User Manual
Contents
1. With any of the words Match Exactly V Auto Filter x a Protein Genes Organism ID APAF1 317 Human 124814 ATPAFL 64756 Human 4901988 IPYPAF1 Human 3612408 w Select Select All a Unselect Add Remove Columns Protein Genes Organism ID Effect of clicking Unfocus nh Effect of clicking Focus The view panel also contains a checkbox toggle to enable the gene view This view is always shown simultaneously with the default protein view and hence doubles the number of networks currently visible Human PROTEIN VIEW co amp amp Le Human GENE VIEW ole e Yeast PROTEIN VIEW l Yeast GENE VIEW Toggling the gene view arranges the protein networks on the left and the gene networks on the right If two organisms were selected for visualization the top two networks will correspond to the first organism and the bottom two networks to the second organism In the gene networks each node corresponds to a gene and a link is drawn between two genes if any of their corresponding protein products interact In cases where genes code for multiple isoforms the gene view can be especially convenient for organizing the display All views remain synchronized by orthology and selecting a node in any network also selects all the corresponding genes proteins and orthologs in the other networks 3 3 Visual Options Legend Panel Visualiz2. Information Selection feat Node Edge First Neighbours F Human Yeast Data Panel w Domain Architecture Visualization V Show pFamA V Show pFamB Loading data for organisms Human and Yeast from directory C Users jim OrthoNetsData iRef_7 0 Select Attributes i n J Node Attribute Browser Edge Attribute Browser lk Network Attribute Browser Domain Visualization Welcome to Cytoscape 2 7 0 Right click drag to ZOOM Middle click drag to PAN By default the genes of interest are all those genes known or predicted to be involved in the chromatin modification biological process 1 2 For some organisms e g human and yeast as shown above this results in large and unwieldy networks and so the next step is to focus or center on a smaller subset of genes or proteins via the protein search selection dialog This search dialog is automatically activated immediately after loading networks and can be activated later via the Select Genes Proteins to Focus menu or by clicking the Focus on Protein button in the top left corner of the view panel see also Section 3 2 select Nodes Manage Plugins iss lB Update Plugins 2 TEIN VIEW lf you want to focus on a particular group of genes proteins and their interaction partners please select them here Otherwise click Finish Advanced Network Merge Sp
3. 15 5 Advanced Options 5 1 Advanced Customization Internally the primary node entity is a protein which is uniquely identified by a numeric iRef identifier ROG Whenever the plugin accepts alternative identifiers such as Entrez Gene identifiers it first automatically maps them to ROGs All node and edge attributes are assigned to ROGs or pairs of ROGs The first time the plugin is run it downloads all data to a ppiPluginData subdirectory under the users home directory This creates the following files and subdirectories iRef_x x attributes na iRef_x x attributes ea iRef_x x organisms Fly Yeast Human Worm Mouse sif iRef_x x orthology orthologs omf iRef_x x pda domain_arch pda These files are covered below iRef_x x attributes na These are the node attribute files containing gene protein aliases and other information extracted from iRef By default the plugin loads all node attribute files found in this directory Thus to add customized node attributes create a formatted node attribute file and copy it to this location Consider the example node attribute file REFSEQ na REFSEQ class java lang String 16979 7 SL56005 722757362 2NP_OO10sS 7162 NP_ O00033 1G NP OSS 714 NP 0007 14 3 52564241 66 7955953 2NP 03293062 NP U32066472 3068764 NP_035707 This is the standard file format for node attributes as covered in the Cytoscape manual http www cytoscape org documentation users html T
4. The Version drop down menu lists the iRef database version that will be loaded Immediately following this drop down menu are a series of 5 checkboxes corresponding to the five model organisms supported by OrthoNets Below this are any custom networks available and add browse buttons to add additional custom networks Customization is covered in further detail in Section 3 After selecting one or two organisms and clicking next the plugin will begin loading gene protein annotations and interaction data into memory Depending on the computer s speed and available RAM this process can take some time but typically finishes within 20 30s on modern hardware After this process completes the networks corresponding to the selections made in the previous step will become visible By default all proteins corresponding to genes of interest see Section 3 2 are shown as nodes and the interactions between them are shown as edges If more than one interaction network was selected both networks will be shown side by side in a tiled arrangement E coxape Deskop New Session ag LLL T E File Edit View Select Layout Plugins Help SE QQQQ B AR FB 3 Human PROTEIN VIEW Control Panel Ka fe Network vizMapper Editor Filters OrthoNets Views Retile Focus on Protein Unfocus rotein View Gene View V Prote Visualization Legend Highlight Homologs Clear Homologs
