Exact matches only
Search in title
Search in content
Search in posts
Search in pages
portfolio
Filter by Categories
ADAM
Architecture
Big Data
Bio-IT World
Bioinformatics
Blog
Cases
cBioPortal
Consultancy
CTMM TraIT
Data Curation
Data Management
FAIR
Galaxy
Harry van Haaften
IMI EMIF
IMI ETRIKS
IMI Translocation
Irina Pulyakhina
Janneke Schoots
Jarno van Erp
Jojanneke Blaak
Joris Borgdorff
Kees van Bochove
Maxim Monait
Natalia Boukharov
OMOP and OHDSI
Open source
Peter Kok
Pieter Lukasse
Piotr Zakrzewski
RADAR CNS
Rianne Jansen
Riza Nugraha
Services
Seth Snel
Sjoerd van Hagen
Software Development
Software Engineering
Software licenses
Software licenses
Solutions
Spark
Team
Topics
Translational Research
tranSMART
TranSMART 2.0
Uncategorized
Ward Weistra
Wibo Pipping

Sjoerd van Hagen

Project Manager / Big Data Architect

@TheHyveNL

In the MRC MATURA project The Hyve has been working with Queen Mary University London to enable GWAS analysis within tranSMART for Postgres, sponsored by Janssen.  As detailed below, we have ported the GWAS functionality in Oracle to Postgres, set up pipelines for loading the reference and analysis data and provided training and example code to the end users. The work was done in collaboration with Mike Barnes and Evan Tzanis in the William Harvey Research Institute Bioinformatics team at Queen Mary University London, and Anthony Rowe from Janssen. The team from The Hyve that worked on this project consisted of Wibo Pipping, Gustavo Lopes and Sjoerd van Hagen.

Requirements

The goal of the project was to enable GWAS analysis in the MATURA project. To establish what was required to do so, Wibo and I have performed a requirements gathering with Mike and Evan during a productive and pleasant, two-day visit to QMUL.

We came up with the following requirements:

  • end users have to be able to perform a GWAS analysis using data in tranSMART to create the cohorts, use plink to do the GWAS analysis and load the results back into tranSMART for visualization
  • it has to work with Postgres
  • Evan would have to be able to set up servers within the MATURA infrastructure
  • additional requirements we would find out from the end users

Activities

To meet these requirements we have successfully:

  • gathered user feedback on the R client and the GWAS viewer
  • gave a training to the end users on how to use the R client to create cohorts and use those to create a phenotype file for plink
  • gave an example of how to run the plink tool to do the GWAS analysis
  • wrote a Python script to transform the plink output into an input file to transmart-batch
  • adapted transmart-batch to be able to load GWAS analyses
  • ported the GWAS code to Postgres (it was only implemented for Oracle)
  • provided Evan with documentation on how to install GWAS-enabled tranSMART
  • provided documentation for the end users on how to use the system

Below is the diagram describing the flow.

transmart

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Results

 

 

 

 

 

 

Leave a Reply

Your email address will not be published. Required fields are marked *

Sjoerd van Hagen

Project Manager / Big Data Architect

@TheHyveNL