— Predicting drug effects in a multimodal biological network

PHC

The fundamental understanding of the human organism and how it is perturbed by variants and drugs forms the basis of delivering tailored care. How can we predict the specific effect a drug has on its known cellular target(s) based on its cellular interaction network?

Apply until April 18, 2021 / Xplorers Camp on May 11, 2021 

Question to be solved

How can we predict the specific effect a drug has on its known cellular target(s) (e.g., activation, inhibition, antagonism) based on its cellular interaction network?

General Background

In personalized healthcare, the fundamental understanding of the human organism and how it is perturbed by variants and drugs forms the basis of delivering care tailored to the individual patient. To enable this, the complex interplay between the biological entities that drive disease at the molecular level can be modeled as a multimodal network which can be leveraged for machine learning applications, including the prediction of polypharmacology, drug combinations, or candidates for drug repurposing.
However, many machine learning applications on biomedical graphs struggle with lacking detail driven by 1) missing data in the data sources used to build the network and 2) an incomplete picture of the organism elucidated by different types of experiments. In particular, the lack of detail on drug effects on their molecular targets can cause issues with utilization of the graph for pharmaceutical applications. In this challenge, we ask for your help to develop computational methods to overcome the missing data problem of drug effects in multimodal networks to open up the full potential of biological networks for personalizing healthcare.

Data Types & Technologies

  • Working with multimodal graphs of biological entities (genes, genetic variants, pathways, diseases, drugs, …)
  • Experience developing machine learning models on graphs
  • Python (or R)
  • Tensorflow 2 or PyTorch

Supporting Material or Links

Needed Skills

  • Self-starter who likes tough machine learning problems and wants to make an impact in healthcare
  • Experience with advanced missing data imputation methods
  • Experience with state-of-the-art mechanistic graph learning methods as well as representation learning and graph embeddings.
  • Experience with biological networks and knowledge graphs can be of advantage.
  • Completed MSc in technical/quantitative field
  • (At least basic) Understanding of pharmacology and biology

Mentor

Dr. Charlotta Fruechtenicht
Senior Data Scientist, PHC Analytics

Form of Cooperation

Preferred scale: 6 months full-time (likely remote) internship

Possible format: Full-time internship, with potential to develop into Master Thesis or part of PhD research project

How to present your Idea

We do not expect a bullet proof solution, but please present your idea on machine learning approaches to solve the task including ideas for suitable training data and validation strategy in a short and concise format (3-5 slides or max 1 page written document plus possible figures to illustrate the concept)

By sending this to us via the submit button you agree to the following:

 

  • you confirm that you are the author of the submission and entitled to dispose of rights of use and exploitation of the contents of your submission, and that you have not yet granted any rights of use and exploitation to third parties that would be infringed by your submission;
  • you grant to Roche Diagnostics GmbH the unrestricted, sublicensable and exclusive right  to use and exploit your submission by all means known today or in the future. This includes without limitation the rights to reproduce, distribute, and exhibit your submission, as well as the right to communicate your submission to the public. You also grant to Roche Diagnostics GmbH the right to edit the submission, to translate it, and to create abbreviations and summaries (abstracts); the aforesaid rights to use and exploit also apply to such edited versions, translations, abbreviations and summaries. 
  • Roche Diagnostics GmbH will designate you as the author of the submission, and will recognize and respect your moral rights in the submission.
  • The relationship you enter into by sending this via the submit button is governed by the laws of the Federal Republic of Germany, and the courts of Germany have international jurisdiction for any disputes arising under or in connection with this relationship..

Submit

Any problems with the submit button? Please send your submission to healthcare.xplorers@roche.com.

 

Further information on our privacy policy can be found here.

Submit

Any problems with the submit button?
Please send your submission to healthcare.xplorers@roche.com.

 

Further information on our privacy policy can be found here.