User Manual

Examples

PTMOracle comes ready with example protein-protein interaction (PPI) networks. Every network that is created by PTMOracle will be put into its own network collection and present the user with a network view. The name and details of each network are provided below. To create any one of these networks, follow the steps provided: Apps -> PTMOracle -> Examples -> network name

Yeast Interactome - Yeast PPI network (referred to as SBI) curated by Pang et al. (2012).
Human Experimental Interactome - Human PPI interaction network (referred to as HI-II-14) derived from Rolland et al. (2014).
Human Literature Interactome - Human PPI interaction network (referred to as Lit-BM-13) derived from Rolland et al. (2014).

Importing Protein Data

PTMOracle allows users to quickly annotated protein nodes in the current network view with different types of protein data. By default, this includes protein sequences, post-translational modifications (PTM) sites and annotations such as domains, motifs and disordered regions. However, by customising PTMOracle, users can also annotate protein nodes in the network with other types of protein data. To learn more about how to customise PTMOracle, see below Customisation. To annotate protein nodes, follow the steps provided: Apps -> PTMOracle -> Import...

Once a file has been chosen, users will be able to see which network will be annotated with protein data as well as the file path of their selected file. Users will also be asked to choose a column in the Cytoscape table that contains the IDs associated the protein data in the selected file. It is important that the IDs in the selected file match the IDs in the chosen column, or else the protein data cannot be mapped onto the protein nodes in the network. In addition, users will be asked to input a 'Source' so that PTMOracle can keep track of which file and protein data has already been imported. This can simply be the name of the file or type of properties that are being imported.

Note that importing protein data will not overwrite previously imported features unless they are from the same source. In addition, importing protein data may take some time depending on the size of the network, size of the file, type of data imported and how much data has already been annotated onto the network. Please ensure that all protein data has been formatted correctly or else you may encounter some problems! For more information, please refer to File Formats.

Once the file has been imported, columns will be created in the Cytoscape node table summarising all the protein data that was annotated onto each protein node. But most importantly, all other features of PTMOracle (OraclePainter, OracleTools and OracleResults) can now be used!

Additional information:

Users may also wish to copy protein data annotated on one Cytoscape network into another Cytoscape network. This can be done under Apps -> PTMOracle -> Preferences -> Mappings . To copy protein data, users must click the 'Add Row' button and fill out the form presented. Drop-down menus will be able to assist users in filling out the information. Users may also wish to remove unwanted protein data mapped onto a network and this can be done by clicking the 'Remove' button.

Exporting Protein Data

PTMOracle also allows users to export all the protein data imported into Cytoscape by PTMOracle. To export protein data, follow the steps provided: Apps -> PTMOracle -> Export...

Once a file has been chosen, users will be able to see all the protein data associated with the network in a table format. Once exported, the results will appear in the chosen file in a tab-delimited format.

OraclePainter

Located in the Cytoscape Control Panel, OraclePainter allows users to visualise PTMs on protein nodes as pie charts in the network view. This enables users to quickly and easily identify proteins with a specific type of PTM, or combination of PTMs. To visualise PTMs on protein nodes: Click the Apply button.

Users may also wish to import or export their colour schemes, as well as create a colour legend. These can be found in the Settings button. The name and details of each painter are provided below.

OracleTools

Located in the "Apps" menu bar under the PTMOracle submenu, OracleTools contains a number of query modules that can utilise the protein data annotated onto protein nodes. It allows users to build complex and systematic searches to construct subnetworks for more targeted investigations. The name and a brief description of each module are provided below. To run a query module on the network, follow the steps provided: Apps -> PTMOracle -> OracleTools -> query tool name

Once a module has been chosen, the user will be presented with a menu containing a brief description of the module as well as several parameters that must be assigned by the user. After the module has run, a column will be added into the Cytoscape node table with the results for each protein.

Calculator - Calculates the number of features or modification types on each protein.
MotifFinder - Finds proteins with a specific motif pattern in the protein sequence.
PairFinder - Finds proteins with specific modification pairs that are separated by a specified number of amino acids.
RegionFinder - Finds proteins that contain a specified modification within a specified region (e.g. domain, motif).

OracleResults

Located in the Cytoscape Results Panel, OracleResults allows users to display protein data associated with any given protein node/s. All protein data is presented to the user in a table format. OracleResults allows users to explore which PTM type or sites might might be involved in interactions and where these interactions might take place. To display OracleResults, ensure that the Results Panel is visible.

To display all protein data associated with any given protein node, users must input the name (or another identifier such as the SUID) of the protein into the search box and click the View button. To display protein data for multiple protein nodes, separate each name with a delimiter (by default, space-delimited). For each protein node, a table of all protein data associated will be presented, along with a text area for the protein sequence if applicable. PTM sites and sequence annotations will also be highlighted in the protein sequence.

Users may wish to adjust the type of delimiter and the column used to search for the protein of interest. These can be found under the Settings button. Users may also wish to display a colour legend to assist in identifying the protein data highlighted on the protein sequence. Users should also be aware of the button with the Refresh icon for updating the table. This is necessary if protein data is already displayed but the user has imported additional protein data.

Customisation

Users can also customise what type of protein data can be imported by PTMOracle. This allows users to annotate protein nodes in Cytoscape with a range of protein data. Please note that customisation of PTMOracle may still be unstable and result in errors or problems. This may require users to re-import previously imported data. For the best PTMOracle functionality, we recommend users to customise PTMOracle before importing data through the import menu. To begin customising PTMOracle, follow the steps provided: Apps -> PTMOracle -> Preferences

Customising PTMOracle involves defining Properties and Keywords . Properties are defined as a specific type of protein data such as PTM or domain. Keywords are considered as denominations of a specific property. For example, phosphorylation and acetylation are PTMs, whereas SH2 and SH3 are domains. In some cases, specific keywords could be differentiated into other subtypes such as phosphoserine and phosphotyrosine.

To define new properties or keywords, users must click the 'Add Row' button and fill out the form presented. Guidelines on how to fill out the form are shown below. Users may also wish to remove unwanted keys or properties from the table by clicking the 'Remove' button.

Defining Properties

Field	Description
Property type	Name of property (e.g. PTM, Disordered, Domain, Motif)
Column name	Name of the column created in the Cytoscape node table
Color	Color code (as hex code) representation of the property when co-located on the protein sequence
Is node property?	Whether the property is for protein nodes or not (default ticked)
Is interval?	Whether the property spans over a number of amino acids or not. For example, PTMs only occur on a single amino acid (hence unticked). Domains occur over a range of amino acids (hence ticked)
Is mergeable?	Whether the property can be clustered or not. For example, Disordered region may be defined between amino acids 1-5 and 3-7, but can instead be merged to 1-7 (hence ticked)
Is column list?	Whether the column created in the Cytoscape node table should be represented as a list or as a boolean (default true)

Defining Keywords

Field	Description
Keyword	Name of the denomination of a specific property. These can be specific (e.g. phosphoserine and phosphotyrosine) or non-specific (phosphorylation and methylation)
Property type	Property that the keyword belongs to (e.g. PTM, Domain). This is based on the property table (see above).
Color	Color code (as hex code) representation of the keyword when visualised with OraclePainter. This is only applicable for PTM properties.
Regular expression pattern	Regular expression pattern which is used recognise keywords. For example, input files may refer to phosphorylation as phosphoserine or phosphotyrosine. A regular expression pattern of phospho will match both and therefore rename them to the Phosphorylation keyword. Note that properties in the input file which can be matched to 2 or more patterns must be resolved by the user upon importing data.

Wiki

PTMOracle / User_Manual