Workflow Entry: Discover_entities

Created at: 10/12/07 @ 21:48:34 Last updated: 10/12/07 @ 22:54:42

Version 2 (latest) (of 2)

Version created on: 10/12/07 @ 21:48:33 by: Marco Roos | Revision comments Expand

Last edited on: 10/12/07 @ 22:54:42 by: Marco Roos

Title: Discover_entities

Type: Taverna 1

Preview

(Click on the image to get the full size)

Download Scalable Diagram (SVG)

Description

This workflow contains the ‘Named Entity Recognize’ web service from the AIDA toolbox, created by Sophia Katrenko. It can be used to discover entities of a certain type (determined by ‘learned_model’) in documents provided in a lucene output format.

Known issues: The output of NErecognize contains concepts with / characters, breaking the xml. For post-processing its results it is better to use string manipulation than xml manipulations. The output is per document, which means entities will be redundant if they occur in more than one document.

Download

Download Workflow File/Package (SCUFL)

Run

Run this Workflow in the Taverna Workbench...

Option 1:

Launch in Taverna

Note: you need to have both the WHIP Launcher and the Taverna myExperiment/WHIP plugin installed on your machine for this to work. See here for information.

Option 2:

Copy and paste this link into File > 'Open workflow location...'
http://www.myexperiment.org/workflows/111/download?version=2
[ More Info Expand ]

Workflow Components

Inputs (2)

Name	Description
input_from_lucene	Example: <?xml version="1.0" encoding="UTF-8"?> <aid:result xmlns:aid="http://aid.vle.org" query="+content:ezh2 +(year:2007^10.0 year:2006^9.0 year:2005^8.0 year:2004^7.0 year:2004^6.0 year:2003^5.0 year:2002^4.0 year:2001^3.0 year:2000^2.0 year:1999)" total="78" time="2"> <doc rank="1" score="0.55880820751190185546875"> <field name="PMID"> <value>15208672</value> </field> <field name="year"> <value>2004</value> </field> <field name="PT"> <value>Journal Article</value> </field> <field name="title"> <value>Activated p53 suppresses the histone methyltransferase EZH2 gene.</value> </field> <field name="content"> <value>... Furthermore, the repression of EZH2 promoter by p53 is dependent on p53 transcriptional target p21(Waf1) inactivating RB/E2F pathways. In addition, the knockdown of EZH2 expression retards cell proliferation and induces G2/M arrest. We suggest that the p53-dependent suppression of EZH2 expression is a novel pathway that contributes to p53-mediated G2/M arrest. EZH2 associated complex possesses HMTase activity and is involved in epigenetic regulation. Activated p53 suppresses EZH2 expression, suggesting a further role for p53 in epigenetic regulation and in the maintenance of genetic stability. Suppression of EZH2 expression in tumors by p53 may lead to novel approaches to control cancer progression.</value> </field> <field name="LuceneDocID"> <value>14861224</value> </field> </doc> </aid:result>
learned_model	Model to discover a set of specific concepts; e.g. the prelearned model named 'MedLine' will make the service discover genomics concepts.

Processors (3)

Name	Type	Description
Default_output_type	stringconstant
Default_input_type	stringconstant
NErecognize	arbitrarywsdl

Outputs (1)

Name	Description
discovered_entities	Entities discoverd in documents provided in lucene output format.

Links (5)

Sink	Source
NErecognize:input_data	input_from_lucene
NErecognize:r_type	learned_model
NErecognize:input_type	Default_input_type:value
NErecognize:output_type	Default_output_type:value
discovered_entities	NErecognize:NErecognizeReturn