University of Heidelberg


Multi-Scale Imaging & Data Mining



An open source data pipeline for (non-) expert users


Open source data pipeline for HCS


The aim of this project is to provide an open source data mining platform - based on the integrative software platform KNIME - for integrating, sharing and processing HCS data using a workflow-oriented architecture which can be used by (non-) experts. Our intention is to develop a platform that covers the entire data pipeline of a high-throughput / high-content screen consisting of


  • Basic compound / instrument / user management
  • Data acquisition using automated microscopy
  • Automated image processing
  • Normalization together with quality control
  • Data storage and archiving using relational databases
  • Data analysis including data modeling and visualization for hit definition
  • Bioinformatics

Thus, the entire software package will enable to (i) store and annotate HCS data; (ii) handle libraries (e.g. siRNAs); (iii) process images; (iv) define hits using multiparamteric classification methods; (v) browse through the data via visualization kits and (vi) answer quality control suits (see Screening pipeline below).


For data processing and analysis a large number of integrations is available. For a simplified overview, see

KNIME integrations

The overview also specifies integrations which are provided by BioQuant / RNAi Screening Facility - they can be downloaded as additional  internal plugins



in the Workflow Repository.



Data pipeline: Ideally, the integrated data base is synchronised with external data bases in order to search for correlations and overlaps.


Screening pipeline [ workflow ]: Example for an entire screening pipeline covering library input files to assay preparation (experiment meta-data), image processing up to statistical evaluation.


Contact: E-Mail (Last update: 29/04/2014)