5. A 12 M eB Ae CU CO a O earn EE E E E A E EEE E E 12 4 1 1 C stOmInNterac Uon NETWOIKS acters concise cease Sava st cage EE EEE E EE T E cueranttu 12 AA CUSTOM genes Of INTELES nesnenin aner E AEE E E EAr 14 4 1 3 Overlaying Experimental Data sainsacsusatsracdncagnscinnensaautaysrnessatutseaser yabateutenstvarndenaiiuansetaauecaeiweneueet lt aaiaacceaesataales 14 S Advanced OPON seca nig cere aanecen EE EA EA EESE E E EAEE EA 16 LSA NAINE CUSEOMIZ a E E E E E measeeeuaneataeehoae 16 Dd SIN COC VION EONS eE E O A oasis demetnccued 18 BG OSSar YON TOMS EA EEE NE A T A A E AE EE E E eT 19 FeR TONCO E E ae tess tire en ean eed E E E E E 20 Introduction The availability of biological information varies widely for different model organisms For example the protein protein interactome of the yeast S cerevisiae is comparatively well characterized whereas only a small fraction of the human interactome is known Conversely there are thousands of human genes that have been linked to disease but there is little equivalent information in yeast The OrthoNets plug in for Cytoscape enables the simultaneous visualization and navigation of interaction networks of multiple organisms supporting knowledge transfer from one organism to another A major difficulty is that protein protein interaction data are often scattered across multiple databases necessitating a potentially long and tedious phase of gathering data prior to any analysis OrthoNets avoid
6. numeric identifier assigned by IREF to every unique protein interaction stored in the database Seldom used in OrthoNets except to specify custom networks see Subsection 4 1 19 ROG The numeric identifier assigned by IREF to every unique protein stored in the database ROGs are the primary identifier used by OrthoNets but are only used internally Except for certain advanced customizations the user can elect to specify their data using the more common Entrez Gene identifier Turinsky A L et al DANCER Disease Annotated Chromatin Epigenetics Resource Nucleic Acids Res Kouzarides T Chromatin modifications and their function Cell 2007 128 4 p 693 705 Razick S G Magklaras and I M Donaldson iReflndex a consolidated protein interaction database with provenance BMC Bioinformatics 2008 9 p 405 4 Pu S et al Expanding the landscape of chromatin modification CM related functional domains and genes in human PLoS One 5 11 p e14122 7 References 1 2 3 20
7. the selected attributes will be shown for all selected nodes Information Selection Ey Node Edge First Neighbours Human Yeast Node Paf1 DISPLAY_NAME Paf1 UNIPROT Q8N7H5 Q9H166 GO_PROCESS GO 0010390 histone monoubiquitination GO 0033523 histone H2B ubiquitination ENTREZGENEID 54623 Select Attributes Select rtf1 Attributes of rtf1 are displayed Finally the First Neighbours tab displays the number of interactors containing the selected protein i e first neighbours and indicates how many of these have diseases and how many are genes of interest This tab also has a Create first neighbour graph button that launches the OrthoNets graphing tool 10 of neighbours 6 distinct attributes 8 z ES intracellular signaling cascade Total of attributes 12 jse a agan histone monoubiquitination CRM GO_PROCESS HGNC MORBID MORBID_FULL TAXONOMY z EE cellular response to insulin 2 histone H2B ubiquitination Chart Types activation of protein kinase i Bar Graph B activity Q Pie Chart iv All positive regulation of establishment of protein localization in plasma membrane GO_COMPONENT peptidyl threonine GO_FUNCTION phosphorylation Graph Options Attribute Number of Nodes Show Display Other attributes histone H2B ubiquitination histone monoubiquitination _ activation of protein kinase B activity cell
8. the Cytoscape plugin manager To access the plugin manager launch Cytoscape and click on Plugins gt Manage Plugins Under the analysis category select OrthoNets and then click install Launching webstart Navigate to http wodaklab org orthonets and click the Webstart tab Depending on the configuration of your browser and installed operating system you may receive security warnings These can be safely disregarded Once Cytoscape starts please proceed to the Section 2 Loading Interaction Networks Alternative standalone installation method Click the OrthoNets standalone version download link from the OrthoNets webpage as shown above and save the jar file to your Cytoscape plugins directory By default the plugins directory is inside the Cytoscape 2 7 installation directory 2 Loading Interaction Networks From the Plugins menu in Cytoscape select OrthoNets gt Networks gt Load Networks If this is the first time running OrthoNets a notification will pop up explaining that additional files need to be downloaded before the plug in can run After these files are downloaded a dialog will prompt you to select the interaction networks you want to work with la PpiPlugin Wizard Choose Version iRef 7 0 v ieannspereedPepbeasnenvent Select two organisms to compare or view the network of one Fly Human _ Mouse Worm _ Yeast Optionally enter your own custom network Browse Cancel
9. 5 or newer available from http www oracle com technetwork java javase downloads index html To install java click the Download JRE button at the above webpage and follow the installation instructions located at http java sun com javase 6 webnotes install index html OrthoNets can require a lot of RAM memory depending on the size of the interaction networks being visualized and queried For the included iRef networks at least 2GB of RAM memory is required and approximately 100MB of disk space If OrthoNets still runs very slowly you may need to increase the amount of memory allocated to java as detailed at http cytoscape wodaklab org wiki How to increase memory for Cytoscape OrthoNets is available via the Cytoscape plugin manager recommended and from http wodaklab org orthonets both as a stand alone package and a webstart version Both packages provide the same features and functionality although the webstart version requires an internet connection to operate and is pre packaged with Cytoscape 2 7 The webstart version is best suited for those who want to 2 explore OrthoNets without the additional complexity of installing the plugin whereas the standalone version is most appropriate for those who want to work with OrthoNets offline or for those who wish to use OrthoNets in tandem with other Cytoscape plugins Installing the standalone version via Cytoscape plugin manager recommended OrthoNets is available via
