You are here

Tanszékcsoporti szeminárium

2014/15 II. félév
SZAB Székház 217. sz.
Horváth Péter (MTA SZBK)
Image analysis and machine learing methods for high-content cancer drug discovery
In this talk I will give an overview of the computational steps in the
analysis of a single cell-based high-content screening, a novel way in
modern drug discovery. First, I will present a new microscopic image
correction method designed to eliminate vignetting and uneven background
effects which, left uncorrected, corrupt intensity-based measurements.
Variational methods for single cell segmentation will be presented. I will
than discuss the Advanced Cell Classifier (ACC) (, a
software tool capable of identifying cellular types based on features
extracted from the image. It provides an interface for a user to
efficiently train machine learning methods to predict various phenotypes.
For cases where discrete cell-based decisions are not suitable, we propose
a method to use multi-parametric regression to analyze continuous
biological phenomena. To improve the learning speed and accuracy, we
recently developed an active learning scheme which automatically selects
the most informative cell samples. Finally, combining the above methods, I
will discuss CL2M (correlative light-light microscopy) a revolutionary
technique for single-cell analysis of various tumor types.