You are here: Home / Colloquium / Summer Semester 2015 / Talk: Thomas Unterthiner

Talk: Thomas Unterthiner

When Apr 23, 2015
from 05:15 PM to 06:15 PM
Add event to calendar vCal
iCal

CE Colloquium

Thursday, April 23, 5:15 pm

Room S3-055 in Science Park 3

 

Title: Toxicity Prediction using Deep Learning

Speaker: Thomas Unterthiner

Affiliation: JKU

Abstract:  Everyday we are exposed to various chemicals via food additives, cleaning and cosmetic products and medicines -- and some of them might be toxic. However testing the toxicity of all existing compounds by biological experiments is neither financially nor logistically feasible. Therefore the government agencies NIH, EPA and FDA launched the Tox21 Data Challenge within the "Toxicology in the 21st Century" (Tox21) initiative. The goal of this challenge was to assess the performance of computational methods in predicting the toxicity of chemical compounds. State of the art toxicity prediction methods build upon specifically-designed chemical descriptors developed over decades. Though Deep Learning is new to the field and was never applied to toxicity prediction before, it clearly outperformed all other participating methods. In this talk we will look behind the curtains of the winning method, and show how deep nets automatically learn to recognize features resembling well-established toxicological properties of chemical compounds. In total, our Deep Learning approach won both of the panel-challenges (nuclear receptors and stress response) as well as the overall Grand Challenge, and thereby sets a new standard in toxicology prediction.

 

Short Bio: Thomas Unterthiner has a BSc in Computer Science and an MSc in Bioinformatics from the Johannes Kepler University Linz (JKU). He is currently working towards a PhD unter the supervision of Prof. Sepp Hochreiter. His research focuses on the theory and application of Machine Learning techniques on biological data, mostly involving Deep Learning and Neural Networks. Recently, he was part of the research team that won the prestigious Tox21 Data Challenge.