10. 7 0 Right click drag to ZOOM Middle click drag to PAN The display can be divided into five main areas the networks themselves and the four main panels the view panel the Visualization Legend panel the Information Selection panel and the customized Cytoscape Data panel 3 1 The Networks Human PROTEIN VIEW o E 88 4 Yeast PROTEIN VIEW fetes Here the networks for two organisms Human left and yeast right have been loaded and the view centered on the human protein Paf1 On the right this centering operation loaded all interactions involving the centered protein as well as all interactions involving interactors of Paf1 The centered proteins are always represented by red nodes and genes of interest are highlighted with a white border On the left the view is centered on all proteins orthologous to yeast Paf1 These nodes are always highlighted in orange In general the centering option applies to all specified genes and all orthologs to these genes Thus it is possible to enter a list of genes from both organisms which would result in a mix of red and orange nodes in both of the networks The left and right networks are synchronized so that selecting a protein in one network automatically selects the orthologous proteins in the other network Hovering the mouse over a node displays the node identifier For the included networks this identifier is the iRef redundant object identifier ROG 3 Right
11. HOSPITAL FOR SICK CHILDREN amp UNIVERSITY OF TORONTO OrthoNets v1 0 For Cytoscape Yanqi Hao Anna Merkoulovitch James Vlasblom Shuye Pu Andrei L Turinsky Denitza Roudeva Brian Turner Jack Greenblatt Shoshana J Wodak 29 11 2010 Hospital for Sick Children Molecular Structure Program and the Dept of Biochemistry University of Toronto Table of Contents DUAN Trae tet crc ea cee ty reece nates ees netse danse mete S E a aceeseanalecouesseeuietes 2 1 Getting and Installing OrthoNets ccscssscssscossccesccscsussccuscousscessesesestcsstesssesscssenasenusecutscssecssccssscusseussosssesseursoereess 2 2 Loading Interaction ING LW ON K Ss ss saescoacccautnrecaccatapareessaneetennssaacedeipaudus esa apnnegaseaecindpaetesaagaaioneantaiarvasaueataitonnnroasnntahasienes 3 3 Working with NETWOIKS sosieciccoscndisaeesseansinsicsadedantevicdenesavddsaceansinbedvensebededeantandegssdanteadedanesctonntdeansionieinncvassdidesseehdaastasnbeaaes 5 ILTEN WOK cers scess tates E cues eau giten es eonsateec vacouasendusaatveseseidecesnesteasapaeencssateee 6 LAVON AS arisen epecrre acetates aio E gate ceec teenth ceases aecevacousaeemecnanendactuastostacetnenenaaceseseccanertonseepsenacoaeceee 7 5 5 Visual Options Legend tNe Mrsccesaeceneqasseanciaecanaaanbasmiaacsamaaneaeoasmasseaese AT EEE 8 S EEEE Se EEE EE E E E teed E E E E E E E 9 sa Tne DI a Pane a E EE sceeossccumeuendecoe docu 12 A EEn Ze El OM E E E E E A E A E A A A E
12. ansparency of all nodes connected to one ea LinkOut re POSEN Selected Nodes r Show Non Adjacent Edges Hide Non Adjacent Edges Dim First Neighbours of the selected nodes Undim Selection Restores the default opacity to the selected nodes First Neighbours Undim Selection Center on Selected Nodes Center on Selected Nodes Applies the centering operation to the selected nodes bypassing the search dialog The last First Neighbours option is a shortcut to the graphing functionality detailed in the information panel Subsection 3 4 3 2 View Panel E Gene View Protein View The view panel contains a Retile button which resets the network windows to their usual side by side arrangement without affecting the displayed layout This is a useful option if the windows have been manually resized and it is desired to restore them back to their default layout The Focus on Protein button is a shortcut to the gene protein search dialog below left and the Unfocus option shows all the proteins corresponding to the genes of interest below right _ z Select Nodes So p pe San mA If you want to focus on a particular group of genes proteins and their interaction partners please select them here Otherwise click Finish Species Human v pafi J Human PROTEIN VIEW eld fad 2 Yeast PROTEIN VIEW fe els With all of the words Search
13. ation Legend Homology Diseases Highlight Homologs A E Homologs of rtf1 i Homologs of Centered nodes E Homologs of CDC73 E Centered nodes Homologs of LEO1 E Homologs of CTR9 This panel is used to affect the display of network nodes based on homology or disease information Under the Homology tab the colours used to identify the centered protein and the orthologous genes proteins are listed Clicking a colored square will automatically select all orthologs assigned to that color or the centered node if the red square is clicked When new networks views are created by centering on proteins the homology legend is cleared and no colours are assigned except red for the centered node and orange for the orthologs of the centered node Colours can be individually chosen for each set of orthologs by right clicking a node of interest and selecting the Display Homolog as discussed in the previous subsection Alternatively clicking the Highlight Homologs in this panel will assign separate colours to each group of orthologous proteins The disease legend is accessible by clicking the Diseases tab in the legend panel Here you can use node shapes to indicate which genes are linked to disease according to OMIM First click Assign Disease Shapes to bring up the disease assignment dialog This lists all diseases that are linked to any of the proteins currently displayed Clic
