(Un-)supervised Learning of Cell Population Structure from Single-Cell Snapshot Data
Event details
| Date | 04.02.2019 |
| Hour | 12:15 |
| Speaker | Prof. Manfred Claassen, ETH Zürich (CH) |
| Location | |
| Category | Conferences - Seminars |
JOINT CHEMICAL and BIOENGINEERING SEMINAR
(sandwiches served)
Abstract:
Rare cell populations play a pivotal role in the initiation and progression of diseases such as cancer. However, the identification of such subpopulations remains a difficult task. I will present our representation learning approaches to detect rare cell subsets associated with disease using high-dimensional single-cell measurements and demonstrate identification of rare CMV infection and multiple sclerosis-associated cell subsets in peripheral blood, and extremely rare leukemic blast populations in minimal residual disease-like situations with frequencies as low as 0.01%, as well as identification of morphological patterns associated with cancer severity.
Bio:
Manfred Claassen joined the Institute of Molecular Systems Biology at the ETH Zurich as an Assistant Professor for computational biology in January 2013.
He has carried out parallel studies in Biochemistry and Computer Science at the University of Tübingen and been awarded a Diploma in Biochemistry in 2004 and a Diploma in Computer Science in 2006. In 2010 he obtained a PhD from ETH Zurich. During his doctoral studies he developed statistical methods to design and validate proteome measurements. In 2011 he moved on for postdoctoral training with Daphne Koller at Stanford University, where he focused on inferring informative network models from single cell resolved perturbation studies.
His research aims at elucidating the composition of heterogeneous cell populations and how these implement function in the context of cancer and immune biology by jointly evaluating single cell and genome wide measurements. The Claassen group builds on concepts from statistics, machine learning and mathematical optimization to develop probabilistic approaches to describe biological systems, learn these descriptions from data and to design experiments to validate hypotheses following from computational analyses.
(sandwiches served)
Abstract:
Rare cell populations play a pivotal role in the initiation and progression of diseases such as cancer. However, the identification of such subpopulations remains a difficult task. I will present our representation learning approaches to detect rare cell subsets associated with disease using high-dimensional single-cell measurements and demonstrate identification of rare CMV infection and multiple sclerosis-associated cell subsets in peripheral blood, and extremely rare leukemic blast populations in minimal residual disease-like situations with frequencies as low as 0.01%, as well as identification of morphological patterns associated with cancer severity.
Bio:
Manfred Claassen joined the Institute of Molecular Systems Biology at the ETH Zurich as an Assistant Professor for computational biology in January 2013.
He has carried out parallel studies in Biochemistry and Computer Science at the University of Tübingen and been awarded a Diploma in Biochemistry in 2004 and a Diploma in Computer Science in 2006. In 2010 he obtained a PhD from ETH Zurich. During his doctoral studies he developed statistical methods to design and validate proteome measurements. In 2011 he moved on for postdoctoral training with Daphne Koller at Stanford University, where he focused on inferring informative network models from single cell resolved perturbation studies.
His research aims at elucidating the composition of heterogeneous cell populations and how these implement function in the context of cancer and immune biology by jointly evaluating single cell and genome wide measurements. The Claassen group builds on concepts from statistics, machine learning and mathematical optimization to develop probabilistic approaches to describe biological systems, learn these descriptions from data and to design experiments to validate hypotheses following from computational analyses.
Practical information
- Informed public
- Free
Organizer
- Profs. Vassily Hatzimanikatis and Bart Deplancke
Contact
- Institute of Bioengineering (IBI, Christina Mattsson)