You are here

| Technology

Computerized Colony Selection Promises to Increase iPSC Generation Efficiency

Comment

Discuss

Review of “A Machine Learning Assisted, Label-free, Non-invasive Approach for Somatic Reprogramming in Induced Pluripotent Stem Cell Colony Formation Detection and Prediction” from Scientific Reports by Stuart P. Atkinson

As we move towards a more technologically advanced society, many simple and complex jobs will soon be the remit of robots under the control of computers. While many sectors see this workplace evolution with a certain amount of mistrust and hesitance, a recent technological advance in the stem cells field will undoubtedly have many researchers jumping for joy!

The study in question, published in Scientific Reports from Xiao Zhang (Chinese Academy of Sciences, Guangzhou, China), reports on a computerized system to aid the generation of induced pluripotent stem cells (iPSCs) via the consistent determination of colony maturation based on morphological [1]. This system promises to reduce the need for intensive training, human error, and variation in lines produced and thereby, significantly increase the efficiency of iPSC generation strategies.

So what are the details of this exciting new advance?

  • Rather than labeling cells or employing invasive techniques, Han et al. developed a computerized system to analyze time-lapse microscopic images of colonies captured during different stage of iPSC reprogramming
  • Reprogramming using an OCT4- GFP knock-in reporter mice embryonic fibroblast (MEF) system permitted the authors to “train” their system employing fluorescence indication as a positive feedback mechanism
    • Verification applied fluorescent immunostaining for pluripotent surface markers (TRA-1-60, TRA-1-81, and SSEA4) and pluripotent gene expression (OCT4, SOX2, and NANOG)
    • Comparisons between algorithm- and manually- “picked” colonies did not uncover significant biological differences
  • Excitingly, this new approach can detect and monitor the earliest morphological changes after the induction of human somatic cell reprogramming by day 7 of the 20 - 24 day process
    • The authors also developed a statistical model to predict the best iPSC selection phase independent of any other resources

Will this new computerized system result in mass redundancies in the laboratory or allow some tired-eyed researchers time to escape the culture hood to accelerate their iPSC-related experiments? Keep tuned to the Stem Cells Portal to find out!

References

  1. Fan K, Zhang S, Zhang Y, et al., A Machine Learning Assisted, Label-free, Non-invasive Approach for Somatic Reprogramming in Induced Pluripotent Stem Cell Colony Formation Detection and Prediction. Scientific Reports 2017;7:13496.