14. clicking a node presents several options in addition to the default options provided by Cytoscape Selecting Display Homolog shows all proteins orthologous to the node using a color coding noted in the legend panel Below the node protein LEO1 in Human has been clicked and assigned the color purple The orthologous nodes in Yeast have simultaneously been color coded purple In some cases additional proteins can become unhidden if the Display Homolog option is selected if those proteins were not adjacent to the centered protein and thus not displayed prior to selecting the Display Homolog option 3312145 Visual Mapping Bypass gt Nested Network b Use Web Services d Hide Node LinkOut Homolog Display Homolog Centralization Undisplay Homolog selected odes First Neighbours The centralization option provides a convenient shortcut for centralizing on a protein of interest bypassing the usual search dialog covered in the previous section If one or more nodes are selected highlighted yellow by clicking them an operation built into Cytoscape the Selected Nodes menu entry will become active exposing several options Note that the same Selected Nodes section is accessible via the Plugins gt OrthoNets menu Hide Non Adjacent Edges Shows Hides all interactions that do not involve one of the selected proteins Patt M esac ad Dim First Neighbours Reduces the tr
15. ecies Human v OrthoNets Network r i ae 7 TT us 7 pa JESE Focus Unfocus Select Genes Proteins to Focus ET gt With all of the words pecs Select Centered Nodes Show all proteins of interest Search With any of the words Selected Nodes gt Select Genes Proteins of Interest i Advanced View Options gt Select CRM proteins as Proteins of Interest ADCO _ Match Exactly V Auto Filter ou ar Homologs Settings Protein Organism ID APAF1 3 124814 ATPAFI Pafi 596227 3612408 w Select Select All a Unselect Add Remove Columns F a Protein Genes Organism ID The search dialog Species drop down enables the search to be constrained by species or to be applied to all species By default the dialog will return all proteins whose protein name Entrez Gene ID or iRefWeb identifier partially match any of the entered search terms Clicking Add Remove columns enables searching based on other criteria For example in the above dialog the column MORBID was added allowing the user to search for genes linked to specific diseases Search hits are selected by highlighting the row corresponding to the desired protein and clicking Select Clicking Finish will then center the network views on those selected genes To undo this operation select Focus Unfocus gt Show all proteins of interest When the Auto Filter is toggled off the
16. ed by pp Most edge attributes are also lists specified as a delimited set of values enclosed in paranthesis For example for the above the edge connecting the protein identified by ROG 139374 to the protein identified by ROG 5124145 is supported by the publications with pubmed identifiers 11283351 and 10655498 The format for an edge attribute is ATTRIBUTE_NAME class ROG pp ROG value value value iRef_x x organisms Fly Yeast Human Worm Mouse sif These are the interaction files themselves for each of the five organisms identified in sif format Each line consists of a pair ROGs delimited by pp ROG pp ROG ROG pp ROG Usually these files should not be modified as custom interaction networks can be added via the GUI facility documented in Subsection 4 1 iRef_x x orthology orthologs omf This file lists all pairs of orthologs between all organisms Each line is consists of a pair of ROGS delimited by tab ROG lt tab gt ROG ROG lt tab gt ROG iRef_x x pda domain_arch pda The domain architectures for each protein are contained in this file one domain per line For example 1000673 2140 DUF3398 PFamA PF11878 Domain of unknown function DUF3398 protein 65 159 1000673 2140 Ded_cyto PFamA PF06920 Dedicator of cytokinesis protein 1930 2107 1001137 752 KA1 PFamA PF02149 Kinase associated domain 1 protein 706 752 The file is contains eight columns delimited by tabs In order these are 1 iRef ROG ide
17. eract with a disease associated protein As above but only interactions involving a protein of interest are shown Only show Proteins of Interest Only show those Proteins of Interest with disease MORBID annotations In addition to the above also show all proteins that interact with a disease associated Protein of Interest As above but only interactions involving a Protein of Interest are shown Shows all proteins of interest regardless of disease association and all disease associated proteins interacting with these proteins of interest Only those 18 Proteins of Interest with diseases and first neighbours with diseases and edges incident with Proteins of Interest edges incident with a protein of interest are shown Shows all disease annotated proteins of interest and all disease associated proteins interacting with these proteins of interest Once again only those interactions involving a protein of interest are shown 6 Glossary of Terms Term Center Centralize DAnCER Focus Genes of Interest Homolog Interolog MORBID OMIM Ortholog Overlay Network PFam Proteins of Interest RIG Description Centering is an operation applied to a group of proteins which hides all proteins not interacting with a centered protein or an ortholog of a centered protein if two networks are loaded and visible Centered proteins are colored as red nodes and if two networks are loaded orthologs of centered pr
