Cell Analysis Software Speeds Search for Potential Eye Cancer Treatment

A new cell tracing algorithm has quickened the discovery of an important, previously unknown role for the tumor suppressor gene Rb, which, when abnormally inactivated, is the probable trigger of the aggressive childhood cancer retinoblastoma.

Intensive efforts are being made to identify and characterize the roles of Rb in eye development, but studies have been greatly hampered by limitations of technology available for finding retinal anomalies at the cellular level.

A major limitation has now been removed by the specialized cell tracing, image analysis program developed for a study of the Rb gene published December 27, 2012, in the journal Proceedings of the National Academy of Sciences of the USA (PNAS). The program allowed researchers to analyze thousands of cells instead of just a few dozen. It could also be used as the basis for developing algorithms specialized for analyzing morphological changes in large numbers of other neurons and other cell types.


"Our paper shows that horizontal neurons known to be deficient in this gene exhibited abnormalities in the way their connecting dendrites were organized after a certain number of days after birth," said author Ryan Kerekes, PhD, of the Department of Energy's Oak Ridge National Laboratories (ORNL; Oak Ridge, TN, USA).

The study, led by senior author Michael A. Dyer, PhD, was a collaborative effort among investigators at ORNL, University of Tennessee (UT) at Knoxville (USA), UT at Memphis (USA), St. Jude Children Research Hospital (Memphis, TN, USA), and the Universidade Federal do Rio de Janeiro (Rio de Janeiro, Brazil).

Dr. Kerekes and ORNL team members Dr. Shaun Gleason and Dr. Mahmut Karakaya developed the computer program and automated tool that traces the very complex and intricate dendritic arbor. This allowed scientists to draw a line along each branch in the neuron's tree of connectors so the branch can be measured in terms of length, angle, and other parameters.

"Previously, this was a very time-consuming and labor-intensive process," Kerekes said. "Existing commercial software tools were not tuned to this particular data and, as a result, produced too many tracing errors."

As a result, only a handful of cells could be analyzed in sufficient detail, according to Kerekes, who noted that the ORNL tracing algorithms achieves the level of accuracy required to analyze thousands of developing neurons.

Related Links:
Oak Ridge (US) National Laboratories
University of Tennessee, Knoxville
Universidade Federal do Rio de Janeiro