Training and Hackathon Sessions



21 Nov 2017:

○ 11:00 - 13:00 Parallel training sessions
      ■ 11-13 Ontologies
      ■ 11-12 NanoQSAR modelling
      ■ 12-13 QSAR and conformal predictions

○ 14:00 - 16:00 Parallel training sessions
      ■ 14-16 Biokinetics modelling
      ■ 14-16 Workflow management systems

○ 16:30 - 17:30 Hackathon sessions
      ■ Introduction to the biokinetics hackathon
      ■ Introduction to the data hackathon on estrogenic activity data

22 Nov 2017:

○ 08:30 - 11:00 Hackathon sessions
      ■ Hands-on session on the topics introduced the day before

Parallel training sessions: 11-13 Ontologies 

Title: Semantic/ontologies
Type: Training
Duration: 120 min


Responsible(s): Egon Willighagen (UM)


Short reminder on how to browse ontologies
■ Introduction to the Jenkins build system, Slimmer and the eNanoMapper ontology
■ Initial population of a general toxicology ontology
■ Work on ISO terminology

Prior knowledge: not required


Jenkins build server build server
■ ontology GitHub repository
■ ontology tests GitHub repository
eNanoMapper ontology browsing tutorial

Parallel training sessions: 14-16 Biokinetics modelling

Title: Biokinetics modelling
Type: Training
Duration: 120 min


Responsible(s): Frederic Bois (INERIS) and Harry Sarimveis (NTUA)


■ Aim: understanding the use, form, inputs and outputs of physiologically based (PBPK) pharmacokinetic models.
■ Expected outcome: the participants should be able to attend usefully the next day hackathon.
■ Requirements: superficial reading of the bibliography given in the resources section below.


■ General presentation of PBPK models (F. Bois, 45 minutes)
■ Presentation of software applications for developing PBPK models. Customising PBPK to individual time-drug concentration data. Creating optimal drug dosage regimens (H. Sarimveis, 35 minutes)


About the software PKSim, see and documentation herein: 

See for example the recent 

Parallel training sessions: 14-16 Workflow management systems

Title: Workflow management systems
(nextflow, squonk)
Type: Training
Duration: 120 min


Responsible(s): Tim Dudgeon (IM), Cedric Notredame
(CRG), Daan Geerke (VU) and Marc van Dijk (VU)


We will describe and demonstrate workflow tools that will be incorporated into the OpenRiskNet platform, and form a key aspect of the ways end users interact with the OpenRiskNet services. This will primarily be a demo of the tools involved and a discussion of how we plan to use and integrate these in OpenRiskNet, along with a limited opportunity to use these tools in a hands-on manner.


1. Introduction to the OpenRiskNet platform
2. Nextflow: orchestrating complex scientific workflows
3. Squonk Computational Notebook: a workflow tool for non-geeks
4. MDStudio: a microservice based workflow system


Nextflow (
Nextflow enables scalable and reproducible scientific workflows using software containers. It allows the adaptation of pipelines written in the most common scripting languages. Its fluent DSL simplifies the implementation and the deployment of complex parallel and reactive workflows on clouds and clusters.

Squonk Computational Notebook (
A web based collaborative workflow building and execution environment, along with data visualisation and analysis capabilities. It includes the ability to easily plug in additional services through Docker images. 

MDStudio deploys microservice based, scalable, reproducible and multi-user ready computational workflows at various levels of abstraction. Everything is a microservice: software, scripts, web services, databases, dedicated HPC pipelines and more. Using only a network connection to communicate, MDStudio microservices natively scale up and out on heterogeneous platforms providing a flexible means of creating workflows tailored to the requirements of a research team.

Parallel training sessions: 11-12 NanoQSAR modelling

Title: NanoQSAR modelling (Jaqpot online platform)
Type: Training
Duration: 60 min


Responsible(s): Philip Doganis (NTUA)


Training aim: Provide a modeling experience on nanoQSAR in two directions: Users that wish an introduction to modeling tools will be able to create their first models using the user interface Jaqpot provides.

Expected outcomes: Provide a foundation course on online modeling to serve as basis for the hackathon and hands-on experience with tools.

Requirements: Understanding of QSAR concepts. 


Parallel training sessions: 12-13 QSAR and conformal predictions

Title: QSAR and conformal predictions (CPSign, ModelingWeb)
Type: Training
Duration: 60 min


Responsible(s): Ola Spjuth (UU)


A short introduction to conformal prediction
■ A hands-on exercise with the CPSign program on command-line, allowing for training models and making predictions using conformal prediction with visualization/interpretation of important substructures
■ A demo of the current status of the web based services (ModelingWeb) wrapping CPSign and offering an API in OpenRiskNet


Documentation for CPSign --

Files access:

Hackathon sessions: Biokinetics hackathon

Title: Biokinetics modelling
Type: workshop/hackathon
Duration: 30 min introduction + 60 min hands-on


Responsible(s): Frederic Bois (INERIS) and
Harry Sarimveis (NTUA)


The aim of this workshop is to create a biokinetics model using the PKSim software and then use ORN functionalities to expose the model as a service on the web.

During this workshop we will:

Show the basic functionalities of Jaqpot API and UI. 
Guide the preparation of biokinetics models using the PKSim software.
Demonstrate how to use the Swagger documentation of JaqPot API to expose the model as a web service and use it for generating time-drug concentration profiles.


Hackathon sessions: Hackathon on estrogenic activity data

Title: Estrogenic activity data for endocrine disruption predictions
Type: Hackathon
Duration: 30 min introduction + 150 min hands-on


Responsible(s): Thomas Exner (DC)


Endocrine disruption is of major regulatory importance. ECHA and EFSA are developing at the moment scientific guidance to enable endocrine disruptors to be identified (unfortunately not public available yet). FDA have created over the years valuable resources in form of the Endocrine Disruptor Knowledge Base (EDKB) and the Estrogenic Activity Database (EADB) and the estrogen and androgen receptors are important targets in EPA’s ToxCast and Tox21.

The hackathon is meant to look at these data sources and different in silico and integrated in silico-in vitro approaches and identify areas in which especially the new FDA data in combination with other data sources could help to improve the existing methods.


EDKB paper:

EDKB paper: 

US EPA Endocrine Disruptor Screening Program – The Pivot:
EDSP21 Dashboard:

US National Toxicology Program:

ToxCast/Tox21: and

FDA case study:

Tox21 models

Ahmed’s models: and

Nicole’s models:, and