18. he first line is the name of the attribute which should match the name of the file minus the na extension This is followed in parentheses by the class of the attribute which can be any type supported by Cytoscape including java lang String java lang Integer or java lang Double In OrthoNets most attributes are list attributes meaning that each protein may be assigned multiple values for that attribute In this case the protein with ROG 1657917 is assigned the NCBI RefSeq aliases 31560057 75788 NP_001033716 etc A list is specified as a delimited set of values enclosed in parentheses In general the format for a node attribute is ATTRIBUTE NAME class ROG value value value iRef_x x attributes ea These are the edge attribute files which store information about the publications and experiment types supporting protein protein interactions The ea file format is the standard edge attribute format documented 16 in the Cytoscape manual OrthoNets loads all edge attributes contained in this directory automatically Consider the example PUBMED ea edge attribute file PUBMED class java lang String 2255804 pp 4772972 14592989 2761514 pp 4705328 11805837 3083028 pp 4629183 18467557 139374 Op 5124145 11283351 10655498 N NN NN NAN The file beings with the name of the attribute followed by the class name in parentheses see na description above An edge is identified by a pair of ROGs separat
19. his will add the network in the Custom section of the dialog The simplest way to provide custom networks is to use the iRefWeb filtering interface available at http wodaklab org iRefWeb This interface allows you to query the iRef database with customized filters These filters act as restrictions on the returned result set For example the user can elect to only consider interactions supported by 2 or more publications from small scale experimental methods 12 s T About iRefWeb File Edit View History Bookmarks Tools Help CB SX A GF hitoy modakiad org iRetWeb Most Visited _ Getting Started a Latest Headlines About iRefWeb Wodak Lab 9 word placing images sic Home Info People Projects Tools Resources Sponsors Blog 7S le q i oS gt 4 7 7 AS 2B eS A AA E Se a ae l 6i ae ho Ve E Papers The iRefWeb interface groups interaction records from the different databases into a single non redundant view In particular iRefWeb facilitates comparing k Done Wodak Lab Tools 23 In collaboration with the Donaldson Research Group l RefWe b Using iRefIndex Version 7 0 About Statistics PubMed Reports PubMed Detail FAQ Release Notes iRefWeb is an interface to a relational database containing the latest build of the Search Tab interaction Reference Index iReflndex which integrates protein interaction data Search Query from eleven different in
20. in the information panel see Subsection 3 4 for display Note that only the first two columns are required but providing gene names is recommended To load an experimental file such as the above select OrthoNets gt Plugins gt Network gt Overlay Edges This will overlay the interaction file but will not change which proteins are centered Therefore overlaying data is usually a two step process First choose the file to overlay and second choose proteins to center on via the Plugins gt OrthoNets gt Focus Unfocus gt Select genes proteins to focus dialog The overlaid data is color coded based on whether the user specified interactions exist in the base network or not Here the base network is either one of the included iRef networks from Mouse Human Worm Fly or Yeast or a user specified custom network see Custom Interaction Networks subsection above Green indicates those edges recorded in the base network that are not present in the overlaid data whereas purple indicates those edges in the overlaid data that are not in the base network Cyan edges are those present in both In addition OrthoNets highlights in gold those edges present in the overlaid data that are not in the base network but whose interologs are present in any of the other four organisms In the below example experimental TAP MS data is overlaid on the human network and the human protein Paf1 has been focused El Human PROTEIN VIEW ecco te ES
21. ing different attribute values or by selecting all attribute values If some but not all of the first neighbours are not assigned an attribute value then by default these nodes are not considered by the graphing tool To group these nodes together into an Other category the user can check the Display Other Attributes option Various options for saving exporting the graph changing the range and zooming are available by right clicking the graph 11 3 5 The Data Panel Data Panel ta Domain Architecture Visualization v Show pFamA Show pFamB Pafi Yeast 0 445 WI Pafi Human o 531 Networks centralized for 1 node Paf1 H Node Attribute Browser Edge Attribute Browser Network Attribute Browser Domain Visualization The data panel is comprised of four tabs the node attribute browser edge attribute browser and network attribute browser are Cytoscape built in features and are documented in the Cytoscape manual These tabs are seldom used in OrthoNets being largely supplanted by the information panel documented in Subsection 3 3 The fourth tab and the one visible by default is the protein domain visualization panel which shows the PFam domain architectures for all currently selected nodes These diagrams represent a protein as a line and the domains as colored rectangles sized proportional to their sequence length Rolling the mouse over these colored rectangles provides additional infor
