IRLAB uses AI methodology

29 March, 2018

 

One of the exciting West Sweden companies included in the AI mapping is IRLAB Therapeutics, which is developing medicines to treat illnesses in the central nervous system (CNS). The company is currently focusing on the development of drugs to assist patients with Parkinson’s disease. Peder Svensson gave an informative presentation and here provides us with some more insights on how the company is using Artificial Intelligence (AI).

”We use AI methodology to discover and develop the candidates that will be most effective and safe as medicines. The biggest challenge to succeed with this is that knowledge about how the brain works, both when it is healthy and when it is ill, is limited.”

”Processes in the brain are complicated and connections between the candidates’ properties in lab models and how they work to counteract patient illness are not linear. To meet this challenge, we have developed a research platform, ISP – Integrative Screening Process, built on system biology.”

“ISP is built around a unique database of systematically gathered and structured information about medicines and test compound’s effects on behaviour, neurotransmission and gene expression in the brain. Data is analysed and processed with machine learning methods, where emphasis is put on gathering data which is fully comparable between candidates and different mechanisms, and which describes the effects of the interaction in the network of the brain, so-called connectivity.

From the challenges highlighted in the report, Peder Svensson ranks the three most important:

  1. Structure and annotate data
  2. Which data sets are the most important ones?
  3. Complexity is a limiting factor

For more information, contact Peder Svensson, Ph.D., Dir Comp Chem & Biol, CIO, IRLAB Therapeutics,  on +46 730-75 77 03 or e-mail peder.svensson@irlab.se