Automating Data Processing in Cancer Research

APEER is helping researchers to analyze data faster and with higher precision

Customer Story

An important aspect of cancer research is understanding why some cancer cells escape chemotherapy and become even more aggressive and resistant to treatment. At the University of Vilnius in Lithuania, Prof. Valius Mindaugas and Nadežda Dreižė use confocal microscopy to understand the impact of selected chemical compounds on the growth and cell signaling behavior of different cancer cell lines. By examining the localization of specific proteins within the cell, with and without chemical treatment, Prof. Mindaugas and his team can draw conclusions about protein function.

Prof. Valius Mindaugas and Nadežda Dreižė collecting data on their ZEISS confocal.

Experimental Challenges

To identify significant trends in these types of experiments, thousands of cells must be analyzed in a reproducible manner. Doing this manually is both challenging and time consuming. Prof. Mindaugas and his team were able to utilize APEER for automatic and unbiased analysis of protein localization – resulting in more precise, impactful analyses.

Experiment Design

In the below example, the team was interested in examining the expression levels and location of different versions of the P53 protein – a tumor suppressor protein that plays a major role in cellular response to DNA damage and other genomic aberrations. In these experiments, DAPI (blue) was used to label the cell nuclei and antibodies were used to fluorescently label p53 (green) and a phosphorylated version of p53 (red) in the HCT116 colorectal cancer cell line.

APEER is able to automatically detect cell nuclei and then measure green and red intensities within each cell nucleus.

APEER Data AnalysisOne example showing detection of cell nuclei (large circles) and then measurement of the phosphorylated version of p53 (the smaller dots). Note that overlapping cell nuclei are rejected from the automatic detection.

Using this approach, the following measurements were automatically computed:

  1. Ratio of red (phosphorylated p53) over green (p53) intensity per nucleus
  2. Number of red (phosphorylated p53) grains over green (p53) intensity per nucleus

This was done with and without treatment with oxaliplatin, a drug commonly used in colorectal cancers.

The first results show that treated cells have a lower average ratio of red (phosphorylated p53) intensity over green (p53) intensity:

The second results show that treated cells have, on average, a higher number of red grains (phosphorylated p53) over green (p53) intensity:

Using APEER, Prof. Mindaugas and his team were able to quickly and accurately analyze large amounts of data. These data show that the localization and expression levels of the different versions of p53 are affected by treatment with the drug. While further experiments are needed to better understand what this means and how this could contribute to more effective treatment strategies, this is an important step in better understanding drug resistance in cancer cells.

More about the Mindaugas lab

More information on APEER

Tags: Cell Biology, Machine Learning