22. king Assign Diamond to all Diseases will change the nodes linked to any disease to a diamond shape Those nodes linked to several diseases will change to octagons and those without any disease association will remain elliptical Alternatively you can select individual diseases from the list and assign them a custom node shape For example below nodes associated with Hyperparathyroidism have been changed to triangles Views Retile Focus on Protein Uni Disease A f tT _ Gene View Protein View Congenital disorder of glycosylation E Hyperparathyroidism Visualization Legend D Hyperparathyroidism jaw tumor syndrome Homology Diseases O yperp y j y Parathyroid adenoma with cystic changes Q Choose Shape Square Triangle Assign Disease Shapes Clear All C Parathyroid carcinoma C Hexagon LS lt gt Rhombus Circle Leave disease shape as circle to ignore Information Selection J Done i Assign rhombus to all diseases _ Cancel _ r l Finally just like the homology legend clicking any of the legend entries will automatically select all of the corresponding nodes 3 4 The Information Panel Information Selection a Node Edge First Neighbours Human Yeast Node Paf1 DISPLAY_NAME Paf1 UNIPROT Q8N7H5 Q9H166 ENTREZGENEID 54623 Select Attributes The info
23. mation about that domain including the name PFam identifier and the amino acid start and end positions of the domain on the peptide Left clicking a domain automatically launches a web browser with the PFam summary for that domain 4 Customization All of the data used by OrthoNets is customizable Section 4 1 covers the basic customizations that are built into the OrthoNets graphical user interface GUI Some file formatting may be required but OrthoNets was designed to make the import of this custom data as simple as possible Section 4 2 covers the more advanced customizations that involve replacing the data files that OrthoNets uses No special GUI facility is provided for this type of customization and formatting these files may require more effort than the basic customizations outlined in 4 1 In addition the advanced customizations will require additional knowledge about how OrthoNets operates internally 4 1 Basic Customization There are three main aspects of OrthoNets that are customizable Custom interaction networks genes of interest and the overlay of experimental data 4 1 1 Custom interaction networks To load a custom interaction network navigate to Plugins gt OrthoNets gt Networks gt Load Network This will launch the first step of a wizard as covered in Section 2 of this document To add a custom network click Browse and select a custom interaction network file format covered below and then click add T
24. ntifier of the protein 2 Length ofthe protein 17 Short name for the domain One of either PFamA or PFamB PFam Identifier Short Description of Domain Amino acid start position of the domain Amino acid end position of the domain ote St a 5 2 Advanced View Options The advanced view options apply to the entire interaction network not just the currently centered view Advanced viewing options can be accessed from Plugins gt OrthoNets gt Advanced View Options Plugins Help Manage Plugins Update Plugins Advanced Network Merge OrthoNets Network FLB TEIN VIEW Focus Unfocus Select Centered Nodes Selected Nodes gt Advanced View Options About Settings Regardless of Proteins of Interest Only Diseases Only Diseases and First Neighbours Only Diseases and First Neighbours and edges incident with Proteins of Interest Only Proteins of Interest Only Proteins of Interest Proteins of Interest with Diseases Proteins of Interest with Diseases and First Neighbours Proteins of Interest with Diseases and First Neighbours and edges incident with Proteins of Interest ALL Proteins of Interest and First Neighbours with diseases and edges incident with Proteins of Interest Regardless of Proteins of Interest gt Only Proteins of Interest gt All nodes with at least one disease MORBID annotation In addition to above also show all proteins that int
25. of interest These genes and proteins corresponding to these genes will always be highlighted with a white border within the OrthoNets network views In addition several of the graphing options and Advanced View Options documented in Section 3 provide customized views based on these genes of interest and their interaction neighbourhood Genes proteins of interest can be specified using the Plugins gt OrthoNets gt Select Genes Proteins of Interest option This raises a dialog very similar to the one used for centering see Section 2 for details 4 1 3 Overlaying Experimental Data In OrthoNets a user specified network can be overlaid on an existing network view This can be used to visually query which interactions in the user network have been previously reported in the literature as 14 recorded in the iRef database and which are novel An experiment file is formatted as a tab delimited list of entrez gene identifiers gene names and several optional data columns prefaced by a header For example the following example is from an affinity purification experiment Bait lt tab gt Protein lt tab gt BaitName lt tab gt PreyName lt tab gt SpectralCount 10428 lt tab gt 10856 lt tab gt CFDP1 lt tab gt RUVBL2 lt tab gt 12 51 8930 lt tab gt 3853 lt tab gt MBD4 lt tab gt KRTo6A lt tab gt 50 434 Here the SpectralCount is a score associated with each interaction This score will be loaded as an edge attribute and can be selected
26. of these gt Search Download Interatome Clear New Search Show Search Help Your query can consi low or both in any combination Filters Expand All Collapse All Show Filter Help Entrez Gene IDs PubMed IDs Protein Aliases Upload a list of gene IDs Source Database Organism Nature of Interaction bind 11207 _ single organism interaction 79002 C unary 2191 _ bind translation 4715 E cross organism interaction 478 IE pairwise 71162 _ biogrid 52082 _ multi subunit 6127 13 1 2 Navigate to http wodaklab org iRefWeb and select the Search Tab Select the Expand All option in the filters section to view the list of available filters Check off the desired organisms If more than one organism is checked then OrthoNets will automatically create separate networks for each organism At this time OrthoNets only supports pairwise physical interactions within the same organism so also check the physical and pairwise options under the Nature of Interaction section and the single organism interaction option under the Organism section Additional filtering options can also be selected 4 After checking the desired filters click Search 5 Once the search is complete click Download Interactome y jl Download Your Interactome 6 Click Download as interaction ID List and save We The download come
27. oteins are colored orange The Disease Annotated Chromatin Epigenetics Resource 1 See Center A user specified list of genes whose corresponding proteins see Proteins of Interest are always highlighted For OrthoNets the term homolog is used synonymously with ortholog See also Ortholog An interaction between two proteins is said to be an interolog if orthologs of those proteins interact The Online Mendelian Inheritance in Man MORBID map is the source of disease gene associations used by OrthoNets See also http www ncbi nim nih gov omim In OrthoNets this is synonymous with MORBID Two genes are orthologs if they derive from a common ancestor The operation of superimposing a user specified network on a network derived from iRef or on another user specified network see Subsection 4 1 1 Edges that are in common or unique to each network are highlighted in different colors Subsection 4 1 3 The source of domain architecture information used by OrthoNets See also http pfam sanger ac uk for more information Proteins of interest are always highlighted with a white border By default the proteins of interest are those known or predicted to be involved in chromatin modification 4 but this can be overridden using the interface documented in Subsection 4 1 2 Proteins of interest are also treated as a separate category in the information panel see Section 3 and affect certain advanced display options The
28. rmation panel is primarily used to display node and edge attributes and largely replaces the Cytoscape node and edge attribute browser This allows for the simultaneous display of node and edge attributes in the information panel and protein domain architectures in the data panel In OrthoNets node attributes include various gene protein aliases such as UniProt Entrez Gene IDs and NCBI RefSeq GO annotations and disease information from OMIM MORBID Edge attributes record information related to each interaction such as the pubmed identifiers of supporting publications and the types of experiments used to detect the interaction To select the node or edge attributes to display for an organism click the Node Edge tab the organism s tab and then Select Attributes This invokes the Select Node Edge Attributes dialog Select Node Attributes x Please select attributes selected attributes are displayed on the right 7 Tr 7 Available Attributes Select Selected Attributes GO_FUNCTION DISPLAY_NAME GO_PROCESS ENTREZGENEID HGNC 4 Unselect UNIPROT MORBID MORBID_FULL REESEN Finish GO_PROCESS GO slim biological process annotations Here the available attributes are listed on the left and those selected for display are listed on the right The select and unselect buttons are used to select unselect attributes for display After clicking finish all of
29. s in two formats the file for import into OrthoNets e MITAB see the help section for details e As a list of Interaction IDs for use in OrthoNets th There will be a pause a few seconds or more depending on the number of records and then your download will begin Note if any interactions are exclusively from BINDTranslation they will be suppressed because it is not yet public Download as MITAB Download as interaction ID list Cancel lal Homo sapiens 0 Alternatively you can specify your own files using one of the two file formats that are supported by OrthoNets for custom networks The first is a newline separated list of iRef interaction identifiers RIGs prefaced by the header iref_interaction_id without quotes For example iref_interaction_id 14 A9 The second file format is a newline separated list of tab delimited pairs of Entrez Gene identifiers prefaced by the header entrezgeneid without quotes For example entrezgeneid 855710 lt tab gt 854063 854663 lt tab gt 854063 856885 lt tab gt 8s54063 For the second file format OrthoNets will assume interactions between all isoforms corresponding to these identifiers i e for a given pair of entrez gene identifiers the assumed interactions are the Cartesian product of the two sets of isoforms 4 1 2 Custom genes of interest By default OrthoNets identifies those genes known or predicted to be involved in chromatin modification as genes
30. s this difficulty by taking advantage of the iRefWeb resource http wodaklab org iRefWeb which provides access to a repository of non redundant interaction data retrieved from ten public interaction databases OrthoNets allows interaction networks from two organisms to be viewed simultaneously and synchronizes the views based on pre computed orthology relationships between proteins genes Each protein is annotated with all aliases available in iRefWeb OMIM disease annotations and domain information from PFAM and every edge is annotated with the number and types of experimental evidence supporting that interaction Up to four simultaneous views are supported for any two organisms These views are derived from the protein interaction network for each organism where the nodes are proteins or genes respectively OrthoNets also integrates with iRefWeb to provide filtered views of the interaction networks and allows the user to quickly center the views on specified genes of interest for example genes involved in a particular cellular processes or cellular localization and infer information between orthologous genes proteins their respective environments in the interaction networks and the disease annotations whenever applicable 1 Getting and Installing OrthoNets Before installing OrthoNets please verify that your computer has Cytoscape 2 7 or newer installed available from http www cytoscape org Cytoscape and OrthoNets require Java
31. search hits are not updated as you type This is a useful feature if pasting in long lists of numeric ids e g Entrez Gene IDs and is often used in tandem with the Match Exactly toggle to quickly locate a specific list of genes If the Use AND Search toggle is activated all the search hits must match all the entered terms although the match can be from any of columns 3 Working with Networks A typical OrthoNets session is shown below E Gytoscape Desktop New Session saul File Edit View Select Layout Plugins Help SARA RBA R 1B j Control Panel F L amp j Human PROTEIN VIEW fg Network VizMapper Editor Filters OrthoNets Views Retile Focus on Protein Unfocus Gene View Protein Viev Visualization Legend Highlight Homologs Clear Homologs _ Homologs of rtf1 Homologs of Centered nodes E Homologs of CDC73 IEA Centered nodes Homologs of LEO1 m Homologs of CTR9 Information Selection Node Edge First Neighbours Node f DISPLAY_NAME rif1 Data Panel fa UNIPROT P53064 P89115 Domain Architecture Visualization ENTREZGENEID 852607 V Show pFamA Show pFamB rtfl Yeast rtf1 Human Networks centralized for 1 node Pafi H Select Attributes z z Steamin Node Attribute Browser Edge Attribute Browser Network Attribute Browser Domain Visualization Welcome to Cytoscape 2
32. teraction databases BioGRID BIND CORUM DIP peant Crenn Mol Ta AN Entel oo HPRD INTACT MINT MPPI MPACT OPHID and BINDTranslation EF Integration is achieved through a rigorously documented procedure for mapping Same Prosi Rewer protein IDs across databases enabling systematic backtracking of the links used Se is err to establish the identity of the interaction partners ee Find interactions and interactively build and download your interactome ae o Download Interatome Clear New Search Show Search Help Your query can consist of sear cha OE the filters below or both in any combination Filters expand All A ollapse All Show Filter Help PubMed IDs Entrez Gene IDs Protein Aliases Source Database Organism Nature of Interaction Number of Publications Lowest Experiment Size LPR Highest Experiment Size HPR Interaction Detection Method Interaction Type Nature of Interaction T unary 0 Organism Vv single organism interaction 70707 A I cross organism interaction 0 IV pairwise 70707 T multisubunit 0 Arabidopsis thaliana 0 I Caenorhabditis elegans 0 T predicted 0 I Campylobacter jejuni 0 I experimental 70707 7 Drosophila melanogaster 0 I Escherichia coli K 12 0 T Escherichia coli 0157 H7 0 l Homo sapiens 0 I genetic 8803 MV physical 70707 T Mus musculus 0 I Schizosaccharomyces pombe 0 1441 Hidden 0 Can match ANY
33. ular response to insulin stimulus intracellular signaling cascade _ negative regulation of protein kinase activity Right click the chart for peptidyl threonine phosphorylation nell saving and editing options positive regulation of establishment of prote Pie Chart Options et uo i This tool provides a visual summary of the interaction neighbourhood of a selected protein Above the gene Paf1 was selected prior to clicking Create First Neighbour Graph As shown in the upper left of this window Paf1 participates in 6 distinct interactions i e has 6 neighbours In the Attributes section of this dialog the user can select which attribute to be graphed in this case the GO Biological Process annotations The total of distinct biological process annotations among the first neighbours of Paf1 is given as 8 in the upper left of the window To reduce display clutter the OrthoNets graphing tool shows only the top 10 most common attribute values as also indicated by of distinct attributes in the upper left Also those attributes that are not set among any of the neighbouring proteins will not be shown in the Attributes section of this dialog e g if none of the neighbouring proteins are linked to a disease in OMIM the MORBID and MORBID_FULL options would not be shown Using the scrollable table at the bottom middle of the dialog the user can override this behaviour by select
Download Pdf Manuals
Related Search