense DFARS 252 227 7015 Technical Data Commercial Items and DFARS 227 7202 3 Rights in Commercial Computer Software or Computer Software Documentation NO WARRANTY TO THE EXTENT ALLOWED BY LOCAL LAW AND EXCEPT TO THE EXTENT AGILENT HAS PROVIDED A SPECIFIC WRITTEN WARRANTY APPLICABLE TO THIS PRODUCT THIS SOFTWARE IS PROVIDED TO YOU AS IS WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND WHETHER ORAL OR WRITTEN EXPRESS OR IMPLIED AGILENT SPECIFICALLY DISCLAIMS ANY IMPLIED WARRANTIES OR CONDITIONS OF MERCHANTABILITY SATISFACTORY QUALITY NON INFRINGEMENT AND FITNESS FOR A PARTICULAR PURPOSE SHOULD THE SOFTWARE PROVE DEFECTIVE YOU ASSUME THE ENTIRE RISK AND COST RESULTING FROM OR RELATING TO THE DEFECT SOME JURISDICTIONS DO NOT ALLOW EXCLUSIONS OF IMPLIED WARRANTIES OR CONDITIONS SO TH
40. rix e g a set of miRNAs with expression measured in the same set of samples and primary data annotation matrix e g pathway or GO annotation of genes Note that in the figure above the boxes representing the various input matrices are drawn to their relative dimensions Consider the case where N number of biological entities such as genes in the Primary Data Matrix M number of biological entities such as miRNAs in the Pivot Data Matrix A number of annotation entries such as pathways or GO terms in the Annotation Matrix Page 5 of 40 Tutorial ENViz 22 Agilent Technologies S number of Samples Then N S dimensions of Primary Data Matrix M S dimensions of Pivot Data Matrix N A dimensions of Annotation Matrix M A dimensions of Enrichments Matrix N M dimensions of Correlations Matrix ENViz analyses a primary data set with respect to a pivot data set and primary data annotation in the following way For each pivot entry e we compute the correlation of pivot data to each element of primary data across all samples e we rank elements of the primary data based on this correlation or anti correlation and e compute the statistical enrichment of annotation elements gene sets in the top of this ranked list based on an MHG minimum hypergeometric 2 3 statistics Details of mHG statistics are explained in 2 3 5 Briefly for each annotation element we do the following We consider top k r
41. s with at least 2 fold change the mHG model seeks enrichment in the top of the ranked list without predefining the top of the list The process results in an enrichment score calculated for each pivot datum and each annotation element e g pathway or GO term The enrichment score is og 9 MHG p value Note that mHG p values are corrected for multiple testing done in MHG calculation but not for number of pivots testes or number of annotation terms in the annotations matrix This enrichment analysis process is also done for anti correlation where the primary data are arranged according to anti correlation with the pivot datum being analysed We then repeat the same process using the second row as the point of focus Page 39 of 40 Tutorial ENViz A Agilent Technologies pivot values primary data genes samples pivot data miRNAs samples At the end of the analysis we generate two enrichment matrices of size MxA one based on positive correlations and one based negative correlations These matrices are used for visualization of analysis results described in Section 3 Page 40 of 40 Tutorial ENViz
42. splaying GO Summary network colored by one pivot Double clicking on a pivot node gray in the enrichment network left shows the GO summary network right panel colored based on the enrichment scores for the selected pivot This GO summary network includes only GO terms enriched in the genes correlated anti correlated to the selected miRNA pivot and their parents JA fo a 3 Si mm ee a ndi hsa miR 16 2 hh O Ob Displaying multiple GO Summary networks colored by multiple pivots Double clicking on a GO node displays multiple different GO Summary networks one per connected pivot up to a configurable limit Each network is populated by the GO terms that have significant enrichment values for the corresponding pivot datum and their parents Page 25 of 40 Tutorial ENViz gh Agilent Technologies 4 Enerly_GOAnalysisResults 1 nese l hsa miR 146b 5p immune 3 hsa miR 146a immune syst 3 L hsa miR 150 immune syste o 2 afk 3 4 Color controls The coloring of annotation nodes in each enrichment network can be adjusted by the color slider Visualization Y Analysis Results RNA_mRNA EnerlyData SetiEnerly_PathwayAnalysisResults mhg Enrichment Display Cutoff Cumulative Enrichment T mia dri When analysis results are visualized for the first time the coloring scheme is defined in the following way Nodes with cumulative enrichment scores i
43. sults archive and click Analyze Note that since GO annotation has more than 11K terms enrichment analysis takes much longer than for pathways For your own data if you don t have an annotation file ready see Section 2 5 on how to generate annotation files using ENViz 2 4 Filtering analysis results by permutation test When the option Fi ter by Permutation Test is selected the following filtering of analysis results based on randomized shuffled data is performed in addition to enrichment analysis To run this analysis you specify e Num Permutations number of permutations to run default is 1000 e Enrichment Score Threshold default is 15 For GO analysis 15 is the recommended threshold for the sample dataset for pathway analysis the recommended threshold is 6 See Section Good analysis practices for guidelines on how to set this threshold for other datasets e Stringency the number of random scores above observed mHG score that still allow passing the filter default is 0 Page 16 of 40 Tutorial ENViz s Agilent Technologies For each permutation samples in the pivot data are randomly shuffled This means that the columns of the matrix are permuted For each pivot with at least one enrichment score above Enrichment Score Threshold the correlations between unchanged primary data and randomized pivot data and the corresponding enrichment scores Sang are calculated If for a given pivot annotation element pair with enr
44. te ready for release We continue to learn from Allan his work is still here for us to follow and trust Above all Allan was a dedicated friend We will always remember Allan s human warmth and his appreciation to the team s work We dedicate ENViz to Allan s memory as a scientist and as a human being Always with a smile and always excited to discuss new science and how software can make it better Page 2 of 40 Tutorial ENViz fe Agilent Technologies ENViz User Tutorial a Welcome to ENViz 1 1 Introduction to ENViz 1 2 Installation 1 3 Input files 1 4 Example dataset NViz Analysis 1 Starting ENViz 2 Setting up input and results files 3 Running analysis 4 Filtering analysis results by permutation test 5 Generating annotation files 6 E 2 2 2 2 2 2 Good analysis practices Visualizing Analysis Results 3 1 Enrichment networks 3 2 Pathway enrichment visualization 3 3 Gene Ontology enrichment visualization 3 4 Viewing and saving enrichment statistics 3 5 Color controls 3 6 Visualization of saved results 3 7 Visualization of generic enrichment data 4 ENViz Operational Notes 4 1 Destroying and recreating network views 4 2 Creating sub networks from ENViz enrichment networks 4 3 Saving and Restoring Sessions 5 Software license 6 Acknowledgments A Appendix Overview of ENViz Enrichment Analysis Page 3 of 40 Tutorial ENViz fee Agilent Technologies 1 Welcome to ENViz ENViz Enrich
45. tics are stored as attributes associated with edges So to see the enrichment statistics for many edges you can use Cytoscape s Table Panel to view and export the edge attributes of an enrichment network The attribute names of the enrichment statistics are prefaced with mHG and the capital names are preferenced with cap For example the B statistic is represented by the attribute NHG capB Here s the steps to view and export an enrichment network s statistics 1 Select edges of interest in your enrichment network by dragging the mouse over them or using other Cytoscape operations for selecting edges e g Select gt Edges 2 Go to the Edge Table and click on the Select All Attributes button the button with checkmarks 3 To export enrichment statistics for all edges in the network use File gt Export gt Table from the Cytoscape menu Page 27 of 40 Tutorial ENViz i Agilent Technologies Session New Session ka a e File Edit View Select Layout Apps Tools Help 9 e i Evi Sods Gog e hsa miR 130b Control Panel Ox iF Enerly_PathwayAnalysisResults 04112014 enrichments n s Homo sapiens fg Network style Select 3 57 Agilent ENviz Network Nodes Edges ca Enerly_PathwayAnalysisResults 04112014 Enerly_PathwayAnalysisResults 04112 340 63 537 274 ssa DNA Replication i 4 y oe
46. utorial ENViz fee Agilent Technologies primary data genes samples So oS ox 2 E samples In the Figure below values for the first pivot are plotted in the top panel samples samples pivot values primary data genes pivot data miRNAs A correlation value is calculated for every row gene in the primary data with respect to the first pivot Correlation could be Pearson or Spearman Correlation values to the first pivot are plotted on the right hand side of the pivot data matrix heatmap Page 37 of 40 Tutorial ENViz fe Agilent Technologies pivot values primary data genes samples correlation pivot data miRNAs samples Next genes rows are ranked according to the correlation with pivot In the Figure below the samples are reordered by monotonically increasing pivot values to produce more clear visualization of the analysis Page 38 of 40 Tutorial ENViz fe Agilent Technologies pivot values primary data genes samples correlation pivot data miRNAs samples We then compute the statistical enrichment of primary data annotation elements gene sets in the top of this ranked list based on MHG minimum hypergeometric Statistics Details of mMHG statistics are explained in 2 3 5 and briefly in section 1 1 above Note that whereas standard statistical enrichment tools set arbitrary thresholds such as top 100 genes top 10 of the genes or gene
47. ve license to download one copy of the Software and to store or run that copy of the Software for internal use and purposes in accordance with this Agreement and the documentation provided with the Software Such documentation may include license terms provided by Agilent s third party suppliers which will apply to the use of the Software and take precedence over these license terms In the absence of documentation specifying the applicable license you may store or run one copy of the Software on one machine or instrument If the Software is otherwise licensed for concurrent or network use you may not allow more than the maximum number of authorized users to access and use the Software concurrently License Restrictions You may make copies or adaptations of the Software for archival purposes or when copying or adaptation is an essential step in the authorized use of the Software but for no other purpose You must reproduce all copyright notices in the original Software on all permitted copies or adaptations You may not copy the Software onto any public or distributed network Upgrades This license does not entitle you to receive upgrades updates or technical support Ownership The Software is owned and copyrighted by Agilent or its third party suppliers Agilent and its third party suppliers retain all right title and interest in the Software Agilent and its third party suppliers may protect their respective rights in the Software in the event
Download Pdf Manuals